Well, welcome everybody. We've got a special guest with us today. This is, Nick Rivett, who is a senior MES consultant here at Parsec. And he's gonna be taking us through some lessons learned from the field with regard to edge devices. This is is an important topic because you never know what kind of gremlins are gonna come up when you're trying to implement technologies like this. And so, Nick's got some firsthand experience that he's gonna share with us today. So I'll let Nick do his thing, give an introduction and you know walk us through some presentation stuff. So Nick, it's all yours. Take it away. Thank you very much Ryan, appreciate that. My name is Nick Rivette. I've got a background in manufacturing systems going back more than twenty five years now, starting in semiconductor and then doing a turn in biotech. And so have a lot of lot of things learned the hard way over the years. So I want to share a little bit of a presentation about edge devices, the lessons we learned from the field, talking about our own edge device solution here at Parsec, as well as general challenges that you're liable to run into that should inform how you approach doing edge solutions, and then, maybe a little bit of a cautionary tale, before wrapping it up. So I will begin by talking a little bit about first off, what are edge devices? They are they are devices that are connected close to the equipment, with with the, intent of simplifying, getting data from the shop floor into your into your manufacturing systems or your data fabric as easily as as seamlessly with this few configurations as possible. And what you see pictured here is our Parsec solution in this edge space, which is always a collection of universal data collectors. You see pictured and then coordinators, which is the little device on the right that knows how to talk to those collectors. So the collectors correct connect directly to sensors and then the coordinators collect the data and pipe it up to your system. As a representation of that kind of architecture using our smarts our smart devices, You have any number of data collectors on the right pictured. Generally in three categories, analog data, just signal meant to be captured, things like temperatures and pressures are like that. Digital recognition of things like photo I being crossed, so counters. And then, you know, even serial solutions, you know, we're all very familiar with scales as a clear example. They often support a serial communication protocol that allows you to gather the to do things like tear the scale and then gather the weight off the scale at request. All of those things feed up through smart data collectors, and then those in turn are able to broadcast to one another to relay and then send information to the smart coordinators, which are connected to the factory network on the floor. And from there, it's just one simple web service connection to get the data into TrakSYS. So very, very easy architecture, very easy picture to draw. But but we all know real life is is got some complications to it. So I want to spend a little bit of time talking about the challenges, that exist when it comes to doing edge devices, whatever your solution is. The first set of challenges around infrastructure on on the factory floor, you may have older networks, you may or may not have IOT segmentation, VLANs, and how those things are set up, how wireless coverage is done across the the factory. All these things can be challenges. When it comes to factory setup and and and its ability to service all the signals, depending on how you how many sensors you have and how much data you're streaming, You have to consider, do you are you gonna affect network throughput and produce latency problems? And then the power and placement devices are always always a challenge. Right? So getting getting power to the device, making sure that it's protected if it needs to be protected, safe safely installing it near machinery, being aware of the fact that noise coming off of some machinery can also interfere with the broadcast of the devices. All these things have to be borne in mind, and often you can't learn all of those things without tripping over one or two along the way and learning how things are affected on the actual floor. Another set of concerns around security and ITOT convergence. You know, are there vulnerabilities? Adding new devices to the floor, increases the likelihood that there could be attacks of those devices. So you need to make sure that you do things like set up proper proper authentication. Don't just leave things on default publicly known credentials. You have to make sure you're working with your security team, and depending on their level of maturity in ITOT convergence, you may be either educating them or them educating you in terms of things that have to be complied with. You know, the differences there are differences in focus. Operations has got more of a focus on keeping the keeping the trains rolling, keeping the material moving through the factory, whereas the the security team is more concerned about risk, which is their role. Patching devices also becomes a concern, you know, even devices that are not Windows based still can be prone to security laws over time occasionally. And another another challenge is around data management and integration. So your machines may have different protocols. How do you accommodate those? Do you use common driver sets? And these are all things getting that data into your MES can be a challenge, depending on the protocol even. Data quality is another concern. So edge devices deliver raw signals, but it has gotta be tied in with context and awareness of things like production or time when things were gathered. The quality is also a concern. It's that you still have the garbage in garbage out concern, and we've seen some of that where, you know, nothing will help you if a photo eye is sensing anywhere from five to a hundred count per single item that goes past. And then and then, from there, also storage and synchronization. And one of the things that strikes me is we tend to think of cloud storage in the in its cheapest form as being inexpensive, but you'd be surprised how that data that data charge can add up over time, especially if you don't have an archival and or even a removal scheme over time. So you have to give those things consideration. There are also organizational and change management challenges. Right? So you may have resistance to change. Production teams are just, you know, disinclined to make any change. And so bringing a new system in place, and using an edge device, All these things create, you know, just plain human resistance challenges. And then also there's there's maintenance elements to these things. There may be a skills gap, we've we've can share some stories about that as well that your team may not have your, that your team or the team at the site may not have experience configuring the edge devices. And if, if you don't, if you don't understand all the things that can go wrong with that, that becomes a challenge to get the right people in the room. I'm coordinating across department and we'll talk a little bit about, you know, the kinds of people who have to be involved with these setups later on in this presentation. So, you know, it can be a challenge for cost and ROI validation. Edge devices, you if you're not able to show the ROI immediately or doesn't map to money, Any any performance improvement solution really only gains its value when it highlights their issues and then the the factory takes action on those issues. If you do too many small pilots, then you can get in a situation where you've got very localized success, but higher up the leadership chain kind of a okay. So what response to those, and then the the ever present worry that the vendor that there's a vendor lock in concern. Edge devices increasingly are are are portable or less fixed protocol, and certainly, you know, the Parsec solution doesn't is more of a work with whatever you've got approach, but not all vendors have the same idea when it comes to that. And then there are also regulatory and environmental factors to be concerned with. So things around compliance and regulated environments, you know, you have to be careful about existing equipment validation. Does it make more sense to put secondary sensing that that is used only for performance and not quality decisions if you've got a validated environment? And then also the environment itself, you know, there are there are harsh conditions to be concerned with, you know, vibration, dust, moisture, powder floating through the air, and one of the more recent implementations I've been involved in. And that that that affecting the signal of the counters is a concern. So all of those add up. So in theory, we've we've got the ability to connect, and that is simple. We've got our our smart devices on the edge and the and the TrakSYS server just listening and, whoever the solution owner is, usually some production area, whether engineering or operations, and then someone on the on the services team, in contact with them. They they set up the network connection to connect the smart coordinators that talks to TrakSYS and it all just works. But the reality is there's several several people and several, other bits of hardware or configuration to concern in the middle of that. Certainly between the smart devices and and the TrakSYS server or your edge device and your, your data collecting solution, There are cabling vendors. For most companies, they don't have these people living in house. They they contract that out. They they run the networking cable to the to the routers and switches, and then you've got two teams within the global IT team that don't necessarily always communicate well with one another, one responsible for networking configuration and another responsible for security and firewall configuration. And you may be in a position where, you know, if your solution owner is not the rare source who already speaks IT, you have to direct the conversation to make sure, oh, okay. Based on what I'm seeing or what you're seeing, perhaps you need to talk to this person on your team internally. So want to want to reflect on that and and talk about a few common takeaway or a few takeaways that you can use to be prepared. One is that that make sure that as much as you can, you are you prefigure your, infrastructure configuration, even before you send the devices because then, this sort of thing can allow your customer to make sure that they're they're ready, they've got all the the powers available, that they've got the the mounting plan, and that they've got, they've got the LAN configuration such that, they have a awareness that the network is connected, even before the devices arrive. You can prepare in advance to make sure that you have the devices online, ready when they're when they're shipped. You can go so far as making sure you you provide to the customer quick start guides, little configuration templates that just provide lead them to the information that they need to gather in order for you to be able to configure the devices to get the signals that you expect to get from each one of them depending on location and even the type of device that we're talking about and the kind of signals coming in off of it. And then making the relationships across the rest of the team. So, you know, I think that as the provider for the implementation, making sure that you're prepared to manage the gray space between the IT department, the OT, even the cabling vendors so that you when things aren't aren't working or, when you're having difficulty getting started, you have the ability to bridge that gray space between, communication if if the solution owner on the customer side is not able to do so. And those were the things I wanted to make sure we talked about today. Ryan, any closing thoughts? Yeah. I had let me fix the camera here. I had You and I chatted about this a lot and I wanna bring up because I think it's interesting. One of the stories that you had where you mentioned companies will outsource like cabling stuff and just how important it is to communicate. We think that, like you said, it's kind of like plug and play with these things. In reality, there's a lot of steps. Just one unforeseen thing was that, sure, the company procured the cabling vendor, the vendor dropped a bunch of lines, ran them all, but didn't label them. So it's not like you can go over to one side of your room and look at the cable and then go to the other side. I mean, right? These are very large areas and maybe these cables are hidden and all that stuff. So just things like that can you know, add delay and stuff that you've not really been thinking about. So I thought that was an interesting story. Yeah. True. Know a true story. A true story. And and, you know, even the level of, like, not not writing down which which of the two inputs on the smart collector was Right. Taking the output on a machine. Like, those those little nitty gritty details that force you to go in and have to look at it. When networking, like you said, with a long run, you may even have to tone out a cable to figure out where it was connected if it's if the if it wasn't clearly annotated. And those things as you say always take time because you've got to get the right equipment in there to do the testing after the fact, and would better to make be to make sure that it's written down as it was written down. Yeah. For sure. For sure. Right. Sticky notes or something. Well, good. Well, hey, look, I appreciate your time today. I'll say to the audience, know, if if you're interested in, you know, looking at edge devices or MES in general, know, feel free to reach out to myself or Nick. Visit parsec corp dot com and yeah, we can get any questions that you have answered. With that, I'll wrap us up here. Thanks a lot, Nick, for your time. Appreciate it. Thanks, Ryan. All right.
