The Connected Engineer

Un podcast per ingegneri, progettisti e innovatori.

Episode 13: Coming Face-to-Face with the Digital Twin


Gavin Quinlan: Welcome to the Connected Engineer, a podcast for engineers, designers, and innovators. I'm your host Gavin Quinlan and each episode I invite experts to discuss the product development challenges companies are facing today as well as some of the new trends in the area of product development.

Many organizations strive to enter the smart connected product market but aren't sure where to start. In today's episode of The Connected Engineer, you will hear a connected device story from Welch Allyn, a leading global manufacturer of medical devices. David Kellner, Senior Manager of Mechanical Engineering, will share how Welch Allyn set new entry stakes in a mature market segment to stimulate growth and frustrate competition. Leveraging connectivity, they established a new level of expectation for their products. David, can I ask you to introduce yourself? Tell us a little bit about your background and help our audience understand a little about who Welch Allyn are.

David Kellner: Sure. Thanks, Gavin. So, I'm currently the Senior Manager of Mechanical Engineering for Welch Allyn. We're a division of the Hill-Rom Company. And I'm responsible for design engineering in our research and development group. While I've worked primarily in electronic vital signs the last several years. I've also worked fairly extensively in vision screening, physical assessment, and digital x-ray. I have my Bachelors in Mechanical and Aeronautical Engineering, and then also have my Masters in Engineering, Global Operations Management.

You know, Welch Allyn was not too long ago acquired by Hill-Rom. But prior to that, we were a family-owned private company for almost exactly 100 years. Now, we're gladly part of a much larger public company which we're really excited about. You know, we've been providing front line care technology to physician's offices, skilled nursing facilities, hospitals for a long time. And we're excited to keep doing that and keep bringing better care environments to our customers so that they can bring better care to their patients.

Gavin: So, let's talk about the product, right? So, we're here to talk about the smart connected product. What is the product?

David: So, in this particular case, the product that I wanted to talk about was the Connex Vital Signs Monitor. This is a new product that came to market for us in 2010 and that was really an inflection point for the company. Prior to 2010, some of our products were connected, kind of connected, some of our producers were really not connected. And, you know, when CVSM came out in the 2010 time frame, that was the point in time where we said, "Look, all products going forward are gonna have some sort of a connectivity strategy." And this product really, you know, kinda set the benchmark. You know, not only does it have Wi-Fi connectivity to the hospital network but it was also, you know, architected from the beginning to be designed for a thing where it could exceed us. So, not only in the clinical workflow is it a connected product but also in the service workflow. It's a connected product all the way back to, you know, Welch Allyn.

Gavin: So, this is a product that measures, the what, the vital signs. So this is a sort of thing, what you'd see on all of the TV shows, sitting beside the bed and beeping and bleeping, and stuff, that product?

David: Yeah, yeah. And we've been in the multiparameter vital signs device market for a long time, and we've made vital signs devices that were just designed to come into a patient's room, collect basic vital signs, and then leave the room. We also made continuous patient monitors that were designed to be left at the bedside, and would continuously monitor the patient's vital signs. And then they would beep an alarm if something went out of range. Now, the devices that we're building are really more hybrid devices where you can change the device to be a sort of a spot check, just sort of wandering, collect vital signs, and wander out device. Or, it could be a continuous bedside patient monitor by just changing what's called a profile on the device. And then not only does the whole user interface change but the behavior behind the parameters also changes. It definitely makes the device exciting for customers because it can kinda do both.

Gavin: Do both jobs. So, early around, I think you used an acronym called CVSM and I was going to ask you what it meant. But I guess it stands for continuous vital sign monitoring, would it be, is that right?

David: It's actually the Connex Vital Signs Monitor. Connex is sort of a sub-brand for Welch Allyn. We have a couple of different products that are in the Connex portfolio. There's a server side central station product and service side connectivity software that goes as part of the suite with the Connex Vital Signs Monitor. The monitor itself is the bedside device that actually goes into the patient room, and that's the primary interaction point between the clinician and the patient.

Gavin: So, do you mind me asking like was there a particular trigger or event that drove the smart connected design thinking?

David: Yeah. You know, for us, this is a very crowded space. If you go to any of the big annual medical device conferences and look for companies that build a multiparameter patient vital signs device or patient monitor that takes, you know, blood pressure, temperature, pulse oximetry, and maybe EKG, you will see dozens and dozens of companies that make these types of devices. The market is extremely crowded. The space is starting to really commoditize. There's a lot of import devices that are coming in that are built over in the Far East that are coming in at very inexpensive price points. And with a lot of big companies all competing for this space, it was really driving price points down and it was driving margins down. So, we had to say, "Look, we're not gonna do the Walmart strategy. We're not going to just move our products over to mainland and try and drive the prices into the floor. We're gonna really sit down, understand the challenges that our customers are having, what motivates them, what frustrates them, you know, what are the problems that they're having in their space, and how can we innovate to make a product that's really different."

You know, when we did that, we noticed that being able to accurately collect electronic vital signs was really no longer a differentiator. When we were doing that back in 2000, 2001 and people were changing from, you know, the sort of analog process of collecting this data, into now doing it digitally, that was exciting. It's not anymore. We also noticed that our customers were rapidly converting over from this really kinda mixed electronic medical records and paper medical records, and driving really hard towards complete electronic medical records. The American Recovery and Reinvestment Act of '09 really put a lot of money on the table for healthcare providers to implement EMR systems. And there's a lot of benefits for hospitals to go with a paperless workflow. Information is available very quickly to the people that need it to make decisions and information doesn't get lost. The right information goes into the right patient record and you don't get transcription errors.

But, in order to really make that whole workflow work, you needed to have a bedside device that could fully record not just the patient's vital signs but could also put that in context of exactly which patient, which clinician, exactly when it happened. And then instantaneously push that information right into the patient record. And we weren't gonna get there by just taking one of our existing devices and making a couple of updates. For us, you know, this had to be a from scratch new product.

Gavin: Yeah. So, like I said, it's not just, "Hey, let's stick a WiFi card in the machine and we're good to go." Like you said, you took that as an opportunity to say, "Let's see what's actually needed here." And I guess to that point, like, how do you determine the extra capabilities or the redesign, you know, requirements that are gonna be needed for something like this? How do you get to that place?

David: Well, you really have to spend a lot of time with your customers and you also have to realize that there are a lot of subtle but very important differences with customers based on geography. So, that standard workflow that you would see in the U.K. or France, or Germany, or in Southeast Asia, it's gonna be different from what you see in the U.S. or Australia, or Canada. There's gonna be a lot of overlap, there's gonna be similarities, but the workflow differences are really important. It's not good enough to just have a device that's got, you know, 15 or 16 different languages on the user interface. You have to understand why hospitals in France take vital signs in a way that's a little bit different than the way it's done in the U.K. or the U.S. And you have to be able to develop a product that kinda deals with that and realize that, while, some people are ready for a WiFi-connected product that wonders around on a, you know, encrypted 802.11 ABG network.

You know, there are other workflows where they're really not at that point with their infrastructure. And they want to just connect a device to a laptop over a USB cable and they want to be able to, you know, push data that way. So, if you know that, you can kind of architect a product, you know, that's going to, you know, work well for the very sophisticated, technically high-end hospital, but it's also gonna work for the smaller facilities that aren't quite there yet. And then, you know, you're gonna always have your special cases like the VA in the U.S. where...Everybody is concerned about privacy of data, but the VA, you know, has more stringent requirements than a lot of other hospitals and they have, you know, special compliance criteria that, you know, you have to meet or you can't sell your product into the VA network. And the VA network, you know, around the world is a really big hospital network, so that's an important customer. So you have to sort of, you know, start to recognize those workflow challenges.

And so a lot of research on the front end goes into understanding exactly what those needs are and how they differ, and then say, "Okay, you know, what architecture can we come up with that is modular, it's adaptable, and, you know, you can build it in such a way that it would work really well here but it would also work over here, and it would also work over there. You know, and do it in a way that's elegant and not, you know, burdensome to the end-user."

Gavin: So, I mean, like, you went in with the objective, so to speak, of a product that could work in basically any hospital across the world. You wanted to have a product that was truly global from the outset.

David: We did. Initially, during the research phase, we partnered with industrial design firm to get a really good understanding of what we were dealing with here. We split our group up and we had a group in Asia, a group in North America, and a group in Europe. And, you know, each group spent a few weeks traveling around their designated region and just spending a day in a different facility. You know, each day spending a, you know, an afternoon or a whole day in a facility. And, you know, looking at their workflow and talking to the nurses and watching how they do their stuff. And following people around and kinda hanging out in the environment to get a feel for, you know, what is it like here, what is it like there?

You know, I got to be on the Europe team and, you know, we were all across Europe in a whole range of different facilities. And, you know, that experience sort of deeply influences your understanding of what will work and what will not work. And so we all came back and became part of the marketing engineering design team for the product. And that was the foundation of knowledge that we use to make a lot of different decisions when we were choosing the touch screen and the back light, and the colors and screen layouts, and what parameters were in the device, and ergonomically how things were organized. It all kinda came back to our experiences, spending time with customers in their environment, and listening to, you know, what was important to them.

Gavin: I think every engineer sort of starts with these objectives at heart but then, you know, sometimes the work gets in the way and maybe you get a little bit disconnected, so to speak from what's going on day to day with the product. So, again, I think I can only add massive value to a device or a design. So, right, you do all this work, you partner with the industrial design company. You probably got, you know, reams and reams of feedback from the different theaters where you've assessed and you, in some way, manage to boil that down to a set of global requirements. In terms of then doing the design itself, you know, and turning that into like a first prototype or the first version, again, like, you know, was it a... Well, we have something that does 50% of this or we have nothing that's really doing any of these things today. So, like, can you give us maybe a little bit of background on the how, what was the next step forward?

David: We have a lot of intellectual property in this space because we had been doing electronic vital signs for some time. So, we were able to leverage our portfolio of sort of underpinning technology. We, you know, we already had our own thermometer. We already had our own blood pressure technology and our own, sort of, blood pressure algorithms, and stuff like that. So, a lot of the component pieces that make up a patient monitor with technology that we already had, we had to figure out a way to sort of package that and make it into a connected product. Luckily, because, you know, we hadn't been in the business of doing wireless continuous patient monitoring for so many years, and had gotten into the business of making our own wireless radios. We had gotten really good at being able to either in batch or continuously stream vital signs information over a network and do it in a way that we wouldn't drop information. And we knew how to do that really well, so we were able to sort of bring that into the mix from our continuous monitoring product portfolio. But we had to sort of figure out and say, "Okay, we're not building an ICU monitor here. We're building really a general care MedSurg floor type device which is very, very simple to use, that could work in all these different geographies. It can handle a lot of different languages and it really needs to be intuitive."

So, you know, we started iterating on, you know, user interface concepts and, you know, mechanical electrical architectures fairly early on. And so there was a lot of formative usability that happened where, you know, we were taking the design language that we had developed based on our initial interactions, and we were building that into sort of physical forms. And then reflecting that back against various customer groups and watching how they would interact with the product, and, you know, what they would sort of just naturally do. And where they would get confused and where they would just, sort of, flow very normally through the work in the way that we expected them to. And then we would just take portions of the design and we will just throw it away and say, "Okay, that was a really cool idea but it just didn't work, so let's get rid of that. Let's do something else."

And so we went through this sort of iterative process where we were building things, 3D printers, screen layouts, and trying to figure out how to build a workflow and a user interaction that was gonna be successful. You know, after several iterations, we could see that we were really starting to dial in. And then, you know, in the background, there was a lot of performance testing that had to happen in the lab to make sure that the product was, you know, not only going to be easy to use but that it was gonna last, you know. So, you know, we cycle tested devices for, you know, tens of thousands, hundreds of thousands of cycles to try and fair it out any little tiny glitches or other things. So that, you know, we would work through them in the lab and our customers wouldn't have to, sort of, suffer them in the field, because we didn't intend to build a couple of these a year. You know, we intended to build thousands upon thousands of these things a year, because, you know, we've always been able to do that.

Gavin: So, in terms of, let's say, just connecting back to the global piece of the conversation. I mean was the product created in a modular sort of way, where if you didn't need like the USB connections, you didn't have to pick that or tick that box in the selection. Or, is it, you know, is it a one size fits all sort of, you know, approach in terms of how the product finished up.

David: When we were all said and done, we decided that there were certain baseline capability that just sort of has to be there. You know, we said, wired Ethernet, USB in, USB out, and a high definition, you know, color touch screen and blood pressure were things that every single monitor absolutely had to have. Beyond that, we said, you know, we want a lot of modularity. Some of our customers have very mature wireless networks. They really want a wireless device. So, we architected in such a way that the wireless radio was sort of a modular component, and you could buy the product with it or buy the product without it. And it was really easy on the production line for them to assemble and configure the device to actually, you know, install the radio card or not install the radio card. Some of the other clinical parameters are very preferential based on region. The type of pulse oximetry that you use tends to be a region by region, facility by facility choice. So, you know, we offered our customers a choice of two really popular OEM pulse oximetry modules. They could choose whichever one they wanted and then on the production line, that was just a module.

We actually design it as a fully enclosed, you know, plastic box that would just drop right into the back of the patient monitor. And then when you put the back cover on, it would hold all these modules together. It also was part of our repair strategy, because if something went wrong in that central module, you could take the back cover off the device. You could slide out a module, slide in another module, and the device on the outside was connected but on the inside it was also connected. All the sensor modules on the inside of the device actually connect with USB cables. So, if you're having a problem with a, you know, pulse oximetry sensor module, you can literally just unplug a USB cable, swap out a module, plug in another USB cable, and put the device back together and you can deploy it back to the floor very quickly. And then you can send that, you know, that defunct module back to Welch Allyn and we'll gladly take care of it, you know, for the customer.

Gavin: So, you mentioned some other sort of let's call them, you know, items there like 3D printing and things like that. Did the project of, you know, creating the smart connected product, has it brought other learning's into play that have been taken into projects that came afterwards?

David: It really did.

Gavin: Just give me one or two ideas. Not that we don't, like I said, just one or two things, maybe again just to like from the audience perspective and I think from another engineer out there who's in the similar type of role. I think sometimes, again, you know, we don't always quantify the benefits that's brought, you know, we can see it in the project itself but sometimes there is that sort of the learning or the new way of doing something that goes into future products, that sometimes isn't quantified or spoken about.

David: Yeah. I think one of the things that was really different about this particular project was that, you know, fairly early on we realized this was really big for us. And while other projects had always been trying to, sort of, strike that balance between time and cost, we realized on this project, it was all about time. It was all about getting to a particular market window in a particular time, and we knew that the vast majority of the cost of the project was gonna be labor. Nitpicking over how much money you're gonna spend on prototypes is really irrelevant. What you need to do is finish the development and get the product to market and start earning revenue. And the faster you can get a great product out, the sooner you're gonna start earning revenue. And the soon you're gonna stop spending tremendous amounts of money on research and development. So, that really became our framework. We were always looking at things from the perspective of time.

So, if 3D printing, a mass quantity of parts was what we needed to do, you know, to get to the next prototype, to be able to tell if this was working or not working in the workflow. We would do that and we wouldn't quibble over the cost of the prototypes. If it meant going to Protomold or another, you know, rapid prototyping house and generating a bunch of parts really quickly, that we could then bring back and use for a thermal test or a drop test, or a vibration test, or a shipping test, or whatever it was. We would do that because, you know, in the end, it really wasn't about the money that we were spending on the parts. It was all about time. It was all about the time of this huge team of engineers that needed to get this product done and needed to get it done to an extremely high level of quality so we get it out the door.

Gavin: Right. So, Dave, from speaking with you just here in the short time that what we've been speaking, I mean it's quite evident that, you know, you worked, I would say like intensely on this project. I'm sure it took a number of years, you know, to deliver. For the engineer out there, for that company out there who, you know, perhaps hasn't ventured into the smart connected product space yet. But, you know, obviously with the knowledge that that's the way things are going I think costumer expectations are, you know, the products are more connected than not now. And, you know, they have to be smart or there's just that expectation, you know? What advice, if any, would you propose to give to a peer or a colleague in terms of trying to do one of these projects or take it on?

David: You know, there's a lot of technology that's available off the shelf these days. From an engineering standpoint it's so easy to drop a WiFi chipset, a Bluetooth radio, a USB port, or any other kind of connected solution into a device these days. The tricky part is not figuring out how to do the engineering work to, you know, to physically put the parts in there. It's trying to figure out why you're doing it, what problem you're trying to solve. And if you understand the customer and what their needs are and what their problems are, and you understand that better than anybody else does, then you can figure out how to do it and how to do it really well. And, you know, you'll really build a great product.

If you're trying to delay your connectivity on top of a product just for the sake of it, you'll end up, you know, with kind of a marginal solution. Then you'll have to iterate the following year with something else and then the following year with something else, and you'll eventually creep up on something. But if you start with like a really deep and thorough understanding of what is the real problem you're trying to solve here and how does connectivity solve that problem, then you can really knock it out of the park.

Gavin: In terms of the product itself and, you know, now that it's out there in the market as in...I think you said the project completed, you know, five years ago or so? Are you able to speak to the success of that product in the marketplace in any way? Again, just to help the listener, you know, have an appreciation for, you know, how did the project go? Okay, so great work, love industrial designing company involved. Obviously, a great team. Did it pay off?

David: Yeah, you know, it really did. As an example, I mean when we first launched this product, you know, we designed our production line. I think we designed our production line so it would build 50, 60, you know, somewhere in the 55, 60 devices per day margin. And within the first two years, we had to redesign the manufacturing process twice to dramatically increase output because we couldn't keep up with demand. And those are really good problems to have. And the other thing is, because the architecture is so modular, and because customers were so excited and responding so well to this product after the initial launch with all this capability, you know, we came back pretty much every year since then and released updates and upgrades. In some cases, we released, you know, a whole new better software capabilities. And, you know, customers that had bought the first device back in 2010 using the Axeda platform we're able to download that as a software upgrade. You know, upgrade a 2010 vintage device to 2016 or 2017 software, and get not bug fixes but like real kind of interesting new functionality.

You know, we also released a bunch of new sensor modules that allow the device to, you know, to do things that it couldn't do initially. And all of that was enabled by the fact that, you know, the product had really taken off. And we really have substantial market share in this electronic vital sign space. And there's a lot of big players in this space, you know, think of a, you know, large medical device company. A lot of companies build multiparameter medical devices, you know, in this vital sign space, you know. Welch Allyn, you know, consistently has the strongest market share in this segment.

Gavin: I think that's a really good evidence that, you know, this type of approach to a product design effort especially, you know, as you move into the smart connected world that can pay significant dividends.

So, that's all for today's episode of The Connected Engineer. I'd like to thank David very much for join us and sharing the Welch Allyn story. My takeaway from our conversation is that the opportunity that new solutions or technology, like the IoT, can provide to companies and engineering teams out there, is to reevaluate existing products and find news way to add value for both their customers and themselves. And I hope, David's story has proved that fact for our next episode.



Gavin Quinlan: Welcome to "The Connected Engineer," a podcast for engineers, designers, and innovators. I'm your host, Gavin Quinlan, and each week, I'll invite experts to discuss the product development challenges we are facing today as well as the trends of tomorrow. On today's episode, we're going to talk about the medical device industry, and the trends that medical device companies should be aware of, and what other industries might learn from it. Today, we have invited Michelle Boucher, VP of Engineering Software Research at Tech-Clarity. Michelle, maybe we could just start with a little introduction of yourself, your background, and what Tech-Clarity does.

Michelle Boucher: I'm a industry analyst, and so I focus on looking at best practices and helping companies improve the way that they develop products and bring products to market. And we focus on a variety of different product development topics, and we conduct our research through surveys and working with end-users, and then we make that research available through research reports, webcasts, and a variety of different ways just to help educate people on how they can improve their business and improve the way they bring products to market. My background is in mechanical engineering, and I've worked for a couple of different manufacturers as well as a few different software companies all focused on product development.

Gavin: Great, thank you very much. So let's get stuck into it. So, what are the main industry trends that medical device companies care about today?

Michelle: So first, what's really, really exciting is that we're expecting a lot of growth in this industry. Just as an example, EvaluateMedTech did some research, and they're projecting a 5.2% growth rate every year for the next five years. So there's amazing amount of opportunity, and that's coming from a few different things. I mean, first, our population is aging. We've got baby boomers that are gonna be needing more devices and better services for their health, and medical devices just...they make people's lives better, so that demand for those products is just gonna grow. And the technologies that are becoming available are just really making a huge difference in helping people. And so what's interesting is there's gonna be all of these exciting opportunities, ways for medical device companies to grow, but there's also some challenges in the industry.

Gavin: So, what has your research shown to be, then, the main challenges that medical device companies are, you know, worrying about day to day or concerned with?

Michelle: There's several different challenges that medical device companies are struggling with. The first is profitability, and that's not really unique to medical devices. That's something many companies struggle with, but medical device companies have some unique circumstances. One, in particular, is that they're so focused on compliance and regulations, and so much of their resources go into making sure the compliance and all of that documentation that it takes those resources away from the ability to be innovative and the ability to focus on quality. So they're missing out on opportunities that could help them create more competitive, more innovative medical devices. And then another one is time to market, which again, isn't unique to medical devices. A lot of companies struggle with time to market, but the challenge for med device is you've got these very long regulatory approval cycles, and so much time goes into that where companies are feeling a lot of pressure. They need to recoup that investment. So they need to be looking at ways to really eliminate any bottleneck, improve efficiencies as much as possible because they are gonna be facing that long regulatory approval cycle at the end, so they can start to see a return on that investment.

Gavin: So, I mean, I think, again, from the audience perspective, the most natural thing people think about is that compliance thing. Like, "Once it's a medical device, we better be sure it's right or else we're gonna be in some trouble." So can you help, let's say, the non-expert user who might be listening to this podcast, understand what the impact of a noncompliant medical device might be on somebody's business?

Michelle: That's a huge issue. McKinsey actually did a bunch of research on noncompliance and captured a bunch of information. They found that non-routine quality events cost the industry $2.5 billion to $5 billion every year, so that's a huge cost right there. Then there's also recalls for different reasons. And the FDA Recall Report actually found that things are actually getting worse, they're not getting better. They found a 97% increase in the number of recalls from 2003 to 2012. So it's just getting that much harder to develop a quality device, that much harder to be compliant. And another thing that McKinsey found was that every year, one company loses 10% of their share value just due to a single quality event.

Gavin: Yeah, and I think we're probably all aware, you know, the very public things that typically happens [inaudible 00:05:18], so they get into the public mindshare pretty quickly.

Michelle: Exactly, exactly. So you can see, like, I mean, if that's...that cost adds up and then it hurts your brand.

Gavin: So, I mean, again, to try and link back then to the earlier thing that...you know, to try and link two challenges, you also have mentioned profitability. So, I mean, with the cost of noncompliance being so high, like, how can companies improve their profitability? It almost seems to be, you know, competing with each other in terms of two objectives.

Michelle: So what's interesting is that some of the research that McKinsey found is that just by adopting quality best practices, that it can reduce your quality-related costs by 20% to 30%. And then just by lowering your cost and having a better reputation because of better quality, that can actually increase your profits, too. And their research is showing that profits would be improved by 3% to 4%.

Gavin: So let's say if I'm a company, and, you know, we talked again earlier on about, like, the opportunity in terms of a marketer, that adoption value, is there some value again out there in terms of, you know, best practice adoption, what that could bring to bear?

Michelle: Yeah, absolutely. They're estimating $3.5 billion, so huge opportunity.

Gavin: Yeah, all big numbers all the way through. So let's go back to, you know, the medical device company's best friend, the FDA. What are they saying about this?

Michelle: Yeah, yeah, I mean, and that's a great point. Compliance is so critical. You need to make sure the FDA is happy. And they actually, they've done some of their own research. They've taken a look at some of their device data, and they've reached very, very similar conclusions to what McKinsey found where we really just need more focus on quality. And they found that the number of issues that they find every year during inspections, that's continuing to remain high. They find that the same issues keep popping up year after year and they even described it as kind of a game of whack-a-mole where the same issue just keeps popping up, and then new issues keep popping up. So what I find really interesting is they found that even a company that is compliant with all the FDA requirements, they may not necessarily make a high-quality device, where a company that may not necessarily comply with everything, they can make a very high-quality device. So it just shows that while compliance is very important, those regulations are absolutely critical, it may not be enough, and that, actually, the realization of the situation has led to the Case for Quality initiative.

Gavin: In terms of the same issues occurring year after year, is it mostly because, again, the companies...I mean, there's a lotta great people working in the medical device business with lots and lots of experience. Is it simply a traceability problem? Is it that things aren't documented? I mean, again, from my limited experience working with a few medical device companies, they have standard operating procedures for everything. So I'm just wondering, like, what's the thing that's tripping everybody up?

Michelle: It's a variety of different things, and there's so many issues that it isn't just one single thing. You mentioned traceability. That is a big thing. If you fix this one thing here, you're just fixing that. If you don't have traceability along your entire life cycle or across your product lines or across the different configurations, it's harder to find where else that problem is gonna be impacted. And if you don't get to the root cause...and devices are so complex now, it's hard to get to that root cause. If you don't get to that, then it's hard to correct it for everything. And one company may have the problem and they fix it, and then another company might have a similar problem. So it's really showing that all these regulatory requirements aren't necessarily...they're putting so much focus on documentation and proving that they're compliant, it's taking some of the focus away from quality.

Gavin: Right, okay. I mean, again, it's just this constant balancing act between...I mean, I love the comment you made earlier about, you know, innovation, in some way, getting compromised because we gotta tick all the boxes, for example, from a compliance perspective.

Michelle: Exactly, exactly.

Gavin: Okay, so let's go back to, you know, the Case for Quality initiative that you talked about. Tell us a little bit about that.

Michelle: So that's an initiative that the FDA launched in 2012, and it really comes out of the research that showed that just a purely regulatory approach is good, but it's just not enough. And, as I just mentioned, it forces companies to focus on documentation rather than innovation, and, as a result, companies are disincentivized to be innovative. So we're missing opportunities to bring better devices to market. So the Case for Quality is really about focusing on the quality of device, and, as a result, we expect that companies will be able to bring better devices to market, so that's gonna provide better patient outcomes. And also, it'll be better opportunities for the manufacturers themselves because they'll get a better return on their investment because the devices are so much better.

Gavin: Again, is that, like, Case for Quality initiative, like, is that then something that, again, somebody listening to the podcast could go onto the FDA website and find material on?

Michelle Yeah, yeah. The FDA has a lot of information on the initiative. There's been a lotta research done on it. And then I should also point out the Case for Quality is an FDA initiative, but the greater focus on quality seems to be kind of a worldwide initiative. And the European markets are also seeing a greater focus on quality because everybody's kind of recognizing the need for a better focus on quality and getting manufacturers to focus on the right things, better devices, better quality versus documentation.

Gavin: So, off that point then, I mean, in terms of your research and the clients you're working with, like, what are companies doing to support that Case for Quality initiative?

Michelle: So one of the things, as you're starting to think about quality, the first thing you wanna do is really focus on the entire life cycle of your product. And you wanna go from a shift from your heavy documentation focus to focusing on the product. So it's really gonna involve shifting from this document-centric approach to more of a product-centric approach, and in using the documentation to support the device rather than the documentation driving the device.

Gavin: Let's take the obvious leap, then, from there. So the company, you know, understands the Case for Quality initiative, they understand that to make a more product-centric approach, this...what does that, you know, mean maybe in more tangible terms or day-to-day changes for how a medical device company might work?

Michelle: So one of the things is we wanna think about the whole life cycle and put together a solution that is going to look at the entire life cycle. And you mentioned traceability earlier. That's a big piece of it. Having the ability to take the patient need that you're solving and trace that to the requirement, trace that to the design, and then on the design end, being able to figure out where's gonna be the mechanical components to electrical itself where...bring all that stuff together, and then trace it through to your quality management system, your supplier management system, your production all the way to service. And having that traceability across your entire life cycle be incredibly helpful because when you do have a quality issue that pops up or you have a change, you can identify exactly where that problem is and get to the root cause. And we find that changes tend to be one of the biggest sources of quality problems because you fix one area, and you don't have an understanding of what else was impacted by that change. So just having the software in place that has that traceability in the platform that can support your entire life cycle can be a huge piece to helping companies adopt the Case for Quality.

Gavin: It's interesting. Like, we had a guest in with us last week who's in a completely different industry, and they were all about requirements. And again, he had a very similar story to say about how much value and profitability he was able to create for his business which was, you know, big engines, which, again, seems totally distracted from medical device and the FDA, etc., but again, it just shows being able to know what you're gonna build, that you're building to what you said you were gonna build, and then if changes happen, what impacts that has, and...yeah.

Michelle: Absolutely, yeah. And I think med device has a lotta unique circumstance just because of the compliance piece, but the focus on quality is something that's really important no matter what industry you're in. We see there's so many challenges just to be competitive in today's market. In the past, so many companies focused on time to market. But focus on quality is really helping companies improve their brand reputation and really creating opportunities.

Gavin: Right. So, I mean, look, there's a lot to consider, obviously, in terms of making that change to the product-centric approach. Am I right in saying that, again, Tech-Clarity and yourself have created, you know, again, some content to assist companies with that particular, you know, thought process?

Michelle: Yeah, yeah, yeah. Actually, we've put together a software selection guide. It's titled "The Medical Device Manufacturer Software Selection Guide" that goes through all of this and talks about the Case for Quality, the considerations you should make, and then all the different stages of the life cycle you should consider. And we've also got checklists for each section of the different software criteria that you should think about so that you can be more successful and bring higher quality, better devices to market.

Gavin: And that's available on tech-clarity.com?

Michelle: It is. It is, yup.

Gavin: So, back at the start when we did the introduction, we've made a little reference to what they addressed earlier, but what do you believe that other industries and maybe...and again, listeners to the podcast who are not involved with medical devices, what do you think that they can take from, you know, the learnings that are coming out of this particular industry, Michelle?

Michelle: Absolutely. Other industries, like medical device, are also struggling with time to market. Products, in general, have just become incredibly complex. Many products have evolved into systems of systems, and you've got so much software, electronics, new materials to consider. So complexity isn't really unique to med devices. You've got a lotta global competition to worry about. And we're finding in our research that it's not good enough anymore to just focus on one thing. You've gotta be able to address all of these things really well. And we're finding that a focus on quality is one of the many things that companies are focused on. And so having a system in place that will help you be able to have that traceability across your product life cycle can really help a company meet all that different criteria and then become even more competitive than they are now.

Gavin: That's all for today's episode of "The Connected Engineer." Thank you very much, Michelle, for joining us. Make sure to subscribe to our podcast on iTunes, and if you're a first-time listener, make sure to look back and listen to our previous episodes. We look forward to you joining us for the next podcast.



Gavin Quinlan: Welcome to "The Connective Engineer," a podcast for engineers, designers, and innovators. I'm your host Gavin Quinlan and each week I'll invite experts to discuss the product development challenges we are facing today as well as the trends of tomorrow.

Gavin: The topic of today's episode is requirements management. Research shows that the number one reason why projects fail is due to incomplete requirements, and one of the most important factors to projects succeeding is clear requirements from the beginning. Today we have invited Teejay Momoh, Global Systems Engineering Solution Architect at Cummins to join us, to speak about Cummins journey with managing requirements. Teejay can I ask you to introduce yourself, maybe give us a little bit about your background, and tell us about Cummins.

Teejay Momoh: Thank you Gavin. My name is Teejay Momoh, originally from Nigeria actually, moved here about 13 years ago now for college in search of a better life, got my...my degree is in electrical engineering, joined Cummins about seven years ago. Here at Cummins, I am the Global Systems Engineering Solution Architect, part of doing that I'm just one leg of a three leg organization that focuses on people, processes, and tools. I am the tool leg, evaluating vendors, different tools to satisfy our systems engineering need. I lead our internal team who's responsible and tasked for, on figuring out solutions within Cummins and maintaining those solutions. So a pretty exciting job, I deal with a lot of people every day in different scenarios and different expectations.

Gavin: Thanks Teejay. I mean, so I know who Cummins are they're obviously, you know, a very big business globally. Maybe for some of our audience who might know what Cummins do, could you give a little bit of an intro as to what products they develop and the markets that they occupy.

Teejay: Absolutely, absolutely I can do that. So Cummins Inc. was founded in 1919, we are a global power leader that designs, manufactures, sales, and services internal combustion engines ranging from 2.8L to 95L. We started as a internal combustion engines and we've been there for a very long time but now we're basically a power leader in both internal combustion and other power sources alternatives, natural gas, things like that so, we're mostly known for our engines that we make for our on highway applications those are a lot of the engines that go into a lot of the semis. We are the world leader in this space. About 65,000 employees strong across 190 countries. We basically divide our business into four major segments: we've got the engine business segment, we've got our component business segment, we've got a distribution business segment, and finally our MAVIS business unit that was just formed by a recent reorg of our power systems business segment that basically manufactures a lot of our power generation products really large engines going from 16L all the way to 95L. We manufacture and order for all diesel engines for all applications for on highway, up highway, and for whatever you can think of, marine, whatever you can do with a diesel engine we're in that space.

Gavin: So, do you mind me asking what...what a 95L engine gets you as for seeing as that feels like a big one?

Teejay: Oh yeah, the 95L engine is one our latest products and it's a jump to see it pushes the limits of engineering in many ways. I believe it's the world's first or the world's fastest high-speed diesel engine, it puts out well over 4,000 horse powers. And it's used for a lot of locomotive applications so trains, plans for it to go into marine as well, so you know yachts and big boats and things like that so.

Gavin: It's actually small little details I always like to learn something of one of these podcasts here so that's my...that's my note for the day.

So, Teejay let's start from the beginning. You know, requirements management, you know a big subject, like what did requirements management look at Cummins let's say at the start of this journey, what was there, how would you describe it?

Teejay: Oh boy, that's a good one. I will start with a quote that I like, it's a quote by George Bernard Shaw that basically says that the single biggest problem in communication is the illusion that it has taken place. Requirement Management in Cummins was not the best just six years ago especially when I started, it wasn't the best before that. This is not our first try to get a handle around requirements and requirements management I will argue that we've historically we've done some requirements because you don't get to be the number one diesel engine manufacturer in the world by guessing right. So we did do some requirements but before we got into this space of requirements management here's what...what things look like. First of all, we had a lot of requirement activities. I specifically use the word activities because there were pockets, there were shadows of excellence, some people did it and some people did not do it, it wasn't across the board. There's a quote that we have here at Cummins and we call it "Unleashing The Power Of Cummins. Can you Gavin, can you guess what that is? Do you know what the power of Cummins is?

Gavin: It's the people I'm sure Teejay.

Teejay: Very good, very good. You're a smart one. A lot of folks that get that question talk about our engines. I mean those are pretty powerful and pull a lot of stuff but unleashing the power of Cummins is our people. That's our greatest asset, they are the cornerstone of what we do here, but just as you would look at that as a very good thing. In the world of requirements management that is actually not a sole good thing, I say this because these people have wonderful minds and can store immense amount of data in their brains and not have to...and not need a system and that works. The problem comes in when they move on that information goes with them. So that was one...that was one of our biggest problems. How do we retain our knowledge right, how do we...how do we keep it within the company? How do we make sure that it's available to the people that need it when they need it? There was a lot of tools being used. We had color coded excel documents, we had Microsoft Word documents, and not only did we have different formats and different tools we had different storage locations.

So it was all over the place, it was all over the place and so people who needed to know the requirements could not find the requirements, people who could find some kind of requirements didn't know what version of the requirements they were looking at right. How many times had those requirements turned and so that becomes the problem when you think about our products that's basically at different generations but it's an iteration involving product over time that we've perfected. When you think about that, knowing the history and knowing what requirements go with the version you're looking at right now at the moment become extremely important. The language, something as simple as the language was a problem, different terms were being used and different interpretations of those terms were being used. When I said, "A." Somebody heard "Z" and that's where my first quote comes into play. I said something, you heard something, you walked away, I said we're on the same page but we never were on the same page.

Gavin: Yeah, I think you're spot on there Teejay. I know I can empathize with you and I'm sure a lot of our audience are going to empathize with that too. I mean and you know like you said, whether it's lots of systems or different descriptions for the same words. I run into that every day you know in my role too. Were there any particular challenges that Cummins were having around requirements management that you're able to talk about with us?

Teejay: Again, it goes back to describing the world before what requirements management was right an inconsistent storage location that was...that was a big challenge right I... I talked about just now. We talk of this one social truth that was nonexistent a few years ago. We're very close to having that now but it wasn't there. Lack of change in notification right. We found a lot of research that has shown that over 30% of our failures and warranty defects come from changes that happened at the interface. That somebody changed the sub-system function and did not communicate it to the neighboring sub-systems right. Not being able to track emergent behavior, that's hard to do today but it's more of a reality today because our product is more complex. Broken traceability, how does your requirement at the component level affect the customer need. Is it dole plated have you missed something? Difficult leaves when you talk about a product has evolved, to reuse and being able to learn from the previous generation of that product becomes extremely important we couldn't do that very effectively.

Lack of standards requirements management structures and processes, everybody was doing their own tidbit, they all did what they thought was right thing to do, but it was different, and it was spread, and the spectrum was all over the place. Having a unified requirements management tool and then getting everybody on the same page even just with the tool. Metrics, measures, being able to collect those was extremely difficult. And you know, there was a system because we're a big system company right, there was a system that got run a few years ago I wanna say four years ago that identified about a 1.5 million dollar cost due to bad and missing requirements, right. One of our directors of engine business systems integration of innovation says it best. He says there are three bad things that can happen. The most dangerous the requirement is the one that you don't know exists. The second most dangerous requirement is the one that you know exists but choose to ignore it. And the third most dangerous requirement is the one without a valid check method. These are all things that we struggled with.

Gavin: Okay, so you mentioned earlier on in your description of your role Teejay that, I think you do like vendor selection and you know you're involved with the tool leg of that, you know, the three legged stool. I guess then you were involved with your setting out what you were looking for in this tool that you just talked about that you're really close to having the whole company standardized on. Like, how did you get that list together or did you sort of set you know here's our top five things we've got to have. I'm sure it was a much longer list now to be fair I'm probably simplifying it too much, but can you give the audience a little bit of an insight into you know what were the driving what were the North star, if you know what I mean, that sort of drove the selection process?

Teejay: Well, the selection process was...I wouldn't say was a long process, but we did do our due diligence. Like I said before this wasn't something that was at least when I got on the project and started leading this work, it wasn't something that was brand new to us, we had several folks who were trying to kick systematization area off in requirements management since 2005. So we had a lot of lessons learned from all of those efforts so when it came to trying to get this whole thing started out. The question wasn't really hey we need to figure out what our requirements are because we had a lot of those already. When we talk about you know basically trying to select a tool there were a couple of things that we took into consideration. First, it should just work, the tool should be simple, the tool should be intuitive, it should not take a Ph.D. degree to figure out how to use the tool. We were looking for a tool that we could embed our processes in right, a tool that dictated high-level processes but had the flexibility for different groups to adopt right. So yes we want to push a standard enterprise wide tool. We want to push enterprise processes but you know we want to keep those processes that are enterprise wide at a 5,000 foot level right, but the tool is still flexible enough to take in groups that didn't have as much resources and still keep them, you know, in a way that they could adhere to the process as well as relaunch projects that actually needed this kind of stuff. It should be the framework.

One of the mistakes we had before was the tool was so complex that we trained one super user and that person was the requirements management person. And that person was expected to do all the requirements management work, that doesn't work. Systems engineering requirements management is a team sport. Everybody has a role to play. So whatever tool we were going to use had to support collaboration and it had to be something that allowed everyone to be able to use the tool quite easily. And going back to you know being intuitive. Reuse was huge for us, requirements reuse, why do we have to spend four to six weeks or even more months trying to relearn things that we already know? So that was big for us, notifications, communications, the tool had to be able to drive things like notification based on good business logic at different levels. User, project system levels, we needed to be able to burgeon control our requirements because we've got platforms that we're working with. We need to be able to know what version of requirements were used when we're looking at a special version of code or of an ending. History, historical reports, we needed to be able to see how things evolved over time. And last but not least and I'm sure with any tool selection anywhere in the world at any big company, security and access control, was critical to selecting a tool. So, those were some of our top when you say North star. That's what I'll put out there.

Gavin: So Teejay like you said Cummins is a big company it's got 55,000 employees worldwide, everything though I've heard you say there again feels to me and I come from a experience of smaller companies, but everything you spoke right there feels it could be applicable in general to anybody who's doing requirements around a project, sort of, I wouldn't say size is immaterial but it feels like they're all good principles just in terms of looking at the subject full stock.

Teejay: I absolutely agree, Cummins is a pretty large company, in that sense, we operate a lot of businesses within the business. There's small projects and there's large projects. For us, we had both spectrums and so we had to have requirements that would cater to both of them that's where the flexibility comes into play.

Gavin: So the next question I have is again a big company I guess you really need a sort of implementation strategy right because it's not just rack up the first day and let's chuck on a few CD's and software and away we go. What, again, would have been your sort of pillars of that strategy?

Teejay : So to start, the one thing that we first did different, was redefine the scope, was understand that we did not need the perfect requirements management solution today. And like every other thing that we did the solution could evolve. Let's figure out what our maturity level is right now and let's deliver a solution that can support that. Another key implementation strategy that we did was the ownership of the initiative. In most companies, IT tends to take over these kinds of projects, but we decided on a different path. We were going to have engineering own design, configuration, and implementation of the requirements solution, because guess who was going to use the tool, engineering. IT still had a function to play, but they would not be the ones the primary implementers of the tool. And then we would focus on what I like to call the three pillars: people, process, tools.

In past attempts, we were too heavy on one of those. What happens when you have one pillar taller than the other pillars, you have a lopsided building. And that's what happens in mixing area too. And finally, there was a focus on the column. And what we call the column here is information flow and I'll explain that here we have to realize that as a technology development organization we did not actually manufacture engines. Our end product was not engines, I know hold your breath, I'm going to explain what I mean. Our manufacturing organization their product was engines, they assembled everything. Our end product was information and understanding how that information went from concept and how it transformed into different analysis tools, requirements, specs, drawings, and models. At the end of the day, our product was information. So understanding that value stream of information from where it's just an idea to something that manufacturing can take and make an engine out of that became information flow. And understanding that piece became extremely important.

Gavin: Like you said, it's that information is what then has a tendency to get reused in the future.

Teejay: Absolutely.

Gavin: It's an interesting perspective because I think if you were to ask somebody straight out and be well obviously the end product here is the engine but that's the sort of, that's the absolute output, but it's actually the data and like you said models and the specs and everything that goes into creating a great engine that you want to be able to reuse and be sure you can get at again easily.

Teejay: Absolutely.

Gavin: Okay, so let's...let's leap forward then into the future, and you're at the other side of this project or at least very close to being at the other side of this project, what is the current state of requirements management at Cummins as we sit here today?

Teejay: The current state of requirements basically right now all IT process deliverables are being supported. System requirements, we're able to capture our system requirements and manage our system requirements. Interface agreement solution, we've been able to put together a solution to manage interfaces and this has been, you know, praised as a real achievement here within the company, but even from external consultants like the requirements experts they saw a solution and said this is probably the most robust interface agreement solution that we have seen. Platform requirements management, everybody talks about managing platforms, product lines, that's where we're going, that's where our business is going. That's where you get efficiencies, that's where you get reuse, we have a solution within our tool, the PTC Integrity Lifecycle Manager, that can support platform requirements management. Compliance requirements solutions, mission compliance solutions, requirements change management that is compliant with CM2 and now finally getting to the point we are developing and we have a solution for global requirement libraries and now we are populating that information with our best knowledge from all our affiliates so that it can be continuously be reused and evolve over time as technology changes.

We have almost 4,000 trained users in our company across the world: USA, India, China, Europe is a major hub. We have other places as well. Beginning from 2017 all programs, all major programs and older now use PTC Integrity Lifecycle Manager as their requirements management tool and of recent has also become our verification management tool. So for the first time in CMI's history we have our verification plans and our requirements management within the same tool. They are able to do things that they could not do before where generating objective metrics and measurements. And this is important right because prior to this it was very subjective. Now we've got great people, we've got honest people so we know that they were doing the right thing but now it's objective. That was subjective, it's not objective. Now, it's objective it's not subjective. All the reports, all the metrics, all the managers, that we get out of the system are the same through every program so now our MRG our leadership are able to look at things relatively and not just in styles anymore. It's really changing the way we do business.

Gavin: And so, adoption to me is always this sort of clearest indicator that the path you've taken and the approach you've implemented is working.

Teejay: Absolutely.

Gavin: So I think big congrats to you, you know, in seeing those numbers grow and develop and like you said if the management is being able to see the numbers, they love the numbers.

Teejay: Show them metrics and measures.

Gavin: Right, exactly, so you've got an audience here listening to you now today. What would be the three to four things and I'm sure you've got a longer list, but what would be the three to four things that you would focus on if you were giving some guidance to somebody who's considering taking this step?

Teejay: First and foremost I would say a committed leader and a clear vision. Somebody who understands where the company needs to go who understands how complex the products are becoming and benefits that you can get because a lot of times especially when you start, most people don't get the vision and so it's important that they have that rock that sees it as clear as day and brings everybody towards that vision so that's the number one thing. Another one I like is organic growth through a value, look for easy wins, look for the low hanging fruits, as you create value people are going to ask for more, as you give them nuggets they can make the connection, they are smart, they're engineers, they will make the connections.

Gavin: Well they can build a bit of belief in where it's going as well. I think that's the other part you want to, you don't want people to think it's you know it's never....it's this great thing but you...we won't see anything for years. I think every business today needs that, let's call them the quick wins or the low hanging fruit but they need something that they can start to believe in you know.

Teejay: Absolutely, so something that they can see and touch and that's what I mean by organic growth through a value. Look to provide value, when you provide value people are going to ask for more value. Of course, it shouldn't be just completely anywhere it should be within your vision and your roadmap of what you're trying to do, but organic growth through a value. A third thing I would say is the 80/20 rule. Again, don't focus on building the perfect solution. Focus on delivering a solution that provides, how does the saying go 80% of the value is in 20% of the features. Your goal should be identifying that 20% feature and it may be different for different organizations and we have to learn that, but don't focus on the 100% solutions. The right solution, right now, for those people is what you should focus on. And finally, choose the right tool, not every tool is going to get you there. Figure out what your requirements are and figure out the vehicle that can support the goals that you have. For us, that was PTC Integrity Lifecycle Management.

Gavin: So what's next now, like you said you're 4,000 users in hundreds on requirements management on the platform on the system. What are the next elephants that are going to get eaten by the Cummins project team, the global systems team?

Teejay: Well, I'll go in Systems Engineering that's what we're going for. So right now, our focus is growing our new product development ecosystem. What other nuggets of data are out there that are part of this development process. That are being handled by other tools. Can we review the amount of tools that are involved in that process? Because once we do that we reduce our training efforts, we reduce our licensing costs, and a bunch of other things. So there's a lot focus on integration with our PLN solution which also happens to be owned by PTC.

And so we are, we're working on integrating with PTC windchill so Integrity Lifecycle Manager integrating within PTC Windchill. I just talked about platform requirements management. This simple fact is everybody's trying to crack this whole product line engineering and platform engineering, some people already have. We're still trying, you know, figure out how we make sense for our business.

We have initiative around model based product line engineering, M.B.P.L.E., so moving from documents based on everything to model-based, unified models, global models. And being able to emigrate that solution also with PTC Integrity Lifecycle Manager.

Our defect management system is currently outside of our systems engineering solution. We are in the process of integrating that as well, and also our risk management. So these are all critical components of new product development. And we're trying to reduce the amount of tools that are involved in trying to achieve these things. We're trying to make sure that all of that data, all of these nuggets of data, talk to each other. So you really have a true ecosystem where every bit of data has a meaningful relationship with another one, and it's something that you can easily consume. Integration is the name of the game, that's what we're doing, that's what's next.

Gavin: So, I guess you know to use the old line a lot done but a lot more to do. That's probably I guess a good summary, is it?

Teejay: Absolutely, I've learned that when you do your job well you don't get less work you get more work, so you said it right.

Gavin: Yep, that is truth. That's all for today's episode of "The Connective Engineer". Thank you very much Teejay for joining us and sharing the common story around requirements management, it's quite obvious your..the passion in your voice and it seems like there's some significant success being achieved there so again thank you for your time and the insight. And I'd just like to ask the audience make sure that you subscribe to our podcast on iTunes and we look forward to welcoming you back for our next episode. Thank you.



Gavin: Benvenuti a "The Connected Engineer", un podcast per ingegneri, progettisti e innovatori. Sono Gavin Quinlan, il moderatore, e ogni settimana inviterò degli esperti a partecipare per parlare delle sfide dello sviluppo prodotto attuali e delle tendenze future. L'argomento dell'episodio di oggi è la gestione del ciclo di vita del prodotto (o PLM) nel cloud. Attualmente sembra che tutto ciò con cui interagiamo si trovi nel cloud: i social network di cui facciamo parte, i servizi di musica che ascoltiamo su telefoni o computer, le foto imbarazzanti delle vacanze che non vogliamo vadano perdute e, sul fronte lavorativo, la posta elettronica e l'archivio dei nostri file. Il cloud è quasi ineludibile e, a dir la verità, perché dovremmo volerlo evitare considerando tutti i vantaggi che ci offre? È quindi del tutto comprensibile che anche il PLM stia passando al cloud. Oggi è con noi Stan Przybylinski, Vice President of Research di CIMdata, società leader nella ricerca, nella consulenza e nella formazione sul PLM. Buongiorno, Stan. Potresti per favore presentarti raccontandoci un po' il tuo background e dirci qualcosa di più su CIMdata?

Stan: Certo, Gavin. CIMdata esiste dai primi anni Ottanta ed è una società di ricerca di mercato e consulenza di gestione strategica che si concentra sul PLM. Abbiamo un'area di attività piuttosto ampia. Copriamo tutti gli aspetti di cui si occupano PTC e i suoi concorrenti, nonché diverse altre aree che da alcuni non vengono prese in considerazione. In ogni caso, collaboriamo con società come PTC e con aziende del settore industriale che hanno bisogno di informazioni di mercato per definire la propria strategia PLM e il percorso da seguire. Attualmente sono Vice President of Research di CIMdata da poco più di sei anni e mezzo. Le mie principali responsabilità riguardano la ricerca quantitativa e qualitativa globale che svolgiamo ogni anno. Prima di entrare in CIMdata, ho lavorato per 7 anni presso un vostro concorrente come manager responsabile dell'intelligence di mercato e competitiva e prima ancora come consulente per CIMdata. Nei 20 anni precedenti avevo lavorato soprattutto nel settore aerospaziale e della difesa, occupandomi in genere di software e ricerca e sviluppo.

Gavin: Quindi una vasta esperienza nel campo del PLM, Stan. Per iniziare, chiariamo al nostro pubblico alcuni termini correlati al PLM e al cloud che costituiranno la base della nostra chiacchierata. Spesso termini come cloud e SaaS vengono usati in modo intercambiabile, ma non sono sinonimi. Spieghiamo quindi che cosa sono il PLM e il cloud e, se possibile, esaminiamo alcune espressioni fornendo per ognuna una spiegazione semplice.

Stan: Bene. Spesso sentiamo l'espressione "as a Service", che serve per definire il livello di servizio offerto nel cloud. Sono state fornite molte definizioni in proposito. Il National Institute of Standards and Technology (NIST) degli Stati Uniti offre una definizione che usiamo volentieri perché è piuttosto semplice. Il livello più basso è l'infrastruttura come servizio o IaaS. Questo livello solitamente non riguarda la maggior parte dei fornitori di soluzioni PLM in quanto fornisce l'hardware essenziale, lasciando al cliente l'implementazione di tutti i componenti. Molti di loro operano al livello successivo della piattaforma come servizio o PaaS. Chi esegue un proprio prodotto software su una piattaforma come AWS rientra in questo modello. Al livello Software as a Service (SaaS) successivo viene effettivamente offerto l'accesso a un determinato insieme di software come servizio. Il software non viene fornito nell'ambiente locale, in cui attualmente sono installate molte soluzioni. Di fatto si fornisce l'accesso al software tramite un browser Web o un'interfaccia utente di altro tipo.

Gavin: Bene. Hai parlato di ambiente locale. Chiariamo questo concetto. Che cosa significa?

Stan: Per ambiente locale si intende ciò che viene eseguito sul proprio computer o su più computer connessi ai server di cui dispone l'azienda. Nella definizione di CIMdata, è il software installato in locale che è prevalente in molte soluzioni PLM.

Gavin: Ok. Altre due espressioni che continuo a incontrare informandomi in materia di PLM e cloud sono "a tenant singolo" e "multi-tenant". Probabilmente sono aspetti più da addetti ai lavori, ma potrebbe essere comunque utile per il pubblico capire che cosa significhino.

Stan: Certo. "A tenant singolo" significa che si ha un server dedicato in cui viene eseguito esclusivamente quel software specifico. "Multi-tenant" è la possibilità per un prodotto software... Si pensi a un database in cui più segmenti costituiscono tenant diversi per l'azienda industriale A, l'azienda industriale B e l'azienda industriale C. I dati sono separati, ma l'esecuzione avviene nello stesso database e, soprattutto, viene offerto lo stesso tipo di funzionalità a tutti gli utenti. L'aspetto chiave del livello SaaS è che garantisce un prodotto coerente, per la maggior parte.

Gavin: Ok. Penso sia un punto di partenza molto utile per il pubblico. Sarebbe corretto sostenere che, allo stato attuale, il PLM nella maggior parte delle organizzazioni viene ancora eseguito in locale?

Stan: Sì. O meglio, la parte della gestione dei dati risale agli anni Ottanta e all'epoca riguardava team oppure gruppi di progettazione tecnica più grandi, mentre adesso le soluzioni sono destinate ad aziende globali. Il concetto di base però è ancora lo stesso. L'elemento principale viene in genere eseguito in locale e molto spesso viene personalizzato dalle aziende nell'ambito dell'installazione locale. L'altro aspetto importante è l'integrazione con gli altri sistemi aziendali e le altre origini dati, nonché con i clienti e le supply chain, se l'azienda decide di adottare questo approccio.

Gavin: Capito. Quali sono i fattori che spingono le organizzazioni a prendere in considerazione il cloud o hanno indotto le organizzazioni che sono passate al cloud a procedere in tal senso?

Stan: Beh, una delle ultime cose che ho menzionato: la presenza di catene del valore e relazioni con i clienti più complesse. È molto più semplice gestirle con una soluzione basata su cloud, invece di preoccuparsi dei propri firewall. Un altro fattore importante sono i costi iniziali più bassi. Mi occupo di questo da oltre 20 anni e in genere chi prende l'iniziativa si trova a dover chiedere budget importanti... Ora le soluzioni disponibili sono molto più sostenibili Si può partire semplicemente con una carta di credito. Se ad esempio si vogliono comprare quattro postazioni per qualche mese, è possibile farlo con una carta di credito, personale o aziendale. La dinamica interessante secondo me è questa: è cambiato il livello del processo decisionale di partenza. Lo stesso vale per i costi iniziali più bassi. I tempi di implementazione possono essere scelti liberamente.

Gavin: Suppongo che i costi iniziali inferiori determinino in qualche modo un time-to-value più rapido per questo test.

Stan: Sì. Certo, se si tiene sempre ben presente l'obiettivo. È necessario restare concentrati sui vantaggi in termini di processi, dati e collaborazione che si desiderano ottenere. Comunque sì, è possibile raggiungere l'obiettivo molto più rapidamente perché non esiste l'ostacolo degli aspetti tecnici.

Gavin: Bene. Questo offre anche flessibilità perché è possibile espandere o ridurre l'ambiente a seconda di come vanno le cose.

Stan: Esattamente. In molti tipi di aziende è davvero necessario, perché i progetti sono molto flessibili. Nelle attività aziendali basate su progetti, soprattutto, si ha un afflusso di talenti destinati a occuparsi di aspetti diversi. Essere in grado di supportarli rapidamente senza contattare il reparto IT e senza tutti questi problemi è davvero un enorme vantaggio per molti tipi di aziende.

Gavin: Bene. Ne consegue una domanda importante: quale ruolo svolge il reparto IT in questo passaggio al cloud? Il suo ruolo è destinato cambiare in relazione al supporto della piattaforma?

Stan: Dipende dall'azienda. In alcune, l'implementazione del PLM viene gestita dal reparto IT. In molte aziende, il reparto IT collabora, nel bene e nel male, con i vertici societari. Esistono molti diversi assetti organizzativi e questo può ridurre la componente politica. Ho sempre sostenuto che il PLM non sia un prodotto ma politica e la politica IT è una parte importante del cloud. Alcuni credono che l'implementazione verrà rallentata perché il reparto IT si opporrà alla riduzione del proprio budget determinata dal fatto che non verrà più usato un server ma verranno semplicemente acquistate licenze software eseguite altrove.

Gavin: Giusto. Suppongo questo sia l'aspetto più ovvio. Come hai detto tu, i costi iniziali saranno inferiori. A lungo termine, inoltre, non saranno necessari grandi acquisti di hardware.

Stan: Esatto. Non è necessario comprare grandi quantità di hardware destinate a perdere valore subito dopo l'acquisto. Esistono molti componenti e, procedendo nel modo giusto, non sono necessarie molte installazioni sui computer locali. Questa enorme spesa può essere evitata.

Gavin: Così come i costi e i tempi di implementazione.

Stan: Esatto. Come i costi di implementazione e di migrazione. Se tutti utilizzano lo stesso insieme di codice di base, la migrazione da una versione all'altra della maggior parte delle applicazioni viene eseguita costantemente. Se si usa Facebook o qualcosa di simile, si può vedere che da un minuto all'altro compaiono nuovi pulsanti e altri elementi dell'interfaccia utente. Dovrebbe quindi essere piuttosto semplice fare altrettanto nei modelli SaaS, eliminando un grosso problema per i responsabili locali.

Gavin: A quanto ho capito, CIMdata ha svolto ricerche sui fornitori di software e sui rispettivi piani in relazione al numero o alla percentuale di clienti che ha già effettuato la transizione o la sta prendendo in seria considerazione. Che cosa puoi raccontarci in proposito?

Stan: Alcuni mesi fa abbiamo effettivamente lanciato un progetto con PTC e molti suoi concorrenti nel mercato del PLM in merito all'adozione del cloud. La premessa era che l'adozione per sistemi aziendali come ERP e CRM è pari, secondo alcuni, addirittura al 20% o 25% del mercato aziendale, mentre il PLM, complessivamente, non si avvicina assolutamente a una percentuale del genere. Qual è quindi la differenza? Intendevamo svolgere qualche ricerca per stabilire se le organizzazioni che procedono all'adozione stiano già utilizzando il cloud. Esaminando alcuni dati (abbiamo ormai diverse centinaia di risposte) risulta che il 35% degli utenti utilizza soluzioni basate su cloud. La principale è Salesforce. Il cloud quindi viene utilizzato, ma non tanto per il PLM, finora, e non così velocemente. Esistono molti utenti, così come molti fornitori, ma non hanno ancora raggiunto la massa critica.

Gavin: La ricerca ha affrontato anche le prospettive della transizione per gli anni a venire?

Stan: Sì. Organizziamo in tutto il mondo un evento, il PLM Market & Industry Forum, cui partecipano responsabili delle organizzazioni fornitrici, come PTC, le aziende concorrenti e i system integrator, e conduciamo sondaggi tra coloro la cui posizione offre una buona visione delle vendite e dei problemi affrontati dai rispettivi clienti chiedendo quale sia la percentuale dei clienti che sta già usando il cloud o sta valutando la possibilità di farlo. Quasi il 45% lo sta almeno prendendo in considerazione in modo concreto. Quando però abbiamo posto la stessa domanda chiedendo "Se così tante persone ora esitano, come sarà la situazione tra tre anni?", è risultato che solo il 18% starà ancora tastando il terreno. Tutti gli altri staranno procedendo con decisione o potrebbero aver già completato la transizione. Gli utenti sono pronti a procedere, a quanto sembra.

Gavin: Bene. Prendendo spunto da una parola che hai appena usato, vorrei parlare delle esitazioni. Dal mio punto di vista, lavorando con organizzazioni di progettazione tecnica prevalentemente di piccole dimensioni, sono sempre stato ben consapevole delle preoccupazioni che sorgono quando la proprietà intellettuale deve essere collocata al di fuori delle quattro mura dell'azienda. Abbiamo sempre sostenuto nelle discussioni che all'esterno è probabilmente più al sicuro che all'interno, ma come viene affrontato questo aspetto?

Stan: La sicurezza è un grosso problema. Purtroppo, quando si verifica un evento pubblico correlato alla sicurezza, i prodotti basati su cloud, che siano coinvolti o meno, fanno un piccolo passo indietro. Ciò che dico sempre, però, è che la sicurezza che si ottiene in queste soluzioni è di gran lunga superiore perché esistono più livelli prima del software e tutti si occupano della sicurezza. Tutti investono più di quanto possa spendere l'azienda cliente e non hanno computer sulle scrivanie con porte USB da cui si possono facilmente estrarre dati, né inviano file per posta elettronica. La sicurezza quindi è un grosso problema, ma esaminando davvero la questione si deve ammettere che non è poi così grosso. Esistono alcuni problemi tecnici, ma niente che non sia stato superato.

Gavin: Credo che un altro aspetto sia che le persone spesso pensano che una soluzione cloud venga fornita come un unico prodotto valido per tutti. Nel caso dei social network, ad esempio, a tutti viene visualizzata la stessa interfaccia. È giusto che i nostri ascoltatori la pensino così o la situazione in realtà è diversa?

Stan: Beh, esiste una differenza tra adattamento e personalizzazione. Anche in Facebook è possibile adattare le funzionalità disponibili, ad esempio escludere determinate persone o controllare in parte ciò che viene visualizzato. Lo stesso vale anche per un'istanza SaaS singola o multi-tenant. Si può impostare l'aspetto desiderato o rendere alcune funzionalità più facilmente accessibili e altre no, a seconda dell'utente. È possibile apportare molte di queste modifiche, ma sono comunque circoscritte dal fornitore del software. Il fornitore deve pianificare la possibilità di farlo, mentre adesso nell'ambiente locale, a seconda di quanto denaro si ha, si può incaricare un system integrator e ottenere esattamente ciò che si desidera.

Questo è un vantaggio e al tempo stesso uno svantaggio del PLM in generale, perché si creano istanze e implementazioni che non possono essere spostate. Sono come bloccate. Non si interviene finché non si verificano problemi. È difficile che venga presa la decisione di agire fintanto che si ha una soluzione che funziona. Prendere un'iniziativa del genere è un impegno. È una decisione che deve essere presa in modo proattivo e apparentemente si passa all'azione principalmente a causa di altre pressioni aziendali. È necessaria una maggiore flessibilità aziendale, che deve essere supportata più rapidamente. Il cloud è molto efficiente sotto questo punto di vista. La transizione sarà agevolata dal fatto che le principali limitazioni tecniche, in termini di dimensioni dei file e aspetti simili, sono state superate da diversi fornitori. Non sono più un grosso problema come un tempo.

Gavin: Esatto. La rete Internet è stata migliorata, è aumentata la disponibilità della larghezza di banda e così via. Passiamo agli aspetti positivi del bilancio e abbandoniamo quindi le esitazioni. Alcuni dei vantaggi per l'azienda in generale sono stati citati esaminando i fattori determinanti di base. Quali sono i netti vantaggi di cui intendono usufruire i vostri clienti che stanno prendendo in considerazione la transizione?

Stan: Nella letteratura sulla strategia degli anni Ottanta e Novanta si parlava dell'importanza di concentrarsi sulle competenze di base, ma purtroppo molte aziende si occupano di IT nonostante non sia l'elemento principale della loro attività. Adottando questo approccio, si può ridurre l'attività IT più di quanto sarebbe altrimenti possibile. A seconda del livello, sarà ancora presente un certo grado di supporto, ma sarà nettamente inferiore rispetto all'esecuzione di propri server e all'installazione di software locale sia nei server che nei computer client in sede. L'azienda non dovrà più occuparsi di questo. Ne deriverà, come abbiano accennato, una maggiore flessibilità. Quando ho iniziato a occuparmi dell'argomento quasi 20 anni fa, le aziende eseguivano queste implementazioni acquistando immediatamente tutti i prodotti software di cui avrebbero potuto avere bisogno per l'intera implementazione per usufruire di un'offerta migliore.

Non è più così. Le aziende desiderano maggiore flessibilità e questo modello è estremamente flessibile. Si possono comprare 10 postazioni oppure 20. In un colloquio, un cliente di PTC ha evidenziato la possibilità di implementare rapidamente istanze a scopo di prova. "Se abbiamo un team di progetto per cui potrebbe essere opportuno estendere l'implementazione per coprire un determinato aspetto della gestione di progetto o della collaborazione, posso ottenere un'istanza e disporre per 3 mesi di 20 licenze che verranno utilizzate dai collaboratori allo scopo". Questo nell'ambiente locale avrebbe comportato enormi complicazioni. Adesso è molto facile poterlo fare, provare e utilizzare il software per gli scopi desiderati, imparare ciò di cui si ha bisogno e quindi smettere, se si decide che è meglio così.

Gavin: Mi sembra estremamente logico. Immagino anche che, come abbiamo detto, la rapidità di implementazione garantisca altri netti vantaggi in termini di time-to-value o time-to-market.

Stan: Certo. L'implementazione è però comunque necessaria. Noi di CIMdata parliamo sempre di persone, processi e tecnologia. In questo modo, la tecnologia, per molte persone, non costituisce più un problema. È ancora necessario, però, gestire i processi e la collaborazione. PTC e i suoi partner di canale dovranno continuare a svolgere questo ruolo perché sarà ancora necessario per progredire. Noi parliamo di Facebook, ma Facebook e il PLM non sono la stessa cosa. Sono entrambi soluzioni Software as a Service, ma con il software correlato al PLM si affrontano problemi molto più complessi.

Gavin: Sì. Suppongo dipenda dal tipo di problemi. Per quanto riguarda il lavoro quotidiano del singolo... A livello aziendale sembra molto chiaro: non è più necessario occuparsi di IT ed è possibile completare l'implementazione ed essere operativi più rapidamente. Questi sono tutti vantaggi tangibili estremamente utili che vengono realizzati dall'azienda, ma per quanto riguarda i singoli membri del team di sviluppo prodotto? Detesto usare questa espressione, ma l'ascoltatore medio del podcast quali vantaggi può trarre personalmente dall'utilizzo del cloud?

Stan: Un vantaggio importante è la possibilità di accedere a ciò di cui ha bisogno quando necessario. In un ambiente locale può essere molto difficile fornire nuove funzionalità, soprattutto temporaneamente. In passato, se si voleva usare qualcosa era necessaria una licenza perpetua in locale, che veniva sfruttata ad esempio per un anno pagando la manutenzione per un determinato periodo di tempo. Adesso, se si desidera utilizzare un particolare software CAM o di realtà aumentata, è tutto molto più semplice. Questo ci riconduce di nuovo al processo decisionale. Dipende dall'azienda, ovviamente, ma può essere molto più facile perché la decisione può essere presa a un livello molto più basso. Tutto il tempo sottratto alle attività IT è tempo in più che i progettisti possono dedicare all'innovazione. Così come i sistemi sono stati progettati, la gestione dei dati rappresenta sempre per i progettisti un grosso fastidio che interrompe il flusso del loro lavoro. Sono molto creativi ed è preferibile non infastidirli troppo.

Gavin: Certo. Finché non usano il disegno sbagliato.

Stan: Proprio così. Questo è il problema. Le persone non se ne occupano volentieri, ma poi finiscono per utilizzare il disegno sbagliato e sprecare diverse ore. Meglio evitarlo. La progettazione dei sistemi è stata migliorata e il cloud garantisce maggiore accessibilità e la possibilità di gestire tutto questo in modo più efficiente, semplificando molto la vita, si spera. Un altro vantaggio importante è la possibilità di accedere alle nuove versioni del software utilizzato quotidianamente. Con l'implementazione di una nuova versione in locale, in base alle roadmap gli utenti sono sempre indietro di due o tre release. Quando vengono introdotte nuove funzionalità, sappiamo come operano le aziende di software: vogliono che gli utenti eseguano in locale le release più recenti e quindi introducono le nuove funzionalità solo nelle ultime release. Questo con il cloud non accade. È possibile accedere molto rapidamente alle nuove funzionalità introdotte.

Gavin: Tornando alla mia esperienza, ovviamente noi fornivamo inizialmente il PLM al team di progettazione tecnica, i cui membri incontravano le maggiori difficoltà nel garantire che le versioni fossero corrette, che ai fornitori venissero inviati i dati giusti e così via. Uno dei valori che tentavamo di trasmettere, però, era il fatto che i loro dati avrebbero potuto essere usati in modo molto più efficiente nel resto dell'azienda e suppongo che l'utilizzo del cloud agevolerebbe anche l'implementazione al di fuori del reparto di progettazione tecnica.

Stan: Certamente. Le catene del valore con i fornitori e i clienti sono ormai molto dinamiche, soprattutto adesso con i prodotti intelligenti e connessi. È una delle caratteristiche più interessanti di tutto questo. Come dico sempre, il lavoro sui prodotti prima terminava con la distribuzione. Si progettava un prodotto, destinato a un determinato consumatore con un particolare insieme di funzionalità, e lo si distribuiva. Al massimo si ricevevano alcune informazioni correlate ai resi, alle richieste di interventi in garanzia o ai reclami. Adesso invece si sa esattamente come i prodotti vengono utilizzati. Il cloud consente di mediare molto più facilmente questo tipo di comunicazione e gestione dei dati e di adottare tutto questo gradualmente. "Adesso che ho questo materiale, sono pronto per l'IoT", "Sono pronto per altri tipi di funzionalità", "Privilegerò l'assistenza sul campo", "Non mi limiterò alla semplice vendita di un prodotto, sono più interessati alle parti di ricambio" oppure "Proverò qualcosa di nuovo". È molto più facile espandere la visione dell'azienda e di ciò che si intende offrire quando si utilizza il cloud.

Gavin: Certo, ha ripercussioni sul flusso di lavoro e sulle attività svolte. Come ho già detto, in base alla mia esperienza personale, anche avere accesso agli aggiornamenti o alle release più recenti di un prodotto può essere in genere molto utile. Negli episodi precedenti abbiamo viziato, per così dire, il nostro pubblico con i racconti di alcuni clienti e colleghi che hanno illustrato i propri successi e come hanno sfruttato in modo più efficiente la tecnologia disponibile. Hai un esempio di un'azienda che ha effettuato la transizione? Se preferisci proteggerne il nome, magari potresti dare un'idea del tipo di azienda e dei successi che ha ottenuto per offrire al pubblico qualcosa di più tangibile.

Stan: Sì. In effetti ho avuto un colloquio con una grande azienda nordamericana di beni di largo consumo, di cui non mi è consentito fare il nome, che utilizza soluzioni basate su cloud di PTC ormai da qualche tempo. Recentemente ha affrontato un'espansione significativa della propria implementazione, ma sfrutta le vostre soluzioni basate su cloud da tempo. Per questa azienda, sono un mezzo per ottenere un time-to-value più rapido e un'implementazione graduale molto più semplice dei diversi prodotti PTC coinvolti. Ha riscontrato vantaggi significativi per i propri utenti perché possono diventare operativi più rapidamente. Non ho però numeri quantitativi da fornire.

Gavin: Capisco. Oggi, Stan, hai dato ai nostri ascoltatori molto su cui riflettere, sia che stiano valutando la possibilità di implementare il PLM a un livello base o di reimplementarlo nel cloud. Se non sbaglio, CIMdata offre programmi di formazione dedicati sulle best practice per il PLM che potranno... [voci sovrapposte]

Stan: Esatto. Offriamo un programma di certificazione per il PLM. La maggior parte dei nostri eventi riguarda le best practice. Ne terremo prossimamente uno denominato "PLM Road Map". A questi eventi intervengono esperti che illustrano le proprie iniziative in merito alla tecnologia e ai processi PLM. È un modo molto efficace per entrare in contatto con queste persone. Abbiamo inoltre i programmi di certificazione cui ho accennato. È questo il tipo di lavoro che svolgiamo con le aziende del settore industriale per aiutarle a diventare rapidamente operative. Teniamo anche workshop e altri corsi di questo tipo.

Gavin: È tutto per questo episodio di "The Connected Engineer". Ringrazio molto Stan per aver partecipato e averci offerto un'eccellente panoramica e molte informazioni utili sul PLM e il cloud. Mi sembra evidente che, come in tutti gli esempi della nostra vita quotidiana in cui il cloud è già presente e ampiamente sfruttato, il PLM nel cloud sia un'area destinata ad affermarsi che potrà offrire vantaggi significativi rispetto all'installazione in locale. Grazie per averci ascoltato. Ricordate di iscrivervi al nostro podcast. Saremo lieti di ritrovarvi per il nostro prossimo episodio.


Gavin Quinlan: Welcome to The Connected Engineer, a podcast for engineers, designers, and innovators. I'm your host Gavin Quinlan and each week I'll invite experts to discuss the product development challenges that companies are facing today, as well as the emerging technology trends that are impacting product development. The topic of today's episode is data management. Sounds simple, right? Everyone knows that you need to manage data well to help make good decisions quickly. Today, we're joined by Anne Moxey and Joe Mathias of Nucleus Research to tell us why data management is such a hot topic right now. Joe, could you introduce yourself? Tell us a bit about your background, and tell us a bit more about Nucleus Research.

Joe Mathias: Absolutely Gavin. So I am a research analyst at Nucleus Research in the data and analytics team working with Anne. And so what we really do, is look at different technology deployments that use data to improve businesses, to improve workflow productivity and ultimately to get a positive return on their investment. And so to tell you a little bit about Nucleus Research, we're a third party independent research firm, we've been located in Boston for 17 years now and really the main message that we're trying to help businesses understand is how they can deploy technology to get a positive ROI. So really a numbers-focused business.

Gavin: Anne, can I ask maybe the same question of you? Just a little introduction for yourself.

Anne Moxey: Sure I'm senior analyst here a Nucleus, also working with Joe in the data and analytics team. And just to add on a little bit more background of Nucleus Research and what we focus on. We are NASBA certified which is the National Association of State Board of Accountancy. So what that means is that we have a very formal methodology in how we analyze data management solution deployments and looking at the return on investment of specific solutions.

Gavin: Thanks very much. Okay Joe, let's get started then. How would you define data management today?

Joe: So to understand data management, you have to understand how much data is being generated by a company. And really it's being generated by all sorts of departments all over the business. So you have data that is often what we call siloed. And by siloed, I mean sort of stuck in one branch of a business that isn't accessible to other businesses. And data can be siloed in sales, marketing, at the executive team operations, across many different aspects of the business. So when we talk about data management, what we're talking about is a couple of different things.

One, getting clean access to the data making sure that the data is prepared in a way that is accessible through applications on your computer so that someone from the finance department can actually look at the data. Two, that the data is collected in one sort of central location. We call that one source of truth that you have one version of the data. What can happen is, we'll talk about this a little bit later, you can get multiple versions that are different, and that can cause problems for understanding the data. And three, that the data is convenient to access. So you're able to figure out what piece of data you want and to get to it quickly through that sort of interface in the application that's intuitive.

Gavin: Okay. Anne, is there anything let's say from an engineering perspective that you could add to that? The commentary from Joe. I mean we talked about the business. I mean again are some of our audience or most of our audience could be coming from an engineering background. Is there anything else you'd like to add?

Anne: Sure. So one of the main benefits you'll be seeing from a data management deployment from an engineering perspective is looking at an increase in productivity. So you're able to access the documents and materials that you need much faster, and you know that it's the correct documents as well. There's less duplication, there's less redundant data throughout the enterprise. So you're able to get trustworthy, clean data at the time that you need it.

Gavin: So Anne, if I could follow that question then with, could you tell us what the current trends you're seeing in the area of data management as you're working with clients today are?

Anne: Well, there's an increase in the complexity with really the three Vs of data management. Volume which is the amount of data that's coming in, Variety of data whether that's structured or unstructured perhaps coming from blogs, perhaps coming from mobile devices. A lot of different types of data coming in. And then also the Velocity of data as to the speed at which it's coming into an application or coming in from different data sources. So that's increased exponentially in the recent years because each person is generating more and more data. And that means that data management solutions are really important topic for any company with big data because as it increases, it's just not possible to manually control all of it efficiently.

Gavin: Yeah. Makes total sense. So going back to let's say efficient then inefficiency Joe, like from the companies that you are working with, what would you say are the biggest challenges they're experiencing with doing good data management?

Joe: Like I had said before, siloing is a huge issue. So if you imagine a company when they start out, a smaller company is able to manage a lot of their information in something as simple as an Excel spreadsheet. Maybe that's a million lines, but when you have a relatively small company with only a few departments, it's easy to keep one application holding all of the relevant data. But once you get a design team, and a production team, and a sales team, and then an executive team all growing and generating their own reports and generating their own data, it doesn't automatically share itself.

And so it ends up that it grows to become this beast that gets out of control, where all of a sudden if the company wakes up one day and they realize, ''Wow, I'm trying to access data that is in a totally different department in their individual application.'' And so we've seen some deployments where they're using a, for example, like a design application to track the product lifecycle, like a PLM solution, product lifecycle management solution. But what the problem is, is that data isn't being translated over into the department that's handling the bill of material.

So you have almost a reduplication of the design team who obviously has to deal with the materials that they're using to build the product that they're doing, but then you also have the buying team who has to deal with their own lists. And so that really the goal for a company that has all these siloed data sections is to just integrate them into one central application.

So the biggest challenge is definitely getting everybody on the same page within the company. And it's going to have a ripple effect where not only does it make it easier to look for data that you were looking for already, but that you'll find data that you didn't know would be beneficial to you within your own department.

Gavin: Sure. I mean even from a personal perspective, I think we all struggle with finding data today whether it's between your email application windows explorer or whatever it's gonna be. So I can only imagine it multiplies when you start to add, like you said, more data more systems.

So Anne, what are our challenges? I mean okay, silos is one thing, but I'm guessing there's probably more areas that even the people who have managed to address the silo side of things that they are challenged with.

Anne: Sure. So a lot of the times what happens when you don't have an effective data management solution is that you end up with unwanted duplication. So if you're trying to do an analysis, a lot of the times a user will pull data into a separate sheet that they can then perform an analysis on, that just means that there's more copies. It means that there can be conflicting data messages coming across. And it just becomes difficult to figure out what is correct because it's all separate. There's nothing consistent and being managed and monitored across your organization.

Gavin: And I guess then you run the risk that you make a decision based on out-of-date information rather than what's up-to-date.

Anne: Precisely. And the other thing too to add on to that is also there's a governance perspective to it as well. You wanna make sure that the right people are accessing the right information, and then you know how they're using it as well. If you're not sure who has access to a data, if they're creating duplications, it can really get out of hand. It can perhaps leave the organization. That would be the worst case scenario in many situations. Or you can end up performing analysis that just has, as you mentioned, the out-of-date data.

Gavin: Okay. So as in most situations by speaking about the challenges we unearth what the particular benefits can be of having good data management. So Joe, again in terms of removing silos, we've sort of indicated that. What other areas are your customers and your clients looking to data management to solve today? What other benefits are they looking to realize from us?

Joe: So this goes along a little bit with the Nucleus Research Methodology which is that we're not only looking at the direct benefits, but also at the indirect benefits. And what we find with data management solutions in our case studies is that a lot of those benefits really are indirect. And so when I talk about just to give a quick background, direct benefits are things like avoided hardware costs. We have a new storage solution that's so much better we need fewer servers. Okay, that's a direct benefit. And it's easy when a company looks at a technology solution to see those benefits right off the bat. What's a little bit harder to see is the indirect benefits. And that typically manifests itself in productivity. Sometimes it's also avoided hires things like that of new employees.

So productivity when we do our data management ROI case studies, we find that productivity is really the biggest benefit that the companies are seeing. And the reason for that is really about the consolidation of a lot of silos. And the way that users from multiple departments can interact with that single source of truth in one application. Because data being so...different departments having different silos of data, it's usually a result of them having different applications. But not only does it make it harder to access data from other people's departments, but the application itself are different in each department.

So the user experience, the way that they are interacting with the data through the application is different in each circumstance. So productivity can be generated out of a single data management solution in a couple of ways. One, it's faster to access the data. So people are performing data analysis or finding data a certain percentage more quickly. And two, they're more accustomed to a single user experience. And so that means Fred from the accounting department can tell Bill from the design department. ''Yeah, go into your data management solution. Okay look at the hub and go access the bill of materials tab and then you'll see exactly what I'm talking about.''

So having the whole company on the same page visually and from a usability perspective, makes it inter-department communication easier, makes it easier to look things up, and it makes it easier to generate reports that are intelligible in each different department.

Gavin: So one of the questions I have then I guess, and maybe this is something you can help you understand better would be, how do companies understand their extra productivity? I mean sometimes I'm not sure companies really know how much time they spend on certain tasks today. So how are they able to estimate or quantify the improvements on that?

Anne: So that can be a different analysis for every company. Sometimes, you'll want to speak to the engineers themselves than say, ''Previously, what were you doing, how are you getting the information, how long did it take for you to perform a query search for example and then what are you doing today?'' And that could be something that you look at a single day as an example, you could be looking at a week as an example and then you would put on the correction coefficient on to the productivity benefits as well. Which Nucleus, our number would be point 75.

So we find that 75% of employees time does go directly into working more. The other 25% could go to Facebook or YouTube whatever else perhaps getting more coffee whatever works. So yeah, productivity is a bit trickier one to define, but it just it's a matter of talking through the process with the end-user themselves.

Gavin: Yeah because I can imagine like I said, it's the indirect one, but it's probably the one that has the greatest overall impact because that extra save time could be used hopefully, not on Facebook. Hopefully, it's on the next product design or the next innovation whatever it's going to be. Not that there anything against this book to be clear. I wanted to go back, Joe, to just one point you made about direct costs. So you said something like hardware would be something. How is the Cloud and the move to the Cloud impacting the whole data management strategy of companies because now that the days of perhaps buying servers are about to disappear?

Joe: Absolutely. It has a big impact actually. And a lot of our new case studies or a more recent case studies not just in data management, but in things like analytics, corporate performance management, things like that everyone is really switching to the Cloud. So the old issues with the Cloud that were being raised a few years ago is the cloud safe, is it really going to be the best way for me to manage my data, I kind of feel more comfortable with the servers in my basement. Those issues have really been worked out. People feel comfortable that the Cloud is at least as safe if not more safe and security in terms of data security for the company than on-premise solutions.

So people are really getting on-board with Cloud versus on-premise. The big benefits of Cloud, there's a couple of benefits. One, is that you have reduced hardware costs. And so not just reduced in that you don't have to pay for them really the cost is offloaded to the vendor, but also that they're more flexible. So a lot of software as a service SAS solutions, allow you to scale what you need in terms of storage or capacity with at an almost real-time rate. So when you buy a bunch of servers, you're locked into what the server capacity is. So if you're using less than, well that's inefficient. If you're using if you need to use more than that, well, you really have to make a sort of conscious investment in that.

With a lot of software as a service, you're able to snap your fingers increase decrease your capacity at will. So it saves people upfront on hardware costs, but also over time it allows them to use their capacity more efficiently. And the second thing is that companies are able to receive updates much more quickly. And a lot of the maintenance is offloaded to the vendor. So with the vendor controlling things from their side, they're able to update basically in real-time almost. It's much faster than on-premise.

And when things go wrong the vendor is right on the other side. They don't need to sort of come in and deal with your on-premise system where you don't need to have an internal team managing and fully managed by them on the back end.

Gavin: And I guess to go back to your three Vs then because of the velocity and the variety side of things and these new data streams that may be coming onboard. That agility I guess is super important.

Anne: It is. And actually, we've done quite a bit of research on quantifying the value of Cloud versus on-premise as well. So this year we published a piece. The Cloud delivers 2.1 times more ROI than on premise solutions on average. And that's as Joe was mentioning because of a lower initial total cost of ownership for TCO and then lower ongoing maintenance costs, but also the ability to update more quickly and as needed. On average, Cloud solutions have a much faster release cadence which means that users are able to access the most recent technology innovations as opposed to on-premise solutions which are static and stuck in more of a historical space for a much longer period of time.

Gavin: Maybe just to follow through them, the ROI since you talked about a 2X for Cloud, in terms of the research that you've done for your clients and...have you been able to ascertain a more overall ROI for data management itself? I mean I know it's a very big subject and everybody's got very specific unique examples. I guess you've managed to collate some data together. Perhaps you could share with the listener.

Anne: Sure. So we went back through the last five years or so of ROI case studies on data management solutions specifically. And we found that on average they return $14.42 for every dollar spent. Now, that's a considerably higher number than most other core enterprise applications. And that's because data is pervasive and really impacts every part of an organization. So in other words, if you put in data management system, you are not only helping out the IT crew, you're helping out your users of CRM, ERP, CPM and really every other application as well, because you're providing cleaner data. And you're also lowering those data silos for the end users themselves.

Gavin: So Joe, one of the acronyms that Anne did mention there was PLM. And I think that's an acronym that might be of interest to the podcast listeners. So could you perhaps address if there is any particular situations or examples that Nucleus have worked with customers on in that area?

Joe: Absolutely so PLM Product Lifecycle Management, ALM Application Lifecycle Management, are two good examples of how data management can have an industry specific impact. And so when we talk about...Anne talked about different core enterprise applications that are affected by data management, things like analytics, enterprise resource planning, ERP, CPM every company has those sort of applications and they benefit from having the data from those different applications codified and brought together. But there are also specific industries that have their own specific departments that would benefit from data management.

So great example is manufacturing. And at Nucleus, we've worked with a couple of PTC customers in manufacturing. And what they have really benefited from is the ability to take data from their CAD users in the design department, and integrate that into the data management solution or from their bill of materials, their bomb management. They're able to integrate that. And so the benefit is when we talk to these customers they said, "Well, one thing that we really liked about PTC was they had a really strong presence in the manufacturing market specifically." And that allows us to use the specific element of their solution that is relevant for our department. So we have a special department and they have a special branch of their data management that services that.

Gavin: Yeah, yeah exactly. So let's try to tag off that. If one of our listeners was encouraged to start investigating data management more seriously, how would you advise them on the process of choosing a Data Management Vendor?

Joe: So Data Management Vendors I think industry specific is really important. Making sure that you feel comfortable that the data management platform that you're choosing checks all of the boxes of all the departments that you have. So if CAD is really important part of your company, you better make sure that there are some CAD compatibility with the solution. But the other thing is, sort of, within that to make sure that you have your data organized in a way that's going to make it easy to hold together in the Data Management Solution.

What I mean by that is understanding the benefits that you're going to expect to get out of the solution before going into investing in one. So an example of that might be understanding that your sales team actually with the PLM solution, is going to have this ability that they didn't have before through the product lifecycle. So if product lifecycle management is the type of solution you're going for, or if that's an important part of your Data Management Project, you wanna understand the benefits that you're going to get you can anticipate them in the buying process.

Gavin: So Anne, I guess this, like everything then, this means that your clients and companies prioritize what matters to them when it comes to data management. Is there a short list of those that people can think about?

Anne: Sure. So I think one important thing to note is that every company has a really long wishlist of technology investments and initiatives that they wanna make, right? So how do you make sure that data management is at the top of your list make sure that that's the right move for you in the right place to put your money? So there's really five key factors that drive a positive ROI, and you wanna evaluate that potential when looking at your solutions. The first is the breadth of the solution. How many people will the new application affect? Of course, the more that interact with it the better. That's very common with Data Management Solutions in particular because as we were discussing earlier, it really affects every application throughout the entire organization.

The other is repeatability. In other words how often is the solution used? And that's the same idea as breadth. The more the more it's accessed the more you'll see benefits. The third part would be the risk of the deployment. Will this costs money if it's implemented incorrectly? And you wanna make sure that you can absorb that if necessary. The fourth is the collaboration component and that's the communication between employees which is often underestimated when looking at new solutions because it's a very expensive time suck for any company.

And one thing with the PLM solutions, for example, is that you're able to have the ubiquitous workflow there at your company. So you're much more easily able to transfer information. And that goes into the fifth factor of driving a positive ROI, which would be knowledge. So that really is what reduces the need to go through the decision making process over and over again to really come to what is often the same conclusion.

Gavin: Okay. Joe just on the knowledge one, we've often heard here about this subject is called the graying of the workforce. Okay, so there's this last knowledge that companies have today, but because of exits for the workforce or people aging especially in an engineering environment, there's a lot of knowledge that gets lost. I mean is this one of the things that you'll put in data management and I guess some obvious could address too? Is that something you've seen anywhere?

Joe: Sure. There's two, I think parts to your question. One is a lot of companies that we've spoken to when we talk to them about, they've just deployed a data management solution, they'll say things like, ''Before we deployed PTC windchill we didn't have change management.'' Change management which is how a document tracking changes to documents over time. ''We didn't have any of that before.'' So it's not that they had an inferior solution. They were upgrading. It was just a business practice that they weren't doing.

And so absolutely, having a historical record of your documents, and your data, and how the data is changing, does help with new employees becoming up to speed, older employees accessing data that they weren't a part of when it was initially generated. Things like that. The other aspect of your question, I think, is how data management enables more powerful analytics. And in terms of knowledge about the data, having the data in one place making sure it's a single source of truth, making sure it's governed correctly is important in itself. But it also allows for the data analysts to really run queries on data that they now they have a greater breadth of access to, and that is more accurate. And so they're really able to derive more insights. It's a sort of a whole another topic analytics. But analytics really does have a big impact across different departments on how the employees understand the baseline data of the business.

Gavin: Yeah I mean again from personal experience I guess it was more around it wasn't just the fact you had easy access to the data, quick access of the data, it was the fact that you could maybe go back and see why was the decision made. That was often a value that could never be uncovered before. So I hope the listeners can see a potential benefit there.

So Anne, maybe back to you. So the individual has decided data management is right, how would you recommend that they make a business case around making an investment in data management?

Anne: So if you're trying to present the business case for a data management implementation to perhaps your CFO or to whoever holds the purse strings, is you really want to have a cut through message that quantifies the benefit of the deployment. So that three numbers that we really focus on at Nucleus would be the ROI, The Return On Investment, the payback period which is the amount of time it takes for you to get your money back and the total cost of ownership or the TCO. And we approach most of these questions with ROI case studies, which illustrates specific examples of how a tech deployment can benefit a business. And you're welcome to access our website to look for more detailed-examples and use those for your business cases as you like.

Gavin: Great. I'm sure the listeners will definitely find those helpful. Joe, where do you see the future of data management going, especially given the new data sources that are appearing like Smart Connected products and Smart Connected operations? So the machines that were in the factories and the amount of data that they create and capture on a daily basis. How do you think this is gonna change, or if they're gonna look at this issue?

Joe: Well, I think as it becomes easier to implement data management solutions, smaller companies are going to start to get on the bandwagon a little bit earlier. Cloud has a lot to do with that. It's a lot easier to do small scale deployments with the cloud as opposed to on- premise. And it's easier to manage. So companies now...smaller, and smaller companies are finding that it's easier to make these deployments. And also I think that companies are going to start deploying data management solutions earlier in the company's lifecycle. Because a lot of what we've seen now like I said we have companies that have worked with Excel for eight years and now all of a sudden they've grown to a point where they can't control their data and they're retroactively trying to wrangle it in.

But I think as companies become more comfortable and understand data management a little bit better and see the need for it a little bit earlier, they're going to realize that if they can get the data under control at Step A or step B, it's gonna save a lot of time and all of the employees working with the solution which is really everybody are going to benefit for the entire period of time. They maybe in the past would have simply been thinking about data management or wouldn't have deployed it yet.

Gavin: That's all for today's episode of The Connected Engineer. Thank you very much to Anne and Joe for joining us on our podcast today, and for explaining and reinforcing the importance of data management. With the volume and complexity of data increasing, I think we can all agree that good data management is a critical piece of a business' strategy. Make sure to subscribe to our podcast on iTunes, and we look forward to welcoming you back for the next episode.



Gavin Quinlan: Hello, everyone, my name is Gavin Quinlan, and welcome to another episode of our podcast, "The Connected Engineer." This week we're gonna talk about one of the real buzz areas of the moment, the area of augmented reality, or AR, as it's more commonly referred to. We'll take a look at the trends, talk about how companies are already using this technology, and the benefits it provides. As in previous episodes, I'm very happy to be joined by some guests who can bring unique perspectives to this discussion, firstly, by Sandra Humphrey, a design automation analyst at Fujitsu, secondly, by Ryan Martin, an analyst at ABI Research, and lastly, by my colleague Matt Sheridan, a senior director for ThingWorx Studio product at PTC. Sandra, can I ask you to give a little introduction on yourself and your role at Fujitsu, please?

Sandra Humphrey: I'm Sandra Humphrey. I work at Fujitsu Network Communications, and we're located in Richardson, Texas. And I support the CAD tools for mechanical design, and I've also supported for electrical design. So our group is design automation, and what our responsibility is, is not only to support the tools but to also do process improvement within the engineering and manufacturing area.

Gavin: Great, thanks very much, Sandra. Ryan, same question to you and perhaps a little insight on what ABI Research does.

Ryan Martin: Absolutely. Again, this is Ryan Martin speaking. I'm a senior analyst at ABI Research, been in the company about two years focusing acutely on new and emerging and, more specifically, transformative technologies, came to ABI from 451 Research, and then before that, I was at Yankee Group. Throughout that tenure, I'd been covering what is now called IoT, but at the time, it was called M2M, and throughout that tenure I had been charged with looking at some of these more emerging technologies.

Gavin: Great, thanks. Finally, Matt, can you explain your role at PTC and what your responsibilities are?

Matt Sheridan: Sure. Thanks, Gavin. So I've been with PTC for many years. Right now, I'm part of go-to market for ThingWorx Studio. I'm understanding what our customers do and the problems they face and how augmented reality can help with that. I'm working on packaging and positioning ThingWorx Studio and doing fun things like this podcast and working with great customers like Sandra and analysts like Ryan.

Gavin: So thanks for those introductions. Ryan, can I ask you to give our audience an overview of what augmented reality is? As I believe a number of people have some misperceptions about what it actually means. Like, there's a distinct different between AR and VR, correct?

Ryan: Absolutely. So, augmented reality is the ability affix content information in your field of view in a heads-up display. Typically, it's a see-through display, often in the form of glasses, but it could be also on a dashboard, it could be in a vehicle, on a widow, on a machine itself. Virtual reality, by contrast, is much more immersive, usually with a binocular display or form factor where you cannot actually see the environment around you.

Gavin: So, are there any trends that are starting to appear when it comes to AR?

Ryan: So I'm of the opinion that augmented reality is one of the technologies that's gonna fundamentally change the way companies do business, the way that we access information, the way that we communicate, and also the cadence at which we're able to do all those things. The reason being that augmented reality provides the opportunity to extend what you would normally experience on a desktop or within the four corners, confines, of a traditional display and bring it into the real world.

Gavin: Okay. So, Matt, what's PTC's perspective on AR, and do we have a particular area of focus?

Matt: So I agree with Ryan's comments. First of all, augmented reality is gonna allow us to bring a digital representation into the physical world. So for PTC, we're very focused again on the industrial enterprise, and we have tons of customers today who build many amazing products. Taking the information that they have existing, whether that be in business systems or with geometry, and being able to bring that and put it in context of the physical product when someone is standing in front of it or using it, that's gonna really shift on how people think about products and work with products in the future.

Gavin: So when you say geometry, you really mean CAD data?

Matt: That's correct.

Gavin: Okay. So given your daily interactions then with customers, Matt, who are working with our AR solution, are you seeing any particular areas that companies are using it for or are starting to think about using it for?

Matt: Yes. So we actually did a survey to some of the customers today who are working with our trial program for ThingWorx Studio, and we definitely see some areas that people are concentrating on. In particular, if you take a look at training and trying to bring knowledge from existing workers or the aging workforce that is leaving to newer workers, in many cases, there's still huge manuals that people are looking at, and they see augmented reality as a way of bringing that knowledge and putting it in a very visual, easy-to-consume context. And again, in context with the physical product makes it even that much more powerful.

Gavin: So, Sandra, if I can come to you, how did Fujitsu get started with augmented reality?

Sandra: Well, I've attended the LabWorx conferences and seen the technology there, and never really thought that we would have the expertise to do something like that with our current staff. And so, as I was looking at some quality reports, I noticed that there was a customer outage that was not detected. And then I started thinking, could ThingWorx help us find these issues quicker or that they would at least be detected? So that's where it really started.

Gavin: Great. So I think it would really help our audience to hear a couple of examples of use cases where AR has already been employed in what Matt referred to as the industrial enterprise. So I know there's an ooh and ah factor that AR has, but I'm working on the basis that real business value can be developed. So, Sandra, would you mind sharing a use case that Fujitsu has worked or is working on?

Sandra: Yes. So we have a couple right now. There's one in our lab to provide sensory [SP] information to other Fujitsu locations in the lab environment. We also are looking at tech hubs in training, being able to use our models directly in their publications instead of having to take an export and redrawing and recreating the graphics and therefore saving, you know, a very substantial amount of money by reusing the model. And then, of course, there's sales and marketing, the ooh and aah factor, but it's not just providing a graphic. You can now provide customers with product that hasn't even been manufactured yet, and give them detail on that product.

Gavin: So they can basically see the product in situ without the product being there by using the using, you know, the glasses or an iPad, like Ryan referred to earlier on.

Sandra: Exactly. I mean, it's not having to send the unit to Canada when you have a customer visit. You can bring your iPad or HoloLens or whatever your choice of communicating is, and it's to scale. You can look inside the unit, walk around the unit, and give them, basically, all the information that they need at that point.

Gavin: Yeah, that sounds amazingly cool. So do you mind if I ask a cheeky little question? How long did it take to create that AR experience?

Sandra: Well, you know, we started the pilot project, and I had a couple within probably four to five days.

Gavin: Wow, that seems incredibly quick.

Sandra: Yes.

Gavin: So, Matt, I'm gonna come back to you. Have you got maybe any other customer use cases that you'd be able to share with us?

Matt: Sure. First, if you don't mind, I'd like to touch upon a couple of things that Sandra talked about and why we're seeing a lot of enthusiasm and excitement with ThingWorx Studio. Just coming back to that last point in terms of the speed with which you can create augmented reality experiences versus other past methods, ThingWorx Studio is allowing us to do that. But I think another important point there is the ability reuse some of the information, especially for many of our PTC customers out there. It pools of CAD data that they have built over the years as part of the design process, and the value of that CAD data maybe is lost now that they've created the drawings and then produced and manufactured the product.

But with augmented reality, we're unleashing another opportunity to reuse that CAD geometry, bring it now again as part of the experience that shows up in this physical, digital world. And as Sandra pointed out, if you have that definition already, then you can immediately start to plug that right into ThingWorx Studio and use it. So those are some great points that you brought up there and, again, kind of validate some of the thought process.

Other examples, though, that you're talking about, again, we do have a very popular program. We have hundreds of customers and participants. Just to kind of touch upon one, Solar Turbines, it's been with us from the beginning of the pilot program. They were also at LabWorx last year and demonstrated what they did with their augmented reality experience. They were looking at it from a manufacturing use case. So how could somebody take the initial design and then map it against the final product? Again, that digital information against the physical, and then see exactly, you know, was this produced to ensure high quality?

Gavin: Okay, so we've got a sort of sales and marketing use case. We've got a manufacturing/quality type use case. What about the designer themselves? Where are we thinking the designer can interact with the product?

Matt: Sure. So, you know, number one, again, the designer is probably working on the CAD geometry, the designs themselves. And when you start doing that, you're talking about design reviews. You're also talking about prototyping and the iterations that you normally go through in a design process. If I could take my geometry that I defined, the CAD model, and then bring it directly to the product that I'm working on or to other designers and say, "Let's review this together," and now I can do it in a virtual environment, sort of like the sales discussion Sandra was talking about, I can see the size, I can walk around it, I can get an understanding. And if you already have some of the product existing, you could put a digital component in the context of the existing physical product so you could see how that might look in terms of a prototype, go back, iterate, do it again very quickly. So it's a lot of thought process there that people are gonna be doing then.

Gavin: So you mean it'll basically cover the entire product development process. AR can be used from front to back end, basically.

Matt: Yeah, yeah, I think so. And even just bringing things to life, like if you had a drawing and you could see now instead of just the 2D version of the drawing, what if I could actually see as the person on the shop floor virtually what this thing would look like in three-dimensional space, that's just gonna open up a whole new understanding.

Gavin: So, Ryan, from a research perspective and the clients that you're interacting with, have you come across any interesting use cases of AR out there that you'd be able to share with us?

Ryan: There are a lot of them, and a lot of really good points raised actually to this point as well, but these include things like track listing and documentation. So the example Matt just gave, that could be extended to also, say, recording a repair or fix or design as it's happening and relaying that to the client or saving it internally, tapping into it later. This includes things like fulfillment, communication collaboration. It could be in warehousing, field of service, manufacturing, or otherwise, certainly for identification, inspection, quality insurance, and telepresence, just as a couple more examples.

But what I think is really unique about augmented reality is the way that this information is conveyed, and it could be certainly something in your hand, like a tablet. But I think ultimately it's gonna be in this heads-up display, digital eyewear type form factor. And the significance of that is that a lot of folks are already wearing glasses. The safety glasses in manufacturing is having the same way the nurse wears scrubs or a firefighter wears firefighting equipment. These are occupational-related tools, for all intents and purposes, and there's no reason, in my opinion, why this technology should not have a place to play in there if it truly transforms the way that employees have access to information and can, you know, perform their jobs better, which is exactly what we're seeing already.

Gavin: So, yeah, this was a point I wanted to make. You said you've been involved with IoT or M2M there for about a decade, and I guess I'm thinking of something like, you know, "Minority Report" or something, you know, that, again, the audience might be able to just recollect back to where it's not just about the visual itself, it's the fact that information of the product can be overlaid as well as the visual. I guess that's where the real power of this stuff starts to come together, and I think Matt used the phrase, you know, that bridging of the physical and digital. I mean, is that what you're seeing with clients you're interacting with?

Ryan: Absolutely, and that's where I think is something like perishable insights. There are information that is being generated but not necessarily transmitter-captured at the point of inception. So what that means, when we think about augmented reality in that context, is that you can not only get more value from operations as they're occurring, but layer that into, say, the CAD data that Matt talked about before to accelerate what I think is the new KPI to wrap around all of this. It's not just ROI. It's about time to value.

So the ability to get a solution like this to market in a matter of days, when we're thinking about something like HoloLens where development is, for all intents and purposes, entirely new, can cost companies, you know, six figures plus, which compared to enterprise mobile app development has a bit of a sticker shock. But when you think of that in the context of time to market, bringing it down to a matter of days and being able to actually provide new tools to employees that immediately get buy-ins, it's a pretty easy sell, especially when we're talking to companies that are, you know, fighting the margins, as is often the case in, say, manufacturing or even things like insurance.

Gavin: Yeah, on a previous podcast, there was a mention about, you know, the fact that the millennial is the new employee, you know, and that that type of individual is not interested in the 300-page manual to learn how to do something. If they can't see it almost on demand, it's like, "Yeah, that's okay. I'll find something else to do." So I guess that's something similar. Matt, I think you had a point to make.

Matt: Yeah, I wanted to just touch back on what Ryan said there about sort of the heads-up display. I think that's where people today, if you ask them about what they're thinking about with augmented reality, they kind of picture that heads-up display. And, you know, we're definitely, no question, moving in that direction, which will have some interesting implications, because do you need to, for example, design a panel into the product to actually show you information if you're gonna have it on a heads-up display of your glasses? So that will be some interesting things to think about as we go into the future just around product design.

But the display itself, number one, a heads-up display is actually interesting because it gives the wearer the ability to absorb more information, so we'll be able to make better decisions by having that. But the heads-up display is only as valuable as what you're seeing. So do I have interesting information, or do I have geometric CAD information that aligns my product? That's one part of being interesting. But another part is, do I actually have the right, correct readings and information coming from the product itself? And that's the IoT side. So if you have the IoT information, that's fantastic, but it might be difficult to display. If you have the display information without the IoT information, you might have an interesting display but not really contextual information that can help you make decisions. But if you can put the two together, then you're on to something.

Gavin: So it feels like we're right at the cusp of something very exciting when it comes to employing augmented reality and industrial enterprise. I'd like to ask each of you to take a step into the future and give me your perspective on where you see AR going for the enterprise. And, Sandra, I'll totally understand again. You know, I know you probably have some secret projects going on at Fujitsu and you don't wanna give the game away, but maybe we could start with you in terms of, you know, what other areas just at a high level perhaps Fujitsu are considering using AR in.

Sandra: The CAD data and putting the visual innovation around it in manufacturing, I believe it makes the worker have to be maybe not as skilled because you can provide more detail, and overlaying it on the product to where it's better than a document that they're actually seeing it. So I think, for us, it's going to be manufacturing is where we're going to be headed, as well as sales and marketing. So far as how sales going to not just provide a cool representation of a design by the unit, but how can they now turn that into a configurable sales tool that they can configure product on the fly and give examples of how it might look in a full rack visually.

Gavin: Yes, and not just the static version but the ability for the, you know, salesperson to actually, like you said, configure it on the fly. That would be pretty incredible, all right.

Sandra: Right, right. Because, you know, some of that stuff's done in video and in other things where they have to go back and, you know, they can't give the customer critical data information so that they can make a decision quickly about that configuration, so you could do that in a more timely fashion.

Gavin: So just for the sake of the audience here again who may not understand the type of products that you're speaking about, what sort of size would a rack or a Fujitsu type of product that would be presented to a customer? How much space would that typically, you know, take up?

Sandra: Well, one rack is 24 inches by 6 or 7 feet tall. So showing a customer that at their site, a fully configured rack that might have anywhere from four to six units in it, would be very difficult, right? So if you can provide a visual, like through HoloLens, of here's what your rack is going to look like, here's where the cabling is going to come out, and, you know, where you're going to need power, and, you know, you could provide several racks actually of what they would be purchasing. So it really does become limitless at that point to what you can provide the potential costumer or an existing costumer about what you're providing.

Gavin: It must provide a very unique experience for Fujitsu, you know, for their customers, you know, to be able to deliver something like that.

Sandra: Yes, yes. And, you know, we've done some other things as well. You know, you talked about the millennials and that sort of thing, and it is attractive to kids coming right out of college who have a very different experience growing up in the time that they did. And a company that has technology such as this is very appealing to those people. And, you know, providing the colleges around here and area of high schools and STEM programs, we're taking these HoloLens and that sort of thing over there so they can see what it's like. They see it on TV, and you know, they're like, "Okay, you know, that's cool." But a kid, you can give this to, and it is pretty exciting, just the whole...

Gavin: Yeah, it just sets a whole different tone of voice for what engineering means today compared to, you know, drawers and drawers of drawings, that sort of picture. You know, that's pretty interesting. Matt, what's your future step? Where do you see it going?

Matt: Well, I think I go back to a little bit of what I said earlier. There's going to be a shift in how we interact with products, and that's where we're gonna be in the future. And I guess sort of the analogy is, you know, today you pick out your phone out of your pocket to do your banking. But 15, 20 years ago, you know, that was not the case. And I think with AR, we're right at the beginning of what's gonna be an incredible ride as we bring in these experiences that connect with the product, and they're just gonna start to allow us to interact, and that could be AR, touching, visual, voice command, haptic. There's all sorts of parts of augmented reality that can be brought into bare when you're dealing with products, so it's gonna be exciting.

Gavin: So, Ryan, let's close that section with you as maybe the person who gets to see the most or deals with customers who are on the bleeding edge of things. What's the analysis firm seeing?

Ryan: So I think that it's tough to dispute the benefits of hands-free and heads-up access to information. And there are tons of use cases that we've discussed where there's certainly relevance today, like, very real, tangible examples of how AR can benefit an organization, department, a firm, or individual. The challenge often is sort of articulating the business case for it, so whether it's developing those ROI scenarios or getting through proofs of concept, getting stuff out of the labs. And in doing so, what we're finding is that the decision maker often is not necessarily the traditional decision maker. It's not gonna be the IT department, although certainly they're gonna have a hand in managing and perhaps in procuring these devices and solutions, but often case is it could be the project manager, it could be the product owner, and it goes all the way down to the people on the factory floor. Like, these are the folks who are actually using the devices, and those are the people who also need to be reassured that this is a technology that's not there to replace what they are currently doing. It's there to help people do their jobs better, and it truly will make a difference.

The question, I think, in the back of a lot of people's minds is, will there be an iPhone moment for augmented reality? I'm not sure, but I think that it's clear, looking at it holistically, things are certainly trending in that direction. Looking a couple years out, I think we're gonna look back and say, "How do we not have this before? How do we do business, you know, before we had augmented reality? What was life like before Post-it notes?" kind of thing.

So I think there's a lot to look forward to and also there's a lot in motion to make this happen, make it possible, make it a reality beyond just the hardware. Because as we talked about before, this isn't just about augmented reality. It's about IoT. It's about greater visibility down to the things that are actually getting connected.

Gavin: Yup, [inaudible 00:23:22] I agree, and I definitely think that the example Sandra gave us is a good one when it comes to how does technology get into the business, which is there's a problem that you identified and then now you're finding a way to solve that particular problem. And as augmented reality continues to mature, it's gonna be able to solve more and more and more problems that the businesses are facing, and give them more opportunity for selling service, operating, whatever the use case might be. But it's still gonna be rooted in improving the business and helping the business, and I guess as people look around what they're doing today, and they haven't solved some of those problems, this is gonna give them the opportunity.

Well, it's another technology to at least investigate to try and deal with those problems. I mean, there's no guarantee, but it's a new, you know, extra accessible piece of technology that perhaps even a couple of years ago you couldn't have considered.

Matt: I agree.

Gavin: So, Matt, you know, listening to this webcast, and if I wanted to learn a little bit more about AR and perhaps even create an AR experience, is there a way that I could get started?

Matt: Yes. So I mentioned a few times for ThingWorx Studio, which is the tool that we offer today, and Sandra mentioned using during our pilot program, we now have online...the ability to try the software for 90 days. You can go to studio.thingworx.com and register, and we'll be happy to sign you up. And you can go through the tutorials and the exercises that we have online to help you get started, and today in the product, we actually have included some initial starter projects. So they're right there when you open the product, and you can immediately start to create something within the first 10 minutes.

Gavin: And from a personal perspective, I mean, is there, you know, an app or something that people could, again, I don't wanna disrespect, but, like, have a play with, you know, in terms of the AR side of things?

Matt: Sure, no, absolutely. So today on the Apple Store, Google Play, and Windows, you can go and get our ThingWorx View app, and the app itself includes some gallery examples. And if you go into the app, there's a quick little instruction on how to get going on that, and you could see some of the examples that we're talking about. And they represent some of the areas we covered. So it shows CAD data and geometry. It shows geometry in some sort of sequencing, which would help, like, the work instructions and training we talked about. It shows some representation of IoT data, gauges, and information coming off of the products, which, again, is all part of that decision-making in getting towards that heads-up information that people are looking for.

Gavin: So that's all for this episode of "The Connected Engineer." I'd like to give a huge thank you to Sandra, Ryan, and Matt for joining us on the podcast today. Even a year ago, the thought of creating an augmented reality experience would probably have conjured images of Hollywood budgets and teams of people, but I trust after our conversation today that you realize this is no longer science fiction but science fact. I encourage you to investigate ThingWorx Studio and experience AR for yourself. Thanks for listening, and don't forget to subscribe to "The Connected Engineer" podcast on iTunes. Thanks a lot.


Episodio 3: Eyes off the Road: The Driving Force behind Autonomous Vehicles


Gavin Quinlan: Welcome to The Connected Engineer, a podcast for the engineers, designers, and innovators. I am your host Gavin Quinlan, and each week I'll invite experts to discuss the product development challenges we're facing today, as well as the trends of tomorrow.

I know we've all seen self-driving cars in the movies, and you can't open the Wall Street Journal without seeing an article about the future of the automotive industry. So today, I've invited David Glass, who is the founder of LHP Engineering Solutions, to give us an inside look into the area of autonomous vehicles, and if you'll excuse the terrible pun, the driving factors that have led him to tackle this area of engineering. Hi Dave and welcome to the podcast. Can you give the audience and I a little on your background, and your education, previous companies you've worked at? It will help the audience get a better appreciation for you and your expertise.

David Glass: Sure. I have a Electrical Engineering Degree and a Business degree from Trine University in Northern Indiana. And I started out my career working in Cummins Engine Company on diesel engine controls. I spent about eight years there and basically moved into a position where I was re-architecting the software platform that Cummins was using across their engine platforms, and helping them to get their product line strategy, leveraging model based design, all the co-generation and really kind of move into leading edge software processes for. So that they could ultimately increase their business size by about four times over the next 15years.

Gavin: Some very deep engineering experience there so, now to the present day. Tell us about your company, LHP Engineering Solutions, and what solutions you provide?

David: We have traditionally been an engineering services company. More recently, we have moved to a model that involves integrating technologies from significant technology partners like National Instruments, and PTC, Microsoft and others. Really, our core competencies are around embedded technologies, your micro controls, and control solutions, along with IoT Connectivity, Telematics, and data and analytics big data. I update solutions for monitoring vehicles, and machine learning, and predictive analytics.

Gavin: So a very broad set of capabilities focused on the sort of automotive area, I guess would be fair. Can you tell me how the company got started?

David: I made a decision to leave Cummins and try to do something different, and actually, Cummins asked that I continue to work with them and eventually I started to grow a services that was supporting Cummins within. But over time we branched out across the automotive industry in a new regions including the Detroit market, Southern California, Colorado, as well as China. And we've just kind of grown over time and started to invest in actual technology solutions that we see the automotive industry really needing to take advantage of all the new technologies that are coming towards the automotive industry now with autonomous vehicles and the connected car, and some of the other challenges they are facing now.

Gavin: How many people work at LHPES?

David: We have 400 now and we have about, close to 300 openings, so we're...we have 10 full-time recruiters just continually recruiting. We also have us a internal university that really pursues talent from the universities and the idea is to get them through a rapid boot camp type scenario, six weeks boot camp, that gets them to a productive state as fast as possible. So we train on certain applications in high need throughout the industry.

Gavin: I understand that about two years ago, you decided to refocus the company more on the area of autonomous vehicles. Do you mind asks me what the catalyst was for that?

David: Was going through my engineering degree, I was also collegian athlete in baseball and basketball, and then I have four kids, three boys that are in sports as well, and coached them throughout the years. And then one of the team mates his name was Josh Bidel [SP], started coaching him when he was about eight years old, and very athletic individual. Grew to about 6'8", and just incredible athleticism and eventually became one of the top basketball players in the state of Indiana. Was recruited by Vermont, who actually just made the NCAA tournament. But about two years ago, on Super Bowl Sunday, he was broadsided by an SUV and he was in a coma for six weeks. And so, the following weekend I was with his parents at the hospital and I was just sitting there thinking now the technology already exists, you know, to avoid an incident like that. And so I guess at that moment in time I just really said, you know, we're the right company, that we can do something. I feel like I had a responsibility to the family and just...even my kids and grandkids had said that, you know, there is a real need to make our roads safer and leverage the technology that already exists today to help save lives and remove much of those incidents.

Gavin: That's an incredibly impressive story. What were the people and your team's reaction when you told them you were going to do this refocus?

David: We were also in a rebranding process so, actually, the announcement publically was made at the rebranding event probably six months later. And it was definitely well received and I think everybody was onboard with the idea, including the community. So, we got really good backing for that idea. The incident reflected the city quite a bit in and across the state of Indiana as well so it was good reception.

Gavin: So there was no doubt among the team that this autonomous vehicle area was going to be great area of focus for the company, a great way to develop the business.

David: No, it really brings everybody together with a mission and vision. You know, I and so I think the ability to be able to leverage the talent we have to do something big for the world can be very exciting for everybody. And not only that but, you know, everybody sees the opportunity and excitement with the technologies being implemented in the automotive space. And so, it's easy for everybody to get behind that, you know, because that's what we're all about anyway.

Gavin: So let's get into a little bit more detail about the engineering of autonomous vehicle. What are the particular engineering challenges that have to be address by your team in the development?

David: Well, there is two big ones that come right to mind, and one is just capacity overall of engineering capability across the industry. It's been fairly fixed over the years. I means there is only so many people that come out of the universities, you know, there's a lot of foreign nationals, but that's also fixed, and then there are visas and so, you know, the number coming in versus the number retire...pretty much the tool of engineering capability and bandwidth at the companies have is just pretty much fixed at the same time. The complexity is growing exponentially and additionally to that, we've also have new forces that are being driven towards the industry. And part of it is the investment autonomous vehicles, there's also the standard that's gonna be required called Functional Safety. And then you also have a number of additional players coming in the space. Technology, big technology players like Apple, Google, Tesla, and Uber. So, they're all buying and competing for the talent, and it's just stretching the capacity even more.

Gavin: So it's extremely competitive to recruit talent given like you said, all the new players in the market, and I guess that goes back to your previous comment of having 300 open positions at LHPES today. What are you doing to, I guess, you know, create that little magnet from your side. You mentioned the university type approach, and the sort of inner university you've created. Can you give us some more details on that?

David: Sure. We've partnered with SAE, and we got started in a model that we've been able to leverage that is really hands-on training and it's not really specifically geared towards making everybody an expert in all aspects of the technology. It's more about getting them to a productive state, confident, a feeling of accomplishment so they can hit the ground running immediately, and come into these opportunities and make a difference right off the back, and then grow their career much more quickly than folks who have traditionally come out of universities. So that's kind of the mission of the group, but it's been very effective and a lot of it is word of mouth, you know, by a person will come in, sign up, get a job and they'll go tell five with their friends, and then they sign up. So, it's kind of been a self building kind of system and it's grown to roughly 160 people, graduates per year, you know, they either fill positions with LHP or fill other positions throughout the industry. So, but on a larger scale what we've realizes that the real solution for the industry, it's not gonna be people that fix it. It's gonna have to be improve processes, improve efficiencies within engineering organizations and technology solutions with tools and leveraging some of the things that have evolved over the past decade like Cloud-Based Systems, and Cloud-Based Testing, and automated testing, and things like that. That we feel will need to be the solution versus just throwing more bodies at each aspect of it.

Gavin: Because there is sort of an upper limit to the amount of people that can be recruited, I guess, given the number of players and the number of graduates, there's just a simple limit.

David: There's that and the time it takes to grow the expertise versus the speed at which the industry is growing and advancing.

Gavin: So, let me just take a slightly different tack. Your facility in the Indiana, I understand there's a pretty cool story about how that came into being. Are you okay to share that with us?

David: The fact is we don't build cars, you know, we don't...we're not automotive manufacture. We also don't build some of the technology like the cameras, or the LiDAR system, or radar systems that actually will go in the vehicles and enable autonomous vehicle. We are the engineering of control software and those kinds of things. So, we start to focus on the standard size of it and when you take it one step further then there's also got to be someplace to do the validation, to meet those standards. So, we started thinking about how can we get a facility in place that we could actually execute those tests. And a few guys came to me, these real estate investors, and they were suggesting they'd like to put a performance track in place and asked if we would be interested in being involved. And I said, "Well, not really, unless, you are willing to invest in developing testing capabilities for autonomous vehicle applications", and at first they were like, "What? You know that sounds like space age", you know, but this was two or three years ago. And but the more they got in it, the more I talked to them and of course, you know, over last two years things have evolved so quickly that it's clear to everybody that this is probably the biggest opportunity that exists for them. So, a lot of OEMs in the area have gotten on-board, a number of technology companies have gotten on-board, a number of universities have gotten on board, as well as the state of Indiana. So, we got a lot of backing, a lot of interest, and it looks like it's gonna move forward. So, we're thinking something like a 300 acre of park that would be fully focused on autonomous vehicle research.

Gavin: Speaking about the testing, bringing me back to that functional safety item that you mentioned. Who's setting these, and, given the speed of innovation in this area, are the certification bodies struggling to keep up?

David: Well the first round of Functional Safety Standards as I am aware of came out in 2011, and since then they've been continuing to work and evolve the standards, as autonomous vehicles become more and more likely in reality, and the next set of standards are supposed to come out in 2018. So, there is…I think there's a National Highways Association, and a Transportation Association that's driving a certain amount of it, and then there's the committee body that is setting a standard that is called ISO 26262. And it's related to the aerospace industry standards that have been in place for the last 15 to 20 years and support the aerospace industry's manufacturing of commercial airline applications.

Gavin: So they are able to reuse some of their standards that have been built by aerospace in the automotive sector.

David: Certainly, yeah. Just with the growing complexity and the growing safety issues that exist with large vehicles on our highways in very congested areas. In fact, I would say the challenges with making the standards for automotive is probably exceeds aerospace from standpoint that in aerospace you're generally going at a certain altitude from point A to point B with not a real high likelihood that you have any, you know, interference. Whereas on the highways with the amount of other obstructions or variability in the driving conditions and everybody being on the same plane that actually makes the challenges much greater.

Gavin: It could become almost like a super air traffic control system that would need to be in place in order for it to handle all the autonomous vehicles that will be on the road, I guess, one way of looking at it.

David: Well, it'll likely be a system of connected vehicles and infrastructure to help them get it. So it will be a, probably, be some sort of a cloud based system that helps sort to navigate the entire traffic and highway inter…you know the system.

Gavin: Yeah, so a system of systems I think is one of the phrases we've used here. So, I think that's a really tangible example of it. So let's take a look at into the future. When I think of autonomous vehicles, I guess I was thinking of driverless cars, but speaking to you today I'm starting to appreciate that all of this technology is gonna start to appear in the cars that we drive every day. It's like the next step in ABS or traction control, is that fair?

David: But yeah, it's been a, you know, so the idea of the ADAS, Advanced Driver Assistance Systems, has been growing already with ADAS. And a lot of cases we maybe and realizing how much logic, and controls, and algorithms have been making decisions, while you're driving, even today. So, I mean, that's gonna continue to grow and I find it interesting too, a lot of people I've heard from say, you know, "Well, I just don't know if I can let go of the wheel, you know, let the car take over", but the reality is t, you know, once you get in and I have, you know, with the Teslas, you know, it takes you about 10 minutes to say, "Wow, this is pretty awesome." And you relax pretty quickly and say, "Yeah, it seems to know what it is doing." So yeah, I think it would be a lot easier transition than people realize and then also, they'll quickly appreciate the advantages it gets them, as far as, you know, to be free to focus on other things or maybe even take a nap, you know.

Gavin: And so, now they can actually text without getting in trouble, I guess that would probably the main thing they want to do. I'm guessing this could have some pretty significant economic impact, like, can you help us understand the areas that might be affected because there's a multitude popping into my head?

David: Yeah, there really is a lot of areas so, I mean just the amount of savings that can be done for the government, you know, related to just the damage, the liability issues, just the amount of money spent on monitoring traffic on the interstate. So, there's many ways that states could save money, the government could save money. And then there is the productivity factor, the amount of work that could be done by people who are today pretty much tied to driving for up to four hours a day going, you know, two hours to work, two hours back, maybe in L.A. or those areas. You know, so productivity is a factor. Also the people who are maybe not mobile today who could be working longer in their career, things like that. So there's many things around productivity, and there's gonna be so much disruption in the industry as well with ride sharing, and also insurance is going to be redesigned for the new way of the world. And so, yeah there's many, many things will come into life as far as changes in the economy I would say, and potential significant economic improvements.

Gavin: Well, I for one won't be crying when the insurance companies have to drop their prices, I can tell you. For the million dollar question, do you see a day like when people driving will become a thing of the past?

David: I think it's possible that young people may not even need to learn to drive, at some point. I think there will be certain cars that are delivered without a steering wheel, but I think the ability to drive will continue to exist for quite a while. But they'll be a lot of technology behind it that will be maybe assessing the driver's capability, whether or not they're intoxicated or distracted. So, you know, I think it's gonna be a blending of…and a kind of an evolution of progress that may eventually get to a point, maybe decades down the road, where maybe you're not even allowed to drive anymore, and so, and this is gonna be interesting to see how it progresses.

Gavin: Yeah, it sure is. So Dave, what are the next steps for LHP Engineering Solutions? What are the next projects you're focused on there for the next couple of years?

David: Well several areas. One is the training systems that we're putting in place to enable more people to get in and contribute so we have that element, you know, advancing the collegian competition that's one piece. And then also, I mean it's really about again integrating the technologies to address the major engineering constraints that exist in the industry today. And so, driving the technology solutions, and automating certain turn the crank engineering activities that are involved with documenting, and certifying, and validating the systems so that they can be approved for commercialization. So I think that's a big part of it. So it will be really focused on investing in the technology systems, the hardware, develop of the platforms, the automated test systems, and the other software systems like test as a service, and cloud-based systems that can really help optimize engineering processes and better utilize engineering assets that are available to companies.

Gavin: So I mean for an engineer, who might be listening to our podcast today, where would they go to find some more information or maybe to even, you know, fire in an application about the 300 open positions?

David: Well, they can come direct to our website and read a lot about some of the things we're doing, and then also a lot of our jobs are posted directly on our website.

Gavin: Great. So, in closing Dave, I'm going back to the catalyst for the company's change of focus that we spoke about earlier. Do you mind me asking what happened with the boy who was in the accident?

David: Yeah, it was certainly a change for the family, change for him lifestyle wise, but the great thing is it turned out to be, I guess you could almost say a miracle. You know, in his recovery. And he…two years later now, he still went to the university where he was recruited in Vermont. In fact, if you watch the game, that where gonna be playing Purdue this week, won't be in uniform, but he'll be with the team. They have him on a ongoing workout routine, and he practices with the team, and obviously, he is not at the level, as he once was or at the D1 level, but the commitment the team has made to him and the family has just been absolutely incredible. So, I'm definitely gonna be rooting for Vermont so, even though, I like Purdue as well but.

Gavin: Well, I think most our listeners will be rooting for them too. Dave, thank you very much for that uplifting story. That's all for another episode of The Connected Engineer. A huge thank you to Dave Glass for joining us. And if you walk away with anything from today's episode, I hope it's not just a greater knowledge of autonomous vehicles, but the importance of needing a combination of great innovation, great technology, and great people to make real change happen. The importance of making a difference and how you can do it with engineering. Make sure to subscribe to our podcast on iTunes, and we look forward to welcome you back for our next episode, where we'll speak Fujitsu about augmented reality, and dispel some of the hype around that subject.


Episodio 2: AR/VR and IoT take flight with Aerospace & Defense


Gavin Quinlan: Welcome to the Connected Engineer, a podcast for engineers, designers, and innovators. I'm your host, Gavin Quinlan, and each week I'll invite experts to discuss the product design challenges we're facing today as well as the trends of tomorrow.

On this episode we will be talking about the Federal Aerospace and Defense industry or what is more commonly known as FA&D. We will discuss how emerging technologies such as the Internet of Things and augmented reality will impact companies who occupy the space. Many companies see CIoT as something that is relevant to consumer products. But not many people realize how important this technology is to aerospace and defense companies. Today, my guests are two colleagues, Ian Bolton and retired United States Air Force two-star general, Brent Baker. Ian is as a technical sales director at PTC and General Baker runs our global federal aerospace and defense business unit. Let's start with the basics. Ian, can you give us a bit on your background and along you've worked with PTC? And what the role of a technical sales director covers.

Ian Bolton: So I've spent pretty much my entire career in aerospace and defense over 27 years. I joined the Royal Air Force in the UK at 17 years. And I was air craft maintainer in the RAF. Did seven years in the Air Force then left and joined an OEM which was BA systems and worked there as a subject matter expert before joining PTC. So my entry into the software industry was with PTC, and I've been with PTC for 10 years.

Gavin: Okay. Thanks Ian. That's an incredibly distinguished and long career there. General Baker, sir, how did you come to work for PTC? And could you give us a little bit of your history with the United States Air Force.

Brent Baker: Sure. Again, I probably served United States Air Force for 37 years. I spent six years as an enlisted member and 31 years as an officer. And I did everything from fuels management, to operational planning, to maintenance, supply chain, big heavy logistics. I had a wonderful opportunity to join PTC, a little over a year ago, I've been with PTC about a year and two months now, and it's just been fascinating for me because one thing my last job, I was the vice commander of Air Force Material Command which has cradle to grave responsibility for all assets and weapon systems in the Air Force. And I thought I knew quite a bit about software , but after I started working with PTC, I really quickly discovered that there are so much amazing software that is available, and in this case a lot of cases out of the box. And so, that's how I really ended up with PTC. It was just a fascinating area for me. I'm just really proud to be part of the team.

Gavin: Thank you so much, General Baker. Hopefully it's obvious to our audience today that we really have two very deep knowledge experts on the area that we're gonna discuss. So, let's get stacked into that, straight away. General Baker, can you explain to our audience why you believe the IoT is so important to FA&D customers?

Brent: There's a couple primary reasons, you know, first of all there's a lot of information that is available when you start reading about IoT. It really is kind of the next iteration of where technology is heading. And actually it's there, quite frankly. I've heard some folks refer to it as a fourth industrial revolution. Another really important opportunity is the United States Department of Finance has a strategy called The Third All Set Strategy where they're trying to take a leap in technology. And I personally believe IoT is and should be a centerpiece of the third all set strategy, but there's IoT and I think the folks will listen that here today that it's such an amazing opportunity. And quite frankly the ideas and options and potential is almost limitless.

Gavin: Okay. Ian, just to try and follow up on that question, could you maybe pick out three or four used cases that you would see that, you know, a company should be looking at almost today or are already looking at today?

Ian: Yeah, sure. So, the air in the industry in some respects has kind of led the industry in small connected products. There are aero space engine manufacturers that have been taking sensor readings and collecting data from complex assets such as air craft engines for over a decade now. And, you know, what's changing in the industry is that the software and the technology that enables you to do this is becoming much more advanced and much more commonplace throughout the enterprise. So some typical applications of that technology would be remote sensing, condition monitoring all looking at the data and performing some sorts of diagnostics or even predictive analytics around the way that the asset is performing and how that could be optimized.

Gavin: General Baker, from a defense perspective are there any of those used cases that are of more relevance or of more interested to companies that, you know, you've worked with the United States Air Force, just for example?

Brent: There are, there a lot of really some pretty amazing used cases and it seems like in the last year or so that we're starting here more and more of the smart connected products and assets coming to fruition quite frankly in the aerospace and defense in particular. You know, a couple that I would think about, this one is health assessments.

We have an organization right now that is doing some pretty amazing health assessment on a particular part of a missile or rocket and they were trying to do this for many years using their software that they created. They had a group of technicians and computer scientists working on that and they were able to basically buy the whole thing works out the box solution, and they were able to complete this project in weeks. I'm literally mean like four weeks they were completely done and then moved on to something else. So, there's also a lot of examples where you could do like logistics operation centers where you monitor aircraft movement around the globe, you could know when an aircraft is landing, you could also know like what work orders are required, or that that aircraft requires, is this the right center for this aircraft to go to? Are the parts available? All that is actually being done today and used in smart connected products. So those are just a couple of examples.

Gavin: And Ian, obviously General Baker has got a very deep expertise like we said in the defense side of things, are those used cases that you've both discussed, are they as relevant for, if you wanna call it the airplane that we all know and love and use every?

Ian: So, I think some of them are and then in some respects commercial aerospace or mass transportation as its own used cases too. So as an example, companies are looking at how to efficiently move passengers, how to efficiently move cargo and luggage. And the IoT can help with sensing the systems that are responsible for that logistics activity and then telling you how well they're performing if you have a breakdown in that process and also in providing metrics in terms of how much cargo or how many passengers have you moved on a certain day. And all this, you know, helps operate just kind of tune that business to perform at the most optimal level.

Brent: You know, I wanted to mention something that Ian said earlier, in the commercial world and even in defense for many years I've been collecting data and information from sensors and or engines, and, but IoT, I think this is really...this technology coming into its own because now you can actually take all of this data which quite frankly a lot of folks didn't even really know what to do with it and you can turn it into very useful information used to, like, big data analytics. So I didn't want that point to go by because that's another really great use of IoT in smart connecting products and assets.

Gavin: So to that end, General Baker, I mean, you mentioned one used case there of the fourth project for the missile manufacture, have you got a few other tangible examples. And again we'll protect the names of the companies here because they're commercially sensitive perhaps. But if you've got a few more examples that you might be able to share with our audience to make this IoT used case more real for them.

Brent: Sure. A couple I would just mention, Newport News Shipbuilding which is a major shipbuilder in Newport News Virginia, they've been using IoT in a particular augmented reality to do some amazing things on the forward aircraft carriers. And they've been using a tool called Beuphoria [SP] to do augmented reality where they were able to take a process that was very manual and labor intensive that would take, you know, they would like plan out what rivets would do on this big piece of metal and it would take them like two or three days to be able to do this, but using augmented reality and IoT, they could do that in like a couple of hours.

The another great example for Newport News which I found amazing is, you know, as they build these carriers, they build them in pieces where they put a lot of box structure there to be able to lift the piece into place and after a while they forget what that added material is. So now they've been using augmented reality to mark that and then later go back and find it using like some type of lens or, you know, glasses, or an iPad, or a phone, and they go back and they can find, "Oh, this piece was supposed to have been removed." So, you know, just amazing time savings and saving hundreds of thousands if not millions and millions of dollars. So, those are just a couple, you know, no kidding used cases where clients have just been able to really make wonderful use of the IoT.

Gavin: Those sorts of stories like really have ground people with this, you know, a phrase that can sometimes be thrown around very easily but you know making it more real for people is very helpful. Ian, General Baker mentioned the phrase AR there where he mentioned at the start of the introduction. This augmented reality side of things, how was that related to IoT and the conversation we're having today?

Ian: So augmented reality is really the best use of technology to take relevant pieces of information about the products and provide that to the end user in very extremely efficient and context specific way. So augmented reality is really best, I believe, when trying to enhance the performance of a worker whether that be on the manufacturing side, on the service side, or in the schoolhouse. I'll give you a couple of examples. So if you need to train somebody on the systems that are part of an F35 air craft, all those air craft are very expensive, you know, in the hundreds of millions of dollars range, and it's not like you can just have one in a schoolhouse.

But with augmented reality you can explore those systems, you can see how the hydraulic cables move the control surfaces and you can interact with the computer model of that extremely complex piece of equipment in a meaningful way. And to give your own maintenance technicians access to that kind of information is really going to transform the way that we're able to train our people to sustain these incredibly complex products.

Gavin: So basically the kids who are using Pokemon Go will be the F35 technicians of tomorrow?

Ian: That's right. And in some cases these kids may never have used traditional technical manuals. In school today they may be using tablets or laptops and when they get out to the aircraft, they'll be using electronic information. In typical aerospace and defense product lifecycle whether we're talking about an air bus A320 or an F35, the end service of these products is incredibly long. We're talking anywhere between 20 years and 50 years. And so, the kids today that are playing Pokemon Go on their iPhones, all the maintainers of five years, six years, eight years from now, and they are used to interacting with this technology. So, to make them efficient, we need to provide them information in a way that they can consume.

Gavin: Yeah, because they're just not interested in carrying big manuals around or trying to sit down in front of a one hundred pages of text in a bazaar, but they need to interact a different way. Is that fair?

Ian: Yes, exactly.

Gavin: Okay. So General Baker, if I'm right, my spies have told me that you might have an interesting story, actually personal experience about the ARPR side of things.

Brent: I have of the…actually, it was an amazing story. So, I was visiting a major OEM agency organization here just about a month or so ago and they're using our euphoria tool for augmented and virtual reality and they asked me if I wanted to be part of an experience, and of course I said yes. So they walked me out in a room, it's was about a 10 by 12 room or so and not a whole lot in there. There's some sensors, and a computer, and some of the glasses that you put on. And so they got me all set up. I put those glasses on and all of a sudden I was right in the middle of a P3 Orion aircraft and the system walked me through how to change out a motor in a windshield wiper which would wipe a motor. And I did that all using augmented and virtual reality. I mean, the maintenance procedures were there, I was inside, it showed me how to remove the part with screws, which bolts to take off. They gave me safety instructions as I went through which just to make sure you're safety aircraft, the power is off. I mean, it was hard to believe that I wasn't standing literally inside of an aircraft and I think that's an important discussion because sometimes I think when we talk about the Internet of Things and augmented and virtual reality which are very connected by the way. I think a lot of folks believe that it's coming down the road or it's gonna happen, but there are many, many used cases where that type of technology is being used today.

Gavin: Okay. Good to know. So, Ian, if one of our audience today wanted to learn a little bit more about how to apply IoT or learn a little bit more about how AR is being used for their business, where could they go to learn a little bit more.

Ian: So I think that two of the best places to go would be, number one, to our website. So on ptc.com, there are different journeys that a customer can take whether be on the manufacturing, or the service, or the digital engineering side, and there are assessment tools that kind of help you charge your path through your IoT journey. So that's a good place to get started to understand where you are today and where you want to be. And also, every year, PTC organizes what we call live works which is really the biggest industrial Internet of Things show that there is on the planet. And I would strongly encourage any interested party to come and attend that event because you'll get to talk to all the customers and you'll get hands on experience of seeing the technology and how it works and listen to case studies from customers that have actually implemented and been successful in this area.

Gavin: Yeah, I mean, that's really at some point I know from my own experiences, there's nothing quite like meeting somebody else who's done the work and having a conversation with them. So that's all for this episode of the connected engineer. I'd like to give a big thank you to Ian and General Baker for joining us and helping us understand what the IoT means to FA&D customers. I hope you took away the importance of taking a step now to understand the IoT as a cornerstone for your digital transformation. If you want to begin your journey, visit ptc.com, and please make sure to subscribe to our podcast on iTunes. I'm gonna look forward to welcome you back to our next episode.


Episodio 1: How to be a Boss at BOM Management


Gavin Quinlan: Benvenuti a The Connected Engineer, un podcast per ingegneri, progettisti e innovatori. Sono Gavin Quinlan, il moderatore, e ogni settimana inviterò degli esperti a parlare delle sfide della progettazione che affrontiamo oggi e delle tendenze future. In questo episodio parleremo di distinta base o meglio di gestione delle distinte base. Molte aziende considerano la gestione delle distinte base come un progetto di totale riorganizzazione. In realtà, iniziare può essere molto più semplice di quanto si immagini. Questa settimana abbiamo con noi Steve Drzewiczewski di iRobot, Eric Horn di Solar Turbines e il mio collega Graham Birch, che forniranno informazioni pratiche, basate su dati reali, sui modi in cui la vostra azienda può gestire le distinte base. Graham Birch è PLM Solutions Director qui in PTC. Iniziamo con qualcosa di semplice, Graham. Qual è la tua formazione e da quanto tempo lavori in PTC?

Graham Birch: Sono quasi 30 anni, cioè gran parte della mia vita non solo lavorativa, che lavoro per PTC. Prima di PTC, ho lavorato nel Regno Unito per un'azienda che progetta e costruisce generatori di turbine, un impianto di produzione di energia su vasta scala. È lì che mi sono occupato per la prima volta di CAD 2D e 3D, anzi uno dei primi compiti che mi hanno assegnato era quello di automatizzare l'estrazione di una distinta base dai diagrammi delle tubazioni e della strumentazione e inserirla nel disegno. Caspita! Ne abbiamo fatta di strada da allora.

Gavin Quinlan: Sicuro. Questo significa che sei piuttosto esperto di gestione delle distinte base e questo oggi sarà utile al nostro pubblico. Per fare una domanda leggermente più difficile che consenta al pubblico di conoscerti un po' meglio. Di cosa si occupa esattamente un PLM Solutions Director?

Graham Birch: In PTC occuparsi di gestione delle soluzioni in realtà significa elaborare una strategia di prodotto lungimirante, nel mio caso per il prodotto Windchill. Parliamo con i clienti per scoprire quali solo i loro problemi, quindi pensiamo a come sviluppare il nostro software per aiutarli.

Gavin Quinlan: Ok. Restiamo sul tema della sessione di oggi. Potresti spiegare al nostro pubblico cosa si intende esattamente per gestione delle distinte base?

Graham Birch: Senz'altro. La distinta base è probabilmente uno dei risultati finali più importanti della progettazione tecnica. Una distinta base può essere considerata come una sorta di ricetta per il prodotto finito. Identifica tutte le parti e i materiali che sono necessari per completare i prodotti finiti e comprende, ad esempio, le parti da acquistare e anche le parti prodotte. La distinta base è dunque realmente una delle principali fonti di informazioni per attività a valle come la produzione, la gestione della conformità, il controllo di qualità e gli acquisti. La sua correttezza garantisce quindi la precisione e la completezza dei dati che contiene, rendendo la distinta base uno strumento assolutamente affidabile. Questo è essenziale per ridurre costosi errori che potrebbero determinare un eccesso di scarti e rielaborazioni, qualità scadente o spreco di tempo.

Gavin Quinlan: Credo che sia un buon insieme di ragioni per cui il nostro pubblico dovrebbe interessarsi oggi di questo argomento. Allora, oltre a Graham, al podcast di oggi partecipano i due ospiti che ho menzionato prima. Steve Drzewiczewski di iRobot ed Eric Horn di Solar Turbines. Entrambi sono esperti del loro settore e possono darci informazioni pratiche e dati reali sulla gestione delle distinte base nelle rispettive aziende. Esamineremo i motivi per cui le loro aziende hanno scelto questo approccio alla gestione delle distinte base, i modi in cui lo hanno attuato e alcuni vantaggi che ne hanno tratto e forse riusciremo anche a parlare di cosa ci riserva il futuro.

[Riproduzione di musica]

Gavin Quinlan: Steve è Senior Manager per le applicazioni collaborative presso iRobot. Allora, Steve, potresti dirci qualcosa in più sul tuo background? Dove svolgi la tua professione? Da quanto tempo lavori in iRobot e quali sono le tue principali responsabilità?

Steve Drzewiczewski: Certo, Gavin. Mi chiamo Steve Drzewiczewski e lavoro per iRobot. Sono in azienda ormai da otto anni. Vi sono arrivato anni fa da PTC, dove ho lavorato come consulente occupandomi inizialmente di progettazione assistita da computer prima di passare al PLM. In iRobot sono responsabile delle applicazioni collaborative, che includono Windchill, SharePoint e Confluence.

Gavin Quinlan: Ho anche Eric qui con noi oggi. Eric Horn è Senior Enterprise Business Process Specialist presso Solar Turbines. Allora, Eric, come per Steve, posso chiederti di dare al nostro pubblico qualche informazione sul tuo background dicendo dove svolgi la tua professione, da quanto tempo lavori in Solar Turbines e quali sono le tue responsabilità in questa azienda?

Eric Horn: Certo, grazie. Essenzialmente sono una sorta di architetto delle soluzioni con formazione in ingegneria meccanica ed esperienza professionale sia nella progettazione tecnica sia nell'amministrazione di sistemi. Lavoro per Solar da quasi sette anni, durante i quali mi sono occupato soprattutto dell'aspetto IT dei processi aziendali e dell'implementazione di soluzioni tecnologiche in ambito aziendale.

Gavin Quinlan: Grazie Eric. Permettimi di tornare a Steve per un secondo. La gestione delle distinte base può essere complicata, per quanto ne sappiamo, quindi, Steve, cosa ha convinto iRobot a focalizzare l'attenzione su questo progetto?

Steve Drzewiczewski: Rientrava nell'evoluzione naturale della gestione di prodotto, così, all'epoca, prima abbiamo iniziato a utilizzare INTRALINK per gestire tutti i dati CAD e parlando con alcuni dei nostri progettisti che erano in azienda prima di me è risultato che erano sempre alla disperata ricerca di modi per fornire informazioni ai nostri produttori. In iRobot non realizziamo i nostri prodotti ma utilizziamo sempre sub-fornitori, quindi essere in grado di comunicare ai sub-fornitori come costruire i prodotti era una sfida importante e decisiva perché non bastava semplicemente acquisire i dati CAD per risolvere il problema. Cercavano sempre modi per tirar fuori i dati, metterli in Excel, manipolarli, aggiungere dati non appartenenti alla modellazione per poi trasferire tutto questo ai sub-fornitori. Trovare un modo migliore per svolgere queste operazioni era una questione importante da risolvere.

Gavin Quinlan: Anche in Solar Turbines, Eric, avete cercato di superare una simile sfida? Anche voi lavorate con sub-fornitori o la situazione in Solar è diversa?

Eric Horn: Abbiamo rapporti d'affari con alcuni sub-fornitori ma abbiamo anche un enorme reparto di produzione interno. Gestiamo le distinte base da moltissimo tempo ma è sempre stata un'attività scollegata dal resto. Il reparto di progettazione tecnica ha compiti specifici e la gestione delle distinte base è completamente separata. L'obiettivo è cercare di evitare fogli di calcolo e attività di "copia e incolla" di tali materiali, talvolta il copia/incolla si estende anche ai disegni CAD. Si tratta di rendere le cose un po' automatiche, almeno un po' di più, con meno procedimenti per tentativi e una maggiore gestione dei dati effettivi. Ci siamo occupati non solo di come gestire la distinta base di progettazione ma anche della gestione della distinta base di produzione.

Gavin Quinlan: Interessante. Graham, hai qualche domanda che ti piacerebbe rivolgere a Steve?

 

Graham Birch: Sì, è veramente interessante. Le distinte base esistono da tempo, quindi non sono una novità. Quello che mi interessa sono le implicazioni di eventuali errori nell'uso di questo strumento. Voglio dire, quali sono le conseguenze negative per le attività a valle quando si fanno errori in primo luogo nella distinta base?

 

Steve Drzewiczewski: Quando realizzi un prodotto commerciale su una scala di un milione di unità all'anno, se sbagli la distinta base le probabilità di ordinare un numero errato di parti aumentano enormemente. Quando eravamo una piccola società, la scala era ovviamente molto inferiore e gli errori erano meno costosi, ma non avevamo tasche altrettanto capienti. Cercando quindi di aumentare la quantità di prodotti realizzata ogni anno, le dimensioni di un ordine sbagliato possono incidere in modo realmente significativo sugli utili.

Gavin Quinlan: Eric, sarebbe la stessa cosa per Solar? Puoi immaginare lo stesso tipo di implicazioni anche per la tua azienda?

Eric Horn: Molto simile. Ci sono implicazioni un po' diverse perché penso che l'intera gestione delle distinte base derivi in realtà dalla necessità di pianificare le risorse aziendali. La gestione di lunghi tempi di consegna per noi è essenziale perché abbiamo parti realmente complicate e costose e cerchiamo quindi di ottenere che specialmente gli articoli con lunghi tempi di consegna vengano ordinati puntualmente in modo da evitare ritardi nella realizzazione del prodotto. La gestione di quelle parti è dunque estremamente importante. Ordinarne in eccesso ha conseguenze altrettanto negative. Vogliamo accertarci che questo non avvenga. Ordinare parti in eccesso comporta costi molto elevati per l'azienda, quindi la gestione è molto importante.

Gavin Quinlan: Quindi, anche se qui abbiamo due aziende molto diverse che realizzano assiemi del tutto differenti in termini di prodotto finale, il problema sembra essere quasi lo stesso e la gestione delle distinte base lo affronta sia per quanto riguarda i milioni di unità che i costi esorbitanti degli assiemi. È così, Steve?

Steve Drzewiczewski: Esatto. Sprecare denaro è sempre un male per l'azienda. Non è una cosa vista di buon occhio.

Gavin Quinlan: Capisco.

Graham Birch: Allora, Eric, come è cambiata la situazione in Solar da quando è stata adottata la gestione delle distinte base? I vantaggi sono percepiti in uguale misura dai reparti di progettazione tecnica e di produzione o uno dei due gode ora di benefici maggiori rispetto all'altro?

Eric Horn: Si tratta di situazioni diverse. Essenzialmente, dato che dovevamo effettuare un cambiamento dei processi di notevole entità, abbiamo iniziato con una sola famiglia di prodotti mentre per altre famiglie diverse la transizione al nuovo corso deve ancora avvenire. Si è già verificato però un enorme miglioramento nel modo di comunicare le distinte base, oltre al fatto che sta diventando più semplice e facile, con conseguente eliminazione dei punti di errore che possono provocare all'azienda molti problemi, smettere di gestire i dati in Excel. Questa è la regola numero uno e l'implementazione dei sistemi consente ora di trasferire i dati dal CAD al PLM all'ERP senza uscire dai sistemi stessi. Il nostro obiettivo è proprio questo. Si tratta di un'evoluzione, quindi le cose sicuramente non cambieranno dall'oggi al domani, ma abbiamo già compiuto dei passi che hanno consentito trasformazioni maggiori. Quali vantaggi sono derivati dall'implementazione di MPM Link per la gestione della distinta base di produzione? La semplice gestione della struttura CAD collegata che consente di creare automaticamente distinte base in PLM è stato il più grande vantaggio che abbiamo ottenuto dall'intero progetto.

Gavin Quinlan: Steve, in iRobot si utilizza ancora Excel o è un tabù?

Steve Drzewiczewski: A un certo punto avevo minacciato di considerare la possibilità di rimuovere Excel da ogni singolo computer in iRobot, ma la proposta ha incontrato molta resistenza. Direi che da sei o sette anni non si utilizza più Excel per gestire le distinte base, quindi ogni cosa viene definita completamente in Windchill e questa è stata per noi una grande vittoria. Quindi, per rispondere alla tua domanda, grazie a Dio, non lo utilizziamo più.

Graham Birch: Eric, so che i progettisti sono per natura un po' avversi al rischio, mentre questo appare un po' un cambiamento nei processi e sembra anche richiedere ai progettisti più di quanto non facessero in passato. Come hai attuato questo cambiamento? Voglio dire, come sei riuscito a compierlo? Come hai ottenuto l'approvazione?

Eric Horn: Ottima domanda. L'ho ottenuta grazie a dirigenti di larghe vedute. La cosa bizzarra è che hai rilevato che può esserci più lavoro da svolgere. In realtà abbiamo constatato il contrario, cioè che hanno effettivamente meno da fare perché gestire il CAD separatamente dalla distinta base comportava sostanzialmente la necessità di svolgere due attività. Ora che questo onere è stato rimosso, possono lavorare con il loro strumento CAD, hanno le distinte base aggiornate automaticamente ed è stato ridotto gran parte del controllo degli errori e del lavoro ridondante. Così ora apprezzano il cambiamento e dicono cose del tipo: "Posso davvero lavorare solo sul mio CAD e la distinta base si aggiorna automaticamente? Perché non lo abbiamo fatto anni fa?".

Graham Birch: Si direbbe che sia stata una facile conquista. È una situazione vantaggiosa per tutti.

Eric Horn: Assolutamente.

Graham Birch: Qual è stata la tua esperienza in iRobot? Hai riscontrato le stesse cose?

Steve Drzewiczewski: Non direi che i nostri progettisti abbiano mostrato grande entusiasmo per l'utilizzo di uno strumento PLM. Indubbiamente ne riconoscono il valore aggiunto. Sicuramente li ha aiutati a non deviare dall'intento di progettazione. Hanno potuto inserire, associare documenti e disegni per analizzare elementi che non avrebbero necessariamente potuto esaminare all'epoca di INTRALINK. Possono fare clic su un pulsante, esportare un report della distinta base direttamente da Windchill e trasmetterlo a un sub-fornitore. È stato un enorme risultato. Man mano che abbiamo fatto progressi lungo il percorso di Windchill, la possibilità di utilizzare package per estrarre tutti i dati in modo da avere realmente un solo repository centrale per tutto è stato un grandissimo vantaggio per quei ragazzi. In precedenza i progettisti avevano sempre difficoltà in questo processo e dovevano continuamente tornare indietro e svolgere attività di qualche tipo per procurare i dati al reparto operativo. Una volta che abbiamo iniziato ad avere tutto centralizzato, il reparto operativo è diventato più autonomo e ha potuto assumersi quel compito togliendolo al reparto di progettazione tecnica. I progettisti erano quindi molto più entusiasti in quanto non più coinvolti in un processo che li distoglieva dalla possibilità di utilizzare gli strumenti CAD e divertirsi a progettare per svolgere invece attività di amministrazione.

Graham Birch: Interessante. Quindi si direbbe che in entrambi i casi, ci sia stata forse un po' di apprensione nell'avviare il processo di cambiamento ma molto consenso e diversi vantaggi percepiti come effetti del cambiamento.

[Riproduzione di musica]

Gavin Quinlan: Permettetemi di fare una piccola digressione per un secondo. Ovviamente, avverto molta positività riguardo al cambiamento, i tanti vantaggi per l'azienda e per il lavoro quotidiano dei progettisti o di chi fa parte del team di sviluppo prodotto. Pur sapendo che a nessuno piace tornare sui propri passi e dire che sarebbe stato possibile fare meglio, però, se poteste tornare indietro e fare qualcosa di diverso, c'è qualcosa che i nostri ascoltatori potrebbero apprendere dalla vostra esperienza con questi processi? Inizio da Steve.

Steve Drzewiczewski: A ripensarci, tutto è iniziato prima del mio arrivo. Dunque sto solo riferendo ciò che è stato detto a me in modo superficiale. Abbiamo eseguito migrazioni dai vecchi sistemi a Windchill quando ci occupavamo solo di pura gestione dei dati CAD. È stato solo dopo due o tre anni dall'inizio del nostro percorso PLM che abbiamo iniziato a convincere gli utenti ad occuparsi di gestione delle distinte base. Avremmo dovuto coinvolgerli prima nel processo, in modo che ogni volta che portavamo un sistema in Windchill fossero subito pronti per occuparsi della gestione dei dati delle distinte base. Probabilmente avremmo dovuto iniziare così, in modo che tutti potessero svolgere gli stessi processi una volta in Windchill. Cercare di gestire due o più processi diversi è stato sicuramente complicato.

Gavin Quinlan: Eric, anche a te la stessa domanda. C'è qualcosa che faresti in modo un po' diverso se potessi tornare indietro?

Eric Horn: Probabilmente avremmo dovuto iniziare prima. Ciò che ha complicato le cose è che utilizzavamo LINQ nel CAD per creare il materiale e superare quell'ostacolo tecnologico è stato il passo più importante da compiere. Era correlato alle modalità di denominazione utilizzate nel CAD, perché ora è richiesta la due diligence e non si possono più inserire numeri parte fittizi nel sistema CAD se si vogliono creare automaticamente le distinte base. Accettare questo cambiamento del processo è difficile perché spesso si inseriscono dati fittizi nello strumento CAD purché l'aspetto del disegno appaia corretto, perché tanto la distinta base è totalmente separata dal progetto. Abituarsi quindi non è facile e neanche tornare indietro. Cosa si fa con tutti i dati legacy che non sono stati salvati nuovamente nello strumento CAD per qualcosa come 10 anni? Alcuni di questi problemi sono inevitabilmente venuti alla luce quando abbiamo attuato l'intero progetto e la trasformazione. Cercare di superare questi ostacoli presenta delle difficoltà ed è stato costoso per l'azienda, ma alla fine ne è valsa la pena.

Steve Drzewiczewski: A proposito di questo, durante la trasformazione dei nostri processi di progettazione, molto spesso abbiamo dovuto tornare indietro e controllare in che modo avevamo eseguito determinate cose, quindi l'eliminazione dell'elenco parti o della distinta base sul disegno è un cambiamento avvenuto nel corso di diversi anni. Si trattava anche di abitudini legate al nostro modo di operare, alla nostra tradizione, che sembrava dovesse continuare per sempre, ma quando abbiamo iniziato a considerare modalità operative diverse, questo atteggiamento ha contribuito a cambiare alcune dipendenze dal passato e abbiamo potuto preoccuparci meno della falsificazione delle distinte base perché ora i dati da inserire nella distinta base sarebbero venuti direttamente da Windchill.

Gavin Quinlan: Quindi direste entrambi che oggi la vostra intera distinta base è riflessa in Windchill e quindi direttamente nei vostri sistemi ERP? Eric, hai citato ERP quindi si tratta attualmente di una corrispondenza biunivoca?

Eric Horn: Sì, per la famiglia di prodotti per cui ora utilizziamo la gestione delle distinte base lo è e semplifica molto le cose anche in questo caso. In passato le discipline erano separate, una per la gestione del CAD e una per la gestione delle distinte base, ma ora invece di due ruoli separati c'è una sola persona che si occupa di entrambe le cose e questo ha reso tutto molto più semplice.

Gavin Quinlan: Steve?

Steve Drzewiczewski: Quindi abbiamo in effetti l'intera distinta base all'interno di Windchill. Avevamo una connessione con il nostro sistema ERP per supportare il settore militare dell'azienda che abbiamo di recente ceduto. Così non facciamo più integrazioni di distinte base in sistemi ERP.

Graham Birch: Poiché alcuni degli ascoltatori del podcast stanno forse prendendo in esame una simile trasformazione della loro organizzazione, quale consiglio darebbero i nostri due ospiti ai clienti che stanno considerando la possibilità di cambiare le modalità di gestione delle distinte base?

Steve Drzewiczewski: Direi che probabilmente uno dei maggiori ostacoli che le aziende devono superare è la sincronizzazione del CAD con... la sincronizzazione dei numeri con le parti WT. Quando abbiamo iniziato, una delle nostre divisioni adottò una numerazione automatica per ogni oggetto, quindi le parti CAD nel disegno e i numeri parte WT erano tutti diversi. Questo tipo di numerazione ha permesso una creazione molto più rapida delle strutture ma era anche più difficile da interpretare per una risorsa umana e questo ha rappresentato probabilmente l'impedimento maggiore di questo approccio. Così, da allora, abbiamo adottato la stessa numerazione per la parte, il disegno e il modello CAD che è umanamente più intellegibile ma richiede più lavoro iniziale. A seconda dell'approccio che si desidera, uno è decisamente più semplice da implementare ma più difficile da interpretare per una risorsa umana.

Eric Horn: Dal mio punto di vista, se si pianifica la pubblicazione in ERP, assicurarsi di definire chiaramente l'obiettivo e assegnare a uno strumento il ruolo di autore e a un altro strumento quello di ricevente. Non si deve permettere che le distinte base possano essere talvolta pubblicate a partire dal PLM e poi inserite nel sistema ERP e successivamente siano accessibili e modificabili separatamente nel sistema ERP. Noi abbiamo disattivato la creazione di elementi e di distinte base con un sistema ERP così tutto passa attraverso il PLM. In questo modo i progettisti usano uno strumento e i responsabili degli acquisti e della pianificazione ne usano un altro e la responsabilità per la creazione e gestione delle distinte base viene assegnata esclusivamente ai progettisti. In questo modo le informazioni vengono trasmesse in una sola direzione e non in modo bidirezionale.

Steve Drzewiczewski: Ottima osservazione. Quando abbiamo realizzato l'integrazione ERP, non abbiamo potuto disattivare la creazione di parti e distinte base soprattutto perché una divisione stava ancora creando cose in modo manuale e l'altra no, e questo ha causato indubbiamente problemi nei casi in cui alcuni hanno preferito saltare il processo Windchill e apportare le modifiche direttamente in ERP provocando problemi a valle.

Gavin Quinlan: Allora Steve, ora che questo percorso nella gestione delle distinte base è concluso, qual è il passo successivo per iRobot in termini di iniziative di sviluppo prodotto? Qual è il prossimo obiettivo da raggiungere?

Steve Drzewiczewski: Ottima domanda. Al momento stiamo pianificando il prossimo aggiornamento da Windchill 10.1 alla versione 11, quindi ci stiamo concentrando su questo. Abbiamo in cantiere alcuni miglioramenti di processi, in particolare il processo di modifica e quello di varianza. Queste sono probabilmente le prossime cose da fare, tornare ad analizzare i processi che abbiamo riprogettato qualche anno fa per migliorarne ulteriormente l'efficienza.

Gavin Quinlan: Perfetto. Eric, anche a te la stessa domanda. Qual è il prossimo ostacolo da superare per Solar?

Eric Horn: Purtroppo dobbiamo fare le cose nel modo più difficile. Ci piace superare più di un ostacolo alla volta. Quindi siamo sempre in evoluzione. Ho accennato al fatto che ci siamo occupati di una sola famiglia di prodotti e per il momento abbiamo in programma di implementare questa e tutte le altre famiglie di prodotti, quindi la situazione è questa e dobbiamo valutare il tempo necessario per realizzare il nostro programma. Separatamente ci stiamo anche occupando di gestione delle distinte base di produzione e pianificazione dei processi, cioè dell'utilizzo dei dati che stiamo recuperando tramite la gestione delle distinte base, quindi stiamo superando anche questo ostacolo. Laddove utilizziamo più distinte base di produzione il team di produzione viene coinvolto prima e l'intero processo è in evoluzione.

Gavin Quinlan: Il titolo del nostro podcast è The Connected Engineer. Ho presentato il podcast dicendo che viviamo in un mondo intelligente e connesso. Ora, posso immaginare che un prodotto iRobot sia già intelligente e connesso. C'è qualche considerazione particolare da fare quando si crea un prodotto intelligente e connesso per quanto riguarda la gestione delle distinte base? C'è qualcosa di speciale che si deve fare in casi del genere?

Steve Drzewiczewski: La sola cosa che abbiamo finito per fare è stata serializzare un numero maggiore di nostre parti per poter raccogliere dati importanti dai robot. Il metodo con cui avevamo inizialmente eseguito questa operazione sulla prima versione del nostro robot connesso era un approccio molto manuale che richiedeva molte persone per l'invio dei dati a diversi sistemi. Ciò che siamo riusciti a fare è stato acquisire le parti da serializzare direttamente in Windchill e inviarle ai diversi sistemi in officina in cui erano necessarie. Sappiamo quali parti verrebbero serializzate direttamente in quanto provenienti da Windchill. Quindi ogni giorno se il progetto fosse cambiato, avremmo visto le modifiche riflesse nell'officina e nei prodotti che si stavano realizzando.

Gavin Quinlan: Cose come i feed dei sensori sono voci effettive delle distinte base o sto fantasticando troppo?

Steve Drzewiczewski: Credo che tu stia fantasticando troppo. In realtà i dati provengono da qualche assieme a livello di scheda che identifichiamo e serializziamo.

Gavin Quinlan: Ok. Eric, non ho una conoscenza simile delle turbine di grandi dimensioni in termini di connettività intelligente. Puoi dirmi se c'è qualcosa di particolare a cui state lavorando e in che modo questo incide o ha inciso sul vostro approccio alla gestione delle distinte base?

Eric Horn: Abbiamo da molto tempo un prodotto intelligente e connesso, ma in effetti non gestiamo la distinta base interna. I sensori, ad esempio, sono un articolo totalmente separato. Nella distinta base è incluso l'intero sistema di controllo, ma non sono specificati i singoli componenti. Disponiamo di attributi che tengono traccia dei componenti essenziali nella distinta base. Il cosiddetto tag ISA è un indicatore di riferimento univoco utilizzato non solo per un assieme ma per un intero prodotto e questo ci permette di collegare i dati dei sensori alla distinta base per individuare la parte che sta trasmettendo i dati. Per la maggior parte, però, questi dati sono in genere monitorati in molti disegni di progettazione e non sempre necessariamente nella distinta base.

Gavin Quinlan: Ok. È tutto per questo episodio di The Connected Engineer. Desidero ringraziare moltissimo Steve ed Eric per aver partecipato e anche il mio collega Graham per averci aiutato a capire cosa può significare la gestione delle distinte base per le iniziative di sviluppo prodotto. Spero che abbiate compreso il ruolo cruciale della distinta base per la trasformazione digitale e quanto sia importante avviare subito questo processo. Se desiderate maggiori informazioni sul percorso di iRobot verso la trasformazione digitale o sulle soluzioni PTC per la gestione delle distinte base, visitate il sito Web all'indirizzo ptc.com/bom. Ricordate di iscrivervi al nostro podcast e saremo lieti di accogliervi al prossimo episodio in cui si parlerà in modo approfondito di realtà aumentata, realtà virtuale e sviluppo prodotto cercando di fare chiarezza in questi argomenti. Grazie ancora.