Tracking without prejudice: IIoT on the shop floor
Manufacturing, engineering and retail: the internet of things is pushing the boundaries of what’s possible in some verticals, while lagging in others. This episode talks to WISER Systems about its 3D tracking solution; a perfect example of effective IIoT.
Show Notes for Series 02 Episode 08
Imagine a 3D model of your industrial facility with objects (or people, controversially) tracked in real time. Perfect for manufacturers, engineering companies, warehouses and anywhere where it’s good to know where stuff is.
Our guest is Logan Maxwell of WISER Systems, who’s at the sharp end of putting this type of cutting edge tech into places that would make a desktop PC or average laptop quake in its virtual boots.
Logan talks about his specific company’s products, but also about perceptions of IoT in general, different ways people try to use the technology, and some use cases that are not quite always benign!
Off the back of a career-ette in start-up manufacturing himself, he’s been with WISER from the beginning, watching it grow from his old professor’s pet project into a thriving business that’s convincing many hardened industry types.
We also touch on how to get positive messages over about technology, and how best to assure people that they’re not going to be replaced by robots if they dare to go to the bathroom too often at work.
Logan’s company, WISER’s website is here:
And Joe’s LinkedIn is here:
Full transcript available.
Joe Green (host): Hello there, welcome to the Tech Means Business podcast. This is a series of audio interviews in which we talk to and chat with people involved in technology that’s specifically focused towards business. The people we talked to are individuals I’ve chosen whom, I think at least, have something to contribute to the area of business technology discourse, over and above what you might call “trotting out marketing collateral.”
Now, I am as fond as the next chap of buzzwords and hip phrases (daddio). So today, I’m delighted to be talking about the Internet of Things, one of those buzz phrases. IoT is one of those areas where the technology affects us all, but we’re mostly unaware of it. Wait at a bus stop in most cities today and the live display board at the bus stop telling you the arrival times of buses: that’s IoT. SMS text messages saying your delivery driver is an hour away: also IoT. And the reason why your air con in your hotel room is set preposterously high: also, unfortunately, IoT.
Some industries are old hands with IoT: manufacturing, engineering, utilities, mining, all these areas have been using IoT for decades on machines, plant, heavy machinery, switches, attenuators, monitors, valves, and so on and so forth. For the rest of us, however, the concept of IoT is a little bit more hazy.
Joining me is Logan Maxwell from WISER systems, who’s made a career out of industrial IoT, or unpronouncably, IIoT as it’s often referred to. (The industrial Internet of Things, IIoT… I’ll our work on it!).
Okay, we’ll talk about WISER’s offering, specifically later on. And we’ll also try and cover off some key concepts behind the technology of the Internet of Things. But first and foremost, let’s welcome Logan Maxwell! Hi, Logan, thanks for coming on the Tech Means Business Podcast, and as has become traditional, please let us know a little bit about you, your career, and how you ended up at WISER.
Logan Maxwell (guest): Sure. Yeah. So I guess to formally introduce my name is Logan Maxwell. And I’ve been working here at WISER systems for I think three years now, which is crazy to think! I started as an engineering graduate from our State University here in North Carolina, NC State.
And out of that, tried to do some entrepreneurial work right out of school. We tried to take a new type of travel mug for coffee and tea with a special kind of thermal layer inside it. So we tried to launch that: it did okay, moderately successful at first, we made some errors in marketing and getting it out there. And as I was working on that, and kind of drove that into the ground, so to speak. I was reached out to by former professor that I had worked with before. He was working on a new sensor technology. And he is the CTO here at WISER systems. So I was brought on really early on at WISER as kind of the first real true employee. And I’ve kind of ridden that wave upwards. So now we’re about quadruple the size, starting at three now we’re at 12. So we’re getting there.
Joe Green (host): So you’re an engineer and the graduate. What sort of engineering did you study? Also, I sense from the entrepreneurial element in that young career that you had a hint of business studies in there as well?
Logan Maxwell (guest): Yes. And strangely enough, it was chemical engineering that I that I studied, and they made a decent attempt at getting us a nice background, in business and presentation, organization strategies; things of that nature. I mean, obviously, the core focus is the engineering, but there is a decent amount of that throughout. They do a pretty good job of that.
Joe Green (host): So you had a call from the old professor. And he’d started WISER or was he thinking about starting a company?
Logan Maxwell (guest): Yeah, so the company had been founded, and it was just getting to the point where they were starting to have pilots, you know, real life product, putting up some prototypes, and starting to get to that growth stage, so to speak. So I joined. You never really know what you’re getting into. But it’s been a fun ride so far.
Joe Green (host): Now, I wasn’t that well equipped, previously about WISER systems, but if I try and tell you what I think I know, you can tell me how far off the mark I am if you’d like! WISER produces systems that model objects in predefined 3d spaces in real time, usually in industrial settings or warehouses or that kind of space? So do you describe it as IoT or industrial IoT: IIoT? Is that the kind of in the right ballpark?
Logan Maxwell (guest): Yes, to some degree. And yes, you’re right on it, it can be considered IoT or IIoT, really, depending on where you put it in. We are designed for heavy industrial environments. So you can think of that as pretty much as industrial as you can get. The sensors are made to work there. And in what they do, instead of imaging, it’s really kind of like a live, always-on asset tracking ability. So, we have little tags that we can place on anything in the facility, be it carts or parts, or people and machines, as they move through facilities and campuses and track them to a very fine degree of accuracy.
Joe Green (host): Now, when you say a campus, that means a specific thing in technology, doesn’t it? Perhaps we should explain: many of our listeners will be thinking about college or university campuses. But here a campus you’re referring to is several buildings or installations that that make up a whole organization, right?
Logan Maxwell (guest): Yes. So when you have multiple buildings under the same organizational title. So, let’s say, I have one building, that’s where they shape the metal, it moves to a second building where they finish the metal, and the third building where they package it and stack it for shipment.
Those are all part of the same process, but they’re in different buildings. So in terms of the way they would link those together, we would just refer to that as a campus.
So with our system, you want to just make sure you’re linking coordinate points. So I can track it from building one and track it as it moves to building two, and so on and so forth.
Joe Green (host): If I were to imagine myself in the shoes of a customer, for instance: I know where all my… let’s call them widgets (as is tradition)… I know where all my widgets are, in production and storage and warehouse, even as they cross the loading bay on pallets and into the back of a truck. So what’s the business benefit for me of that very detailed and very specific knowledge?
Logan Maxwell (guest): It’s going to depend on how you implement it. So we can take this in a few different verticals. So if I’m utilizing this in a healthcare setting, the business benefit is to find that item, that crash cart, immediately, and try to save a life with it.
If I’m using this in a manufacturing procedure, which tends to be where we’re installed, in a lot of cases, I could put this on a work in progress. And I could track its motion through the facility, and know when it hits every station. So not only can I gather metrics on the movement, timing, and analytics of each thing as it moves through my process, but I can also trigger alerts, and send folks to do certain things as it moves through: saving time. A lot of times it’s an efficiency tool. But it really is going to depend on which vertical and industry you’re placing these on.
Joe Green (host): Okay, so we have the hardware side of this and can see in real-time (as near as dammit) where tagged items are. I’m assuming that each of these sensors or tags is reporting back to a piece of software somehow. Is that installed out in the cloud? Or is it installed in house — held in house?
Logan Maxwell (guest): Yeah, so you have, as you mentioned, you have some tags on your carts and parts, you have some WISER antennas in the rafters and a rough 75 to 150 foot grid depending on how open your facility is. And then those are all reporting back to WISER software, which is typically run on a virtual Windows machine on site, and using existing servers.
It can be hosted in the cloud, you know, we don’t necessarily host anything, which has been preferred. And most of the industries we’re in are in aerospace and defense and high level manufacturing. They prefer to host on site.
Joe Green (host): I can see the security implications of that for sure. So for now, we’ve got this information flowing in, but what then?
I was reading this morning actually that 70% of internet traffic is in fact, API traffic. And that’s course machine to machine data flows, as opposed to me sharing pictures of the new puppy on Facebook!
So now, I don’t want to talk about APIs specifically. But I’m assuming that somewhere in the WISER software, there must be sockets, APIs, different ways that the information that’s thus far captured can be transmitted to other machines or other software instances. Right?
Logan Maxwell (guest): Yeah, exactly. The fundamental way people use this system is gathering the data. Right.
So that’s the one of the more important aspects of the system: we’ve tried to make it pretty flexible in the sense of getting the data to the end user in a few different ways. So we have directed database linkups, we have API, we’re going to do a publish-subscribe model pretty soon for socket based architectures. But exactly, yeah, you need to get the data out of the system to get some actionable insight from it.
Joe Green (host): Yes, I can imagine why you wouldn’t want necessarily want a standard dashboard. I mean, if you run a healthcare facility, (you mentioned crash carts, for example), then you need a very different set of views from, for instance, a manufacturing facility. And that agnosticism means that the data can be fed, anyhow, anywhere. So does the system output XML, for instance? So how would I get hold of that data out onto the larger network?
Logan Maxwell (guest): Yes, you’re right, in a sense of just trying to deliver the data as the core product, as opposed to a set dashboard, you’re right on there.
So the data that most folks use out of the system is essentially the tag positions, the tag times and the tag IDs and battery levels for low battery alerts, that sort of thing. So that can come out direct to a database, but it also can come out of an API via serialized JSon. I’m happy to send you more technical documentation on that if you want to post that. But essentially, we’re trying to get: “a tag 42 was at coordinate position XYZ at time 1218.” And we’re just trying to get that to the user. And so they can, or we can bring in a partner that has a custom set dashboard for that industry, or they can just utilize their existing ERP systems. And use the data we’re sending to automate processes.
Joe Green (host): I understand the logic of that entirely, but I have to hold back my inner nerd. You know… talking more about restful API’s, JSon that like!
So I’m gonna have to move the focus on to IoT’s place in various industries. From my outsider’s point of view, I’ve seen logistics and supply chain companies, in this last year, for obvious reasons, come to the forefront of not only the mainstream, but also the technology press.
Do you think that IoT usage is now so prevalent in the logistics sector, that it’ll be those companies in supply and delivery, for instance, who are going to be driving IoT advances? I mean, that kind of IoT “crown”, as it were, has sat on the head, in the past, of engineering and manufacturing. But do you think it’s logistics and supply chain sector that’s leading the way now?
Logan Maxwell (guest): And that’s an interesting question, because I have speak from a reference point of our technology to some degree, which might have a different answer than IoT as a whole.
So in terms of who’s utilizing our technology, we’ve seen logistics and supply chain actually really not utilizing our system because they need more globally connected IoT infrastructure. So I think the answer is, from what I see in the market is in general, is yes, I think they’re looking for new technologies for connected global supply chain management. And some of that’s with some pretty neat stuff. There’s some new companies going out with, you know, low orbit satellites, and small passive tags, basically kind of a replicated GPS model with very small, low power, if not passive tags, I believe at this point. So pretty cool stuff coming out of there.
For us, it’s really been larger manufacturing companies that missed the first wave. They’ve kind of missed a lot of people moving into, heck, even active RFID in places, and so they’re kind of skipping and going, “Hey, let’s make the factory of the future. Let’s bring in the newest technology that works.” And they put it in and make a new event. So that’s actually where we’ve seen, it has been these kind of industries that traditionally were lagging behind or trying to catch up and trying to find the the creme de la creme so to speak.
Joe Green (host): Yeah, I think that’s something that pure IT professionals don’t really appreciate too often. I mean, we’re so enamored by new shiny things, that we want to throw old stuff away and replace every five years or every year, actually, if we can get away with it!
But in some industries, we’re talking much longer lifespans for equipment and machinery, 20 years, maybe, maybe more, maybe 50. Each machine and device, of course, is going to come laden with sensors, attenuators, controls and the like, all of which are still connected and still working. So often, we’re backporting existing technology into these devices. And those devices, of course, as I say, represent massive investments. And that’s not often appreciated, is it?
Logan Maxwell (guest): Now, if you’re talking to the folks that are actually running the processes that are on the ground in the facility, working the machines, stability, and reliability is favored over everything else. You can show them something new and flashy. But it’s got to work. And it’s got to work every time. And it has to work with the legacy stuff.
They’re already running, as you said, they’ve purchased machines and things that are meant to be there for, you know, quarter of a century. And so it’s sometimes difficult for these industries to really innovate when they’re locked into these older machines that are actually what’s creating the product.
Joe Green (host): You mentioned the people operating the machines there. I think that’s an interesting area to discuss with with regards to technology. We’re talking here about environments that are dirty and noisy, sometimes potentially dangerous. And so it’s understandable that people on the shop floor (sometimes literally) they would approach technology with some suspicion, because you know, they’re not in environments that technology is particularly suited to at the end of the day
For you guys, it must be an uphill battle to go into these literally hostile environments as a young tech company, and state: Look, what we’re offering is just as hard and reliable as those devices over there that last 30 years. Right?
Logan Maxwell (guest): Yeah, you’re absolutely right. It’s a very large hurdle. And, you know, I don’t want to be run out on a rail. So we might have to go back and delete this section after I say this! But, you know, I think, industry 4.0, IoT, all of the buzzwords that we’re moving into, have lacked an impact and a punch, because there’s a disconnect between the technology and the implementation. There’s a lot of people out there putting together demonstrations, slick pilots, but have they actually installed the system at scale? A facility isn’t actually being used?
And so the only way we’ve been able to really overcome that hurdle as a smaller young company with very new technology…
I mean, this is beyond, you know, Bluetooth tracking! This is kind of the new, new thing! We have to literally take it to the facility, work with the people in the facility, to set it up and demonstrate it live to them, functioning while the production is occurring, right? So all the machines are running, the welders are on, we’re in an automotive facility, we’ve got earplugs in, and we’re running the system off of a little laptop, and we’re watching the assets move around in real time. That’s really the only way we can overcome that burden of truth. You know: show don’t tell, which I think in general, and again, don’t run me out with pitchforks, is the failing of IoT in general, is marketing flair without real demonstratable installations.
Joe Green (host): That’s something I think that chimes with me every day when I write about technology. What doesn’t sell a product or a solution isn’t that it’s new with a capital ‘N’ or that it handles API’s really well in a very efficient manner — unless it’s an API gateway, and that’s its primary purpose, of course!
But what people are interested in is getting the job done, and going home earlier than they did last week, or going into the boardroom and being able to show one of those graphs that heads up and to the right.
And that’s a disconnect, isn’t it, I think between technologists and a normal people, if you like?
Logan Maxwell (guest): It ties into the earlier conversation of presenting them with a brand new interface, brand new dashboard and additional stuff that in their minds is just going to break; you know, so it’s adding complexity to their process.
So you have to fight against that, especially in our work. I mean, I’m speaking in terms of real time location in that market, but having to be as transparent as possible. We say, “Hey, we’re gonna install these up in the rafters, you’ll never see them. And then data will just get fed to your existing systems.”
So their job is “I need to track parts, work order seven, I go to the same terminal, I’ve always gone to, I type in the same thing I’ve always typed, and it’s just automated there. Right?
So having to simplify that process, because in the end, they don’t really care, typically, what tag is on the item. They want to be able to go to the terminal and find it. And it has to work every time.
Joe Green (host): We mentioned suspicion there as in the suspicion like, “Is this gonna work? Is it gonna break? We’ve seen this kind of crap before.” Another definition of suspicion, or at least another concept around it, is that when people get new technology foisted on them, almost it’s that suspicion that technology as an entity will replace them that will put them out of a job literally. Is that something you find in general at WISER and in the IoT industry as a whole?
Logan Maxwell (guest): We’ve definitely encountered it and WISER specifically, I mean, I can speak of a few individual cases where that occurred. But it doesn’t tend to be a factor if you’re tracking work in progress, or parts or materials or carts or bins.
You know, we have a few people side-eyeing the antennas as we’re installing them, and we have to tell them that they’re not cameras. But really where we run into a lot of resistance in that area, is when they’re on personnel, and tracking the actual folks moving around the facility, you know. It truly it does get a bit Orwellian at times, when you’re putting this on people, and you look on the map and you see everybody moving around in real time.
There’s reasons that you would do this, that are benign. We’ve had folks in industry use us for time and motion studies, for instance: they’ll temporarily place it on a group of folks, watch them move around the facility, they get nice digitized spaghetti diagrams, and they can rework the facility to prevent extra steps, essentially. And that’s relatively benign, and they are doing that anyways.
But we have had a few groups, also putting this on people to tie to productivity, you know. So they’re wanting to say, “Hey, how long is this person on the machine? How long? Are they off the machine? And what did that end up as, resulting in terms of their individual productivity?” So there has been a significant amount of resistance to that, which is understandable.
Joe Green (host): Yes, I don’t think that the supplier of IoT, or any type of tech, I guess, is responsible for that. It’s like all these things. You know, if you buy everyone a copy of Excel, there’ll always be people who prefer a pen and paper and a calculator. I think it’s all it’s all in the presentation to state that a product is benign and positive. But to do so honestly, I think that’s important.
I can imagine people looking up though at the sensors in the rafters, and I’m wondering whether they are being tracked and maybe reprimanded if their smoke breaks are too long. Is that just a case of communication? Or are there in your experience, misuses of IoT?
Logan Maxwell (guest): Yeah, it comes down to company by company basis, and the management culture of that facility, at least, I’m speaking purely from anecdotal personal experience here. So, you know, take this with a grain of salt. But what I’ve seen is in facilities where there’s a very kind of punitive and reward-based system, those are where you get the most suspicious glances in the immediate, saying, “Hey, you know, is this going to track me? Is this going to track me when I go to the bathroom?”
So I think that comes a lot down to management styles within the organization. So as you mentioned, if there’s an open flow of communication as to what this is doing, and why the hope would be best case, even if it was to directly replace a current job that was being done by human, which WISER has seen, that person would then be able to be reutilized in a different way, in the facility or organization and thus improve their processes. Right?
So it doesn’t necessarily mean that, you know, a third of the workforce is going to get laid off and robots are going to take over. But certainly there’s always going to be a suspicion of that and how it’s received ends up coming from the organization implementing it as you said.
So, you know, WISER have the technology and we’ll give it to the organization but we’ve seen it deployed very benignly, everything’s great. And we’ve also seen a lot of suspicion around it. It kind of comes down to the organization.
Joe Green (host): Okay, Logan, so I want to change tack again here, and talk about communication in the technological sense. We’ve done some work in the past with companies using LoRa, and other protocols used in IoT. And of course, we’ve got, you know, buzzphrase number 46b, we’ve got 5G looming on the horizon as the next big protocol, one, which will probably have the biggest impact in IoT, of course, rather than in the consumer market, I suspect, at least initially.
So, without giving away trade secrets, can we talk about the nitty gritty of the WISER system? How are those sensor arrays talking to the base software?
Logan Maxwell (guest): So we’re using what’s called ultra wideband. We can use a number of different channels within the subset that’s called ultra wideband, but essentially, it’s just a very wide, short pulse. And we’re using sophisticated timing algorithms essentially.
So you can think of it like, our little antennas that are in the rafters are all cell towers, all the little tags are little teeny cell phones, and they’re all just making calls, every so often, and are user configurable, right?
So I could ping once a second or multiple times a second, or once every few minutes, that sort of thing. And when that call is made the antennas, essentially, all know what time it is, and know what time they received the call — that sort of thing. And that’s all in very short, wide pulses, which, due to the physical nature of the wave, make it a lot easier to pick up in reflected crowded environments.
And so all that’s pinging back and forth in the facility, there’s a little WISER gateway box that is given an IP address, and that sends it all back to the software, which we mentioned can just be hosted in a VM on site or in the cloud (if it has access to that IP address). And that’s all then calculated. And from there, we send the coordinate points.
Joe Green (host): And so, are the calculations are being done by the sensors up in the rafters, or is the grunt work being done by the server?
Logan Maxwell (guest): Most of the grunt work and calculation is done on the server side with the software there. There are some calculations done on the antennas, but the majority is done on the software.
Joe Green (host): So earlier on, I touched on 5G: a dread catchphrase, like AI, that always get mentioned, usually as indicative of THE FUTURE! However, as someone who lives and breathes this stuff, do you see a killer 5G product? Do you see 5G changing things overnight? Or will it just create a step change or two?
Logan Maxwell (guest): Ah, somewhere in between the two! I mean, it’s a step change, it’s a larger step change, right? So to us it’s okay, we’re moving from X speed to Y speed, you know. So it’s viewed a bit as just kind of an enhancement to what already exists in terms of the speed that we’re able to send data.
For us, as a company, it’s not something we could really utilize on our technology side, inherently, because the bands aren’t really wide enough to do what we do. The new Wi Fi protocol is exciting to us, because it might be wide enough to do some calculations. So that would be very interesting. And we’re looking at that.
But in terms of 5G, I think it’s picking up a lot of marketing traction. And we’ve seen a pickup of folks just kind of using it as blanket terminology to explain new technology.
So we’ve actually bid and won some “5G” factory proposals, even though our our technology is not really technically under that umbrella.
Joe Green (host): So back in the day, we used to ask guests whether there was an event that you guys would be attending or a trade stand, maybe where people could come and speak to you guys. But sadly, that’s all passed, at least for now. So if people want a bit more information, what should they do?
Logan Maxwell (guest): You know, more information is always at our website, you can enter in a form and somebody from our organization will reach out and answer some questions for you.
Joe Green (host): Yeah, one thing that any demo of the technology that WISER systems offers is that it can have really cool visuals, like glowing dots moving around a 3d space in real time, which is going to take many boxes for the for the marketing department. I mean, god, try selling a hyperconvergent networking software or load balancing software running in a container! Anyway, I digress. We’ve run out of time, so check the links in the show notes.
And I just have to say a big thank you to Logan Maxwell of WISER systems for joining me today.
Logan Maxwell (guest): Okay, Joe, I really appreciate it.
Joe Green (host): Right. I’m off to find some pictures of models smiling and pointing at laptop for the visuals as part of my new marketing drive for this podcast. Or perhaps a businessman punching the air on the steps that lead up to an office block…
Join me next time on the Tech Means Business podcast. See you soon.
23 September 2020
9 September 2020