CPG Manufacturing Masterclass: Fixing "Pilot Purgatory" with UNS & Data Foundation

speaker-0: Welcome to this Abelara livestream for the Prove It series. I'm here with Dylan Dufresne. He's the architect, co-founder of Abelara. I'm Zach Scriven, digital marketing manager and solutions consultant for Abelara. And we're here with our partner, Ricardo Santos from DuPinix. Today we're going to be talking about Prove It and I'm going kick it over to Dylan.
speaker-1: Yeah, so this one we're going to go into a little bit more about the details of what the virtual factory is, what are the use cases, what industry are they in, and why we brought in Tupinix to help out. So yeah, we're just going to start off a little bit with some backstory here. The unified namespace for Proveit this year is split off into three different enterprises. One of them is Enterprise A, B, and C. We chose to use Enterprise B, so that's the one we're going to go into detail today on. It is a multi-site company, a bottling company, and Yeah, being a bottling company, that's where we brought in Ricardo. But going back to why they brought us in, we're going to talk about some of the hypothetical situation and the use cases and the value we would bring that we'll be talking about on stage. So being a multi-site bottling company, Capyauer Inc. is what they're called. And they approached Abelara after completing a unified namespace proof of concept with another integrant. They had invested significant capital into what they were told would be their single source of truth. And then... They realized that even though it was promised to be the foundation for the digital transformation, the backbone of industry 4.0, they were still stuck. Six months later, years later, pilot purgatory, the unified namespace was in place. We had an MQTT broker, but nobody was able to use it for anything. So while they had a proof of concept that proved that they could connect and collect and store data, there was data flowing from PLCs into an MQTT broker. topics that were published and dashboards that were shown. There was a MES database that had some extra context and some master data in it, but they still didn't quite have the visibility insight or the single source of truth that they were promised. So yeah, worst thing they found out is that they didn't actually own it. And that's where we get into the story of VendorLock and all of that. So they brought us in because while they had their MQTT broker and they could see the temperature of the tanks, they could see the part counts off the robots and the palletizers, couldn't quite know what that all meant. So the MQTT broker might have a state of running or idle, but it didn't, what states were possible, didn't exist. It could tell you what was in a tank, what material was there in that moment, but not what material could be there. There was no master data definitions. And yes, it was in that MES database, but that's where VendorLock came in. So they put in the database, but they couldn't access it. They couldn't query that database and get that information and make use of it without contacting their integrator. and paying for more phases. really the application was inaccessible. They had dashboards, they couldn't extend them. They had a database, they couldn't access it. All they had was this data in an MQTT broker without any context, no units of measurement, nothing. So they called us, Abelara came in and our job was to scale this proof of concept and get them out of purgatory and get the pilot into a scalable solution. So to do that, we have to add context. So they already had real-time telemetry data from their equipment, things like temperature, flow rates, weights, and pressures. They had equipment state codes, like we said before, running idle downtime, et cetera. And they had some OWE calculations and handful of other NQTT topics, asset IDs, descriptions, things like that, organized by site or area in an ISA-90. But they didn't have any way of tracking that material. They didn't have any production context or state history, and they had no work order integration. And they definitely couldn't see what was in inventory anywhere. So they have their scaling problems, they had ownership problems, they didn't really understand the architecture, and they didn't have any ownership or independence on being able to scale it themselves. So we came in and did an assessment, this is again for people just joining, theoretical based on the Proovit Cappy Hour Enterprise Company in the virtual factory. And in this situation, Ibelara comes in, we did an assessment, we looked at the UNS and talked to everybody in that company, from the CEO down to the operators. to identify where they wanted to go, where they expected to have gone, and why they weren't there yet. And it confirmed what we already suspected. The existing unified namespace was just simple telemetry data without any real context. It was unified, it was a single namespace, but it was hardly a full unified namespace. So yeah, it connected equipment to a broker but didn't unify operations. There was no semantic model, no master data alignment, and no integration with business systems. And they had no way of adding context over time. So we began our architecture work and you can see some of that in previous videos in this series if you check them out on YouTube. And after we got that architecture in, we were working with them. We worked with them to install the infrastructure, get everything ready to go. But architecture alone doesn't fill bottles. So that's when it came time to talk full implementation. And that's where we brought in a partner who is an expert in this space who knows bottling like the back of their hand. We brought in Tupinex and the rest is history. So we'll let Ricardo introduce himself here. and talk a little bit more about capybara and bottling.
speaker-2: Welcome Ricardo. Hello guys, yes. Thank you. Yeah, thank you for having me here. So it's really exciting. Our first live stream for hopefully the first of many to come. So ⁓ on this partnership that we're doing, being together, prove it. So we are getting ready, getting a lot of parts moving here on the background. So this is great. Great start. So first I want to introduce myself, let everybody know who I am and why you should listen to me. Right? If I'm so called the industry expert on the CPG and food and beverage. So my name is Ricardo Santos. This accent comes all the way down from Brazil. I was born and raised in there. went to engineering school. I started my career down there. about 2009, 2010, I went straight into CPG, straight into food and beverage. My career, I started working straight to Coca-Cola, PepsiCo, Unilever. So I didn't know much what it's like to work in For The Mom and Pop shop. So there's nothing wrong with it. was just that my experiences, when I go to a customer, they usually have ERP, they usually have MES. So I know a lot of people's experiences say, well, not everybody have MES, but because I work with these big players, that's... That's the history, right? So about 10 years ago, I moved to the U S and again, working for food and beverage, doing PLC controls and MES databases. You were on that live stream with Travis Cox and Remos. You guys were talking about how things used to be back in the days. I was using a lot of APIs, making ODBC connections on Factor Talk View and trying to call, you know, start procedures from there. So we all know how using those tools used to be like. So that's why I consider myself kind of the legacy, if you will, or industry. three system integrator doing PLC controls and using those former tools. I went to work as an innovation and digital transformation engineer for Nestle. So overseeing all North America and Canada, right? So that's when I got myself more into the industry four kind of projects. And I used to do things kind of the old way. So all these problems that we talk about the unified namespace, that all these problems that unified namespace. solves all these data silos. I didn't hear about it. I lived it. So I've done it. I was there. So unfortunately I learned the hard way. So we spun up some initiative here with really no vision. We're trying to get value out of it. It didn't really work. And then you go to the next one and this system doesn't really connect. You start to make a lot of point to point connections. Right. So that's what I lived firsthand until I found the community and learning a new and better way to do it.
speaker-0: I got a quick question. We got a second for Dylan and for Ricardo. And I don't want to hijack the stream because I know we got a very packed agenda for the audience today. I was talking to one of my coworkers here at Kilm. We all work at different companies. And I was like, ⁓ what do you do? He's a process engineer. And he designs, believe it or not, manufacturing facilities for food and beverage. And I was talking to him about what we do. And it was funny. He said, You know, there are new plans going in today where the said, you know, for a lot of our customers, as long as the batch gets produced, that's good enough for them. And it blows my mind that like this, this, world is like foreign concepts to a lot of people. Like even someone designing the plant is digital fluency is relatively low. Now, no, no disrespect to him. You know, he's a process engineer, right? But it's like, you know, the customers aren't asking for it. Like, and I'm like, How do you address, like real simply, Dylan, how would you address that, like, the need for digital, right? Like, I talked about perfect batch, like, collecting data, analyzing it to create the best perfect batch. You can't do that without AI or collecting data in the first place. what was your guy, like, how would you address the need for digital for CPG?
speaker-1: I think the first part of this, and this might be the more controversial one, is not everybody needs it at the stage they're in either. It's usually good to start with some form of it because it's easier to start with it than it is to add it later, but some people do just need to make that first batch. You can't improve a batch if you can't make a batch. And a lot of places, especially with the markets that they're in, some industries don't have the same level of competition as others. So sometimes just making it is good enough to set you apart. But if you want to scale from there, you've really got three different ways to do that as a business. So like when we come in and do workshops or anything like that, when we're talking, especially to the leadership teams and the executives, there's kind of three different goals that they usually fall into. You either want to increase revenue, you want to decrease cost, or you want to increase quality. Sometimes you need digital to get there if you're already doing everything else well enough. But if you don't have the processes in place, you don't have the people hired, you don't have that process even engineered or designed yet, let alone the business process, then technology and as much digital anything as you want won't do much good if you don't have something to digitize first.
speaker-2: Back to Zach's point, how easy it is to tap into these machines once they are already running. Because if you have the strategy from the beginning where I want to make data my primary commodity and therefore I will buy this machine from you if you make this data accessible. Because now I try to go into, especially on the filling. machines and a lot of that that we will talk about here today. No, I will not give you the PLC backup. You cannot get any data from it. And if you do, I will charge you per tag and then maybe I'll put in a data concentrator. And as long as you pay me per tag, I'll make them available to you.
speaker-1: Yeah. And calendar, I said before, some people are lucky enough to start Greenfield and they can start to standardize on some of these things upfront. And with the right digital fluency, I think you would call it, you know that questions to ask upfront. So even if you're not going to put in the full digital architecture from day one, you should at least know enough to talk to your machine builders and the OEMs that you're buying this equipment from and say, I need the data in this way and this format and... accessible and this option. You don't want to get locked in because we're not thinking about it at all. You'll oftentimes get locked into like what you were saying, Ricardo, where you can't get the PLC backup. You can't control it. You can't even access it without breaking warranties or and once you start breaking that warranty, you don't get support anymore and things kind of fall apart from there. So a lot of people don't go down this path because they asked their OEM too late about the device connectivity piece. When it's an afterthought, it's harder for them to do too. mean, they're not always trying to be malicious either. I mean, if they've already written the code and commissioned the system, there is a lot of risk on their side of letting everybody access it. I mean, we were also on the control side here. mean, while I don't always agree with the methodology, I can at least empathize with not wanting to support something when other people are in there making changes that you don't know about. It makes it really hard to support. And while I always err on the side of ownership and training, I have the empathize with why they go that way too. The other one is once the project's done and they've already written the code in a certain way, it can oftentimes be a lot of work for them to go back and redo things or add new things on top of or bolt things onto the side of and they shouldn't be expected to do that for free. So the important thing is to know upfront to what questions to ask and what foundation to be laid. And then from there, maybe you don't need a whole digital architecture or a unified namespace, but you know that when you do need it, it's ready and you can connect to it.
speaker-2: Or in this case, on the CAPI hour, you start collecting the data, you make the data available in there, but sometimes can lack documentation on what's available in there, or what was the intent? I know that we always talked about this, right? So what are the problem you're trying to solve in this CAPI hour will be a great example of something that happens all the time, which is... Hey, we have the data here. Great. We have the infrastructure and then what are the use cases? And I feel like that's good time to bring me in because I know the bottling companies. I know how CPG works and we can make use cases together. We, customer usually knows, right? They have somebody with a spreadsheet that is collecting that data manually in there. That's a great place to start. I always say, right? So if you have somebody with a spreadsheet, let's talk to that guy. let's try to make that process an automated data collection for that one. So I think that's what we're going to talk about here today that we are already talking. Especially once you get up to more of a... You start growing and then you get management positions in there. You start asking questions. So how much does my product cost to make? So how much energy I'm consuming? So different skill numbers, right? So at 8 ounces bottle versus a 16 ounces bottle, I have different packs. So in this virtual factory, have, so how much per skill number does it really cost or do you just average it out for the month? And then eventually you average it out for the year and you say, well, an average is this much, right? But which product, which skill number is more efficient to run? Which one you should be trying to push more or? less or which one you should maybe eliminate from your portfolio.
speaker-0: If Glenn, I know Glenn's watching, he would be saying, and I don't know what my variance is on my gross margin until the quarter end. And I'd love to know that in real time.
speaker-1: Yeah, it ties in too. mean, just because a SKU takes more to make or costs more to make or runs the line slower, you also need to know what does it cost? What's the market demand and what's the supply chain for it look like as well? So then that balance and that trade off between how hard is it? How expensive is it to make versus how easy is it to sell? And how much does the market want it comes into that balance where you can start making decisions based on real data. How expensive is it? Does the market even want it? And find those balances and those ratios to make those real decisions that you couldn't otherwise make outside of this gut feel or somebody spending a month digging through Excel sheets and ERP tables to run a report on.
speaker-0: The other thing that I just quick thought back on my point was like, if the manufacturer is demand constrained and not production constrained, then they probably would be better off like investing in product or marketing or sales. You know, but if you're looking at capacity issues, that's where digital data would obviously come in, you know.
speaker-2: and Sorry, I was going to say because sometimes you're saying, you can produce more if you make your machines more efficient. But like you said, if you don't have demand, if you can't sell more, right? And then now, can you produce the same thing at a lower cost with a better quality?
speaker-1: digital data.
speaker-0: Let me ask you this then. Are you seeing more CPG companies doing their own manufacturing or are you seeing more manufacturing as a service to then serve CPG startups until they reach a level of volume where then we're going to take manufacturing in-house once they're like a Chobani or you know like a Nature's Bakery or something like that. Because if you're at a small scale you're right it's like millions of dollars why would you invest so much into when you could just go to a Copac or maybe Copac would be for the packaging but like a co-manufacturing facility.
speaker-2: Well, that's what I talk about the buying behavior. What I see as a buying behavior is, hey, I need to solve this problem, right? So I'm installing a new line or I want a new filling machine. So a lot of times is I go to different filling machines vendors and then I do, I built what I call my best buy model and then I set supplier A, B and C and then I pick between these three vendors. That's what I see a lot, right? And now the problem is with technology can be a real problem too. Right? Because and then you say, what software vendor should I go to? Because again, you're going looking for the magic software out there. You're looking for the monolithic approach, right? So which software vendor I want to do digital transformation, which they call digital transformation software because that's the buzzwords that they sell, right? And that's a problem because now you're using the same buying behavior that you had for machines for software, right? Which We have similar experience here. It can go with software vendors and integrators. I think Dylan and I, have very similar vision on this, right? With similar outcome. Dylan and Abelara, they come in very good in this because, and then you're agnostic to the software. trying to, you're looking at this problem you're trying to solve, going out there and helping them right through... Awareness, right? I like to talk about the add car change management, right? So you bring awareness first, so then you create desire, knowledge and all that. So Abelara can bring all that in. And I think when we come in to PNIC for especially for CPG here is that we follow Abelara's guidelines and architecture. So the same way that you have your Realtor, if you bought a house, you know that they like to work close with the lender. And why is that? because they both want the deal to go through. And Abelara and 2Penix is the same thing. We both want the customer to be successful. We are on their corner, right, protecting their best interest and not that software A, B or C, but really the customer. And that's why I trust Abelara to go out there to the customers. And I do believe that they trust me to... pick after them and do the implementation services following their guidelines, following their architecture because we will both be successful together. So that's my message.
speaker-0: We're the dream team you guys heard it here. I love that analogy by the way nice
speaker-2: There we go. It's nice. Hey, I do have that presentation. Do want to, can I jump into more maybe of an educational for engineering people that are going out to prove it. They will get to know enterprise B bottling and what are the use cases that people try to solve. Right. So this is for everybody, engineers for whoever's out there. me.
speaker-1: start sharing your screen and I'll frame it here quick. The Ricardo's got a long history at a lot of different CPG and especially a lot of different beverages, beverage manufacturing plants, better bottling plants. And from that he's learned a lot of the things that can do with the stuff I was talking about, increase revenue, decrease cost or increase quality. And he's got a presentation here right now that's going to go a little bit into the weeds, but also stay high enough level that you can follow along. and really show some of the math and the engineering and how we get to those outcomes. And some of the stuff that we're going to be doing for ProveIt so that you can see how we implemented it, either in future webinars or on the floor and in our presentation. So yeah, Ricardo.
speaker-2: Very good, great. That's it. So let me share that in here now. And I'm pretty sure you guys can see the screen, right? So this is officially the Capy Hour logo. No, no, it's not official, but it can be, right? The Capy Hour Inc. soda bottling company. That's Enterprise B. Back to Zach's question in there. Why should you go digital? This is usually my question for data visualization, right? So I think about Homer Simpson. Homer Simpson just looking in there. I ask three questions. Where are we today? Maybe it's half day through production, we are 40 % of target. How did we get here? So can we see trends because our mind is capable to predict where we're going to go based on previous patterns. So how did we get here? Well, we had a rough start. We had a changeover earlier, couple of hiccups, but now it's been steady. All right. based on the current state, where are we going? Now, well, we're going to be about an hour late. We're going to have to pay everybody over time, right? So that's usually the questions for data visualization that I like to ask. I cannot see you guys in here, but I believe that you're with me, Yeah, we're. Yep. All right. So getting down to the weeds, what are the most common challenges in beverage specifically? Waste is usually a good use case.
speaker-0: We're good, yeah.
speaker-2: So how do we fill a bottle? So we know that we have 500 milliliters. So if you think like me, I did a metric system for 27 years in my head and then I had to get to the Imperial with fluid of ounces, but I still think as metric first. So 500 milliliters, how do I fill a bottle? So if I fill 10 bottles, does it mean that I use five liters? No, right? So let's talk about how do you fill bottle. So if the label says the volume is 500, legal and regulatory requirements ask for what's called a required overfill. So you have to overfill the bottle, right? So this is to ensure that the customers receive the net content that is on the label, right? So plus the volume label, now you have the required overfill with the tolerances that ⁓ like Zach said that I know Glenn would love. So this is mandatory. It's in green because I have no choice. I gotta do it. Now, the first villain in here, it is what we call the access overfill. So if we set the machine to fill 508 and it fills 10 milliliters extra, now you're getting to 518 per bottle. To the customer is free product. I love it. I get a little bit more, but... Worse than that, the carbonated beverages specifically, they need some head space over here. So if you don't have that head space, you're going to go through a vision machine system like let's say a Presco or whatever vision you have and it's going to kick out and you're going to increase the rejects because of the internal pressure.
speaker-0: Overfill it and then potentially lose the whole pro- whole can.
speaker-2: Right, right. The whole can, that's the same thing with the pressure. Funny enough, that's why the can has the round bottom in there, specifically for managing the internal pressure. And to Zach's point is, I'm just giving free product. No, you're increasing the rejects as well. So now with a flow meter that we have, now talking about the data that we have on Enterprise B, we do have a flow meter. We can totalize. We know the average prepared bottle that is going into the bottle. So remember that we had 518 in here, but we see 524 average in the flow meter when you do the calculation. So where is the rest going? That's why we call this spillage, right? Spillage is also something you want to reduce because this is again product going down to the drain. This can be a little spikes usually because when you're filling a line usually what we do what we call flush to drain. So you're filling the line all the way down to the drain. So you will see a spike in the beginning and then the actual spillage will be much lower than that. But average over time, average a shift or a peel if you produce two water, you'll see, let's say, six milliliters. Now we have 16 milliliters per bottle of waste. Now, just doing a very quick math in here, right? So I know for in the... The grocery store, at the grocery store, we're talking about maybe $1 per liter, 1.2, but let's say that this cost me 70 cents. And then in this specific site, I have six production lines, each filler, high-speed filling, 1200 bottles per minute or somewhere in there. So let's say at 80 % OEE, 24 hours run on almost $8 million bottles a day. It's $87,000 every single day. I'm not okay with burning $87,000 every day. Are you? So pretty much that's... Right. Yeah, yeah. Well, what if it's burning just wasting, right? Investing is...
speaker-0: Depends on what, but... different. ⁓
speaker-2: Right, exactly. Well, I don't spend elsewhere with technology, right? Well, but that's not all. So also you save it in logistics. So a good project that we're doing, we call it the perfect pallet. So it is if we reduce that volume, can we fill, can we fit an extra pallet in the truck? Now we are also saving in logistics, right? So now we can, you know, if we're going to do bottles, ⁓ What's our waste and trying to reduce that waste because now we have visibility, right? Because now we can actually measure that. Any questions, comments so far on this one?
speaker-1: It's all making sense to me, I'm liking it. ⁓
speaker-0: in the chat so far but this is really good.
speaker-2: All right. Well, that was the main one that I wanted to show as far as, you know, presentation visual, but there's a lot more, right? So when I go for commissioning, especially a new bottle, what always blows people mind is that about 60 % of the time we spend on CIP, which is always, people are not usually ready for this, you know, they're than half of the time. Yeah. So let's
speaker-0: Half the time. product changeover or why?
speaker-2: Everything. So let's say I'm setting up a new production line in here, right? So you're talking about receiving raw materials, receiving, mixing, batching. We spent more than half of the time commissioning CIP. I'm talking about commissioning time. Producing product is easy. It's usually pretty straightforward, right? Once you did all your IO test and everything, and then you're going to production, right? But CIP is where usually we spend most of the time. And I work a lot for dairy. So dairy is also, you know, you can't have milk in a vet for more than eight hours, for example. Right. So FDA has a very particular rule there that's called the PMO, pasteurized milk ordinance, that you have HTSTs, which is high temperature, short time, so this pasteurizes and all that. So you have to be constantly cleaning. So you're always cleaning equipment. So funny enough, Zach, In one of these, customer wanted to do, they had two CIP skids. They wanted to put a third one. We put a CIP report in there. What we found was that CIP skid two was being utilized at 100 % and then production was getting late because we were waiting for the CIP skid to be available. Right? But what we found was that CIP skid number one was available at 40 was being utilized at 40%. So now, but by just doing a few mechanical changes on what we call the CIP routes, now we balanced these two and they didn't have to spend millions on another CIP skid. All they needed was data visibility to make a decision. And instead of buying another CIP skid, all they did was rerouting CIP, right? Cool one, OEE. This is a little different on food and beverage than usually when you're doing machine centers, right? Because these guys usually we want to see, we want to visualize OE by shift, right? Or by week or by PO. Most of the times it's by PO. These guys, the big guys usually get a production order from Walmart, from Target, and they pass it down from ERP to the MES. They produce to order. So I want to know. what's my efficiency on that PO, right? ⁓ So this is why I think when it gets tricky when you try to put OE into MQTT is that that's the OE now, but is that an average for the day? Is that an average for the shift? And I think these are the things that I like to specify when I'm building a dashboard. Does it make sense? Okay. You guys can stop me if I'm talking.
speaker-1: Now this is all good.
speaker-2: curve as well. So now I have the machine, what's it rated for, right? So what is the machine speed setting? So am I pushing all the machines to its max to what was designed? Funny enough, we put this into a customer and we see that every night the machines slow down and right before the next shift starts, the machine, the speed goes up. We found this because one of the operators don't like to clear gems as much. So if he runs the machine slower, it's not going to have as much gems. So he doesn't have to work as much. Not here to cut our tail on anybody, but it's just one of the things that say, hey, we need to have visibility on what's the machine speed has been running and we need to set triggers off, you know, if it's not running up to speed and therefore also.
speaker-0: Another question to ask in that scenario is, that a worthwhile trade off? And you can't answer that question without the data.
speaker-1: And that's I was just going to say too, because the data would also if you have the right data, sometimes the operators running it slower because it actually means it runs all day. So sometimes lowering the machine speed can actually increase the productivity over time.
speaker-0: So the standard rate then for the OEE calculation?
speaker-1: Yeah, typically that's going to depend on how they're calculating OE though. And that's a whole rather whole other topic.
speaker-0: podcast.
speaker-2: Yeah, yeah, but the speed setting, what it should be running for sure, we would lower that, you know what I mean?
speaker-0: Or run it at the maximum that's recommended.
speaker-1: mean, OEE set points going to have how many things are supposed to come off of that machine in a given duration, more so than an actual speed setting. So sometimes lowering the speed will increase the uptime and actually get you closer to that target count part, count number as well. So when you have the rest of the data around you can.
speaker-2: The rabbit hole you're talking about is that what I have in here is bottleneck, right? If that machine is a bottleneck, then you have a problem, right? Because now you're having to decrease for the whole line. But if that machine is not the bottleneck, that's a good thing to do. If the machine is a bottleneck, I would want to do a top three reasons or top five reasons why that machine is stopping and get together with maintenance and say, hey, let's tackle these big monsters. Let's see. Why is this machine stopping? And let's try to push it, right?
speaker-1: Yeah, the real rabbit hole I was talking about, that's a big one too, is if and how and why we're calculating OWE. Some places are going to calculate OWE to try to identify the bottlenecks and find more information so they can go drill down and do their root cause and go deeper. Other people are calculating OWE so that they can have a fancy number to show people. And what set points you put into your OWE calculation and how valuable OWE is, is dependent on how and why you're using it.
speaker-2: That's even that stuff. Usually times you have a we have the true OE and you have the OE to please management. that don't think I got there today.
speaker-1: If you're trying to get real OE, you're not going to adjust the set point in the calculation. If you're trying to get a P's management OE and just have a number that doesn't mean anything but makes people happy, then you might adjust the set point down to make it look.
speaker-0: It's also safe to say, manually calculated OEE is never real OEE. ⁓
speaker-2: So, I think I want to talk about the quality here. This is two use cases that we did as well. So this label one, it's very interesting. ⁓ this customer was expanding, they installed two lines. So they took part of their warehouse to build two lines in there. So now it's a non-temperature controlled room and they're getting more rejects on label. They are able to figure it out that is condensation, right? ⁓ Then let's put a fan on it and let's run 24 seven that fan and get rid of the condensation or ⁓ so that worked. And then while we came in with data is installing ⁓ relative humidity and looking at the ambient temperature and also an API to the weather. And so now you can correlate the reject number according to ⁓ the environment. So I think this is when really unifying data comes in, because now it's not just the machine data, but also the mapping where the machine is, which area it is, is it a temperature control room and all that. And the same thing with these trade packers. We were trying to figure it out. were engineering was there, say, hey, we did this. We have no idea. Sometimes the trade packer, we have a lot of gems. We cannot figure it out. I go talk to an operator and an old lady, very, very cool person. And then she said, I know what it is. I've told them. I said, okay, let me hear. Same thing. So whenever they get these tray packers from outside, they store in a room and they get warped. So she said, if these trays are there for more than a week, I know that I'm going to have problems. So, and they don't run the tray packer all the time. So by simply managing when they get these trays and where they store, now they have a better quality rate out of the tray packer just because ⁓ it was getting warped in storage.
speaker-1: Yep. We got a good question here in chat. I'm going to jump in on is CPG has been asked to follow the same quality and production standards as the pharma industry. How can we show that implementing a UNS can help achieve this level of quality? And my immediate answer to that is the keyword there is going to be help, but implementing a UNS in and of itself will not affect your quality one way or the other. The UNS is one of many ways that you can make your data visible, connected, and be able to provide information. And that information is something that you or other automated systems can act on. And that is what will affect your quality, ideally for the better. But if you're going from traditional DCS systems and siloed data systems in general, you're going to have, you usually don't have very good visibility into your data. You might have an old historian of one of a dozen different names. that has its own charting tool and maybe has some API stuff that a couple people know how to use if you're paying for it or might have some CLI tools if you're paying for those. But usually it's hard to utilize the data in those historians. And if you go from something like that to either augmenting it in various ways or connecting it through something like that provides you to, that allows you to use that data to create actual information with actual context that can then be acted on by various. things, whether it's people or processes or other technology that automates that. But Tercato, curious your take on that too.
speaker-2: That's really good. I'm thinking about, because I never did a UNS for Pharma, I've done a lot of work for Pharma with the 2021 CFR Part 11. Data tampering. So I think we can still be compliant. That's a good question to you, Dylan. What's your take on it? My question back to you, is ⁓ using all that to be compliant with data tampering, I guess. data transformation in UNS. ⁓ I would be curious to work. I haven't done so.
speaker-1: Yeah, I don't have much pharmacy experience specifically, but in general with food and beverage or got a lot of egg with a lot of various FDA style requirements. not just a UNS, but having a robust data background, having a robust data architecture and infrastructure and a plan for what data is where, how it interacts together. Maybe not eliminating, but reducing the point to point connections and then having the systems on place for full audit ability of every change. If somebody is going to tamper with the data, either don't allow it to happen or at least track that it did. Some industries and some places will need to go back and modify batch records and others are mandated that they can't, but in either way, if somebody is able to, and they do, need to know that it happened. Who did it from what station, from what client, from what project and what login and what time and what day and why. And being able to have that traceability and then connecting the data. in a more cohesive way allows for much better genealogy and traceability as well, not only of the product or the processes, but also of the data itself. And by having that.
speaker-2: Yep. Sorry, go ahead. I was going to talk about something. I was going to talk about the dairy industry specifically on that PMO that I was talking about. what we have to have in there is your temperature transmitter in there needs to be sealed by the instinct specter. That goes into a chart recorder, which if you go to dairies now, they do have the circular chart recorder where they have to change that paper, right? That's very common.
speaker-1: Yeah, yeah, headblown here. Go for it.
speaker-2: To this day, I know that Anderson and some of these other ones, they are coming up with new ones. ⁓ And now I think it's just the market adjusting to trusting those. You know what I mean? But even VFDs. So we have a legal box on the spasterizers, for example. We have something that's called the legal box. It's a PLC that a enclosure is sealed by the state inspector. Even VFD drivers, we do have to take that HMI out of it. We need to have a state inspector sealing there too. It needs to be hardwired. Cannot be over Ethernet IP. You need to do hardwire and put set points in there. It's crazy, you know?
speaker-1: Definitely. think that's a good segue into your next bullet point there with traceability.
speaker-2: Yeah, so on that one, think I have a couple. The ones that I can talk about that I did recently was some preforms, right? So on this one, for example, on this Enterprise B, we don't have right now, which then I would assume that the injection molding machines, they just come preforms, which is, if you're all familiar, is they get the resin, right? The liquid resin they put into those injection moldings, and then they get what's called the preform, which is a... tiny bottle before it blows. So now where are you getting this from? Which batch is that? That is something that usually is not tracked. And then now you get more waste, more reject down the line and usually we don't know where this came from. An interesting one is with caps. So sometimes we just had a very high cap rejection on that. bottle in there you see was crooked or it didn't fit the bottle quite well. So once we start tracking that, we saw that the B number on the lot is one. It was when we had higher rejects asking the supplier. So yes, this is, this is my machine number B, you know, so I have A, B and C. So now you stop immediately receiving any caps from that supplier that comes with the lot B, right? And say, Hey, either You stop until you can prove to me that you did your center lining and everything is good and things like that, you know.
speaker-1: Before moving on to the next slide here, then I did want to hit one more question from chat on LinkedIn. Ryan asks, generally, would you give people the option to select a timeframe for OBE reports? And my quick answer to that is generally speaking, I very well lean into dashboards over reports for most things that aren't regulatory or just somebody's personal preference. And with any dashboard of any KPI or any aggregate, I would always give them various time slices as well as custom. selected timeframes as well. very, I always lean towards interactive dashboards where you can either go back into history, view the live one and just kind of make a seamless experience. And then sometimes it's even better than a real report just to be able to take a screenshot and print that to save it as a JPEG or, and sometimes for regulation, you do need actual PDFs.
speaker-0: By the way, this, by the way matters how you collect and store and calculate OEE. If you're not calculating the raw, like if you're not storing the raw events and the machine state transitions and the counts, you can't, you can't slice OEE and dice it however you want. If you just have like, all right, a real time, like you have a PLC tag that says OEE is 80 % and you're just historizing that. You'll never be able to go back and, and re-slice that data. So. And that's our our MES bootcamp fans out there
speaker-1: part of the That's a big part of the traceability, not only in the genealogy of not only the process and the products, but also, as we said earlier, the data itself. So as long as you're collecting all that raw data, all the raw values, you can recalculate whatever you want on whatever timeframes you want. And even more so with various tools, especially things like flow software and stuff, you can come back and even if I wasn't calculating this KPI six months ago, I've had that data. So now I can go back and back calculate KPIs I didn't know I needed.
speaker-2: tricky part to slice OE, I think it is scheduling. Because otherwise you're slicing teap. You know what mean?
speaker-1: But if you have the schedule, historized as what it was in that moment, as well as all the actual data points, then you can say, what was it for this half hour block, this hour block from this 15 minute block and as long as you have your.
speaker-0: want to see OEE by shift, OEE by operator, OEE by production run or OEE by product. Like those are all the things that I would want to see.
speaker-1: Yep. Yeah, and this doesn't preclude precalculating certain things for various reports and stuff as well. But when you have the real data, then you can make it more accessible for people who want to drill down and play with it and slide things around and really get into the weeds.
speaker-0: Use the gap, basically. Right.
speaker-2: Very good. Cool.
speaker-1: Good drop on the song there. Little teaser. All right.
speaker-2: If I move to the third one here just in a time check, think we're... So now not just to beverage. So now I'm getting a little bit out of just the beverage and bottling, but most of manufacturing, especially when you have a production line in CPG, the highest ROIs that I see, it's usually on utilities. And I did want to talk a little bit about this.
speaker-1: ⁓ we're good. We got time.
speaker-2: What I call the top energy consumers, right? I like going straight to when you talk about energy a lot of people go straight to electricity But just remember that steam, compressed air, other things are also energy, right? Gas. Yeah. Right. And now this is something that I failed in the past and if I had if I was adopting a unified namespace I know for sure this could have been a lot better, which is
speaker-1: steam compressed air, gas.
speaker-2: For example, this one, I'm trying to correlate how many gigajoules per ton of product. This was a good metric, but energy, have a lot of power monitors, 500s, GE, multi-link and all that into one silo. And then we had production data in another silo, right? And it was very difficult to integrate these two. So we had to use OLEDB and going to iFix Historian using a SQL bridge and doing all this kind of spaghetti because we already have point-to-point connections and after a lot of extra programming I got it but at that time I didn't know about unified namespace and especially how we could solve this problem using MQTT broker that would have been a dream you know
speaker-1: And I do one clear, one piece there too, for some people who might watch later or during the MQTT broker and the unified namespace is going to do a really good job of your overall values for various energy readings. You're still going to need a lot more of the more traditional, more complex, often vendor driven solutions for some of the full scalar data, the full spectrum stuff that for real deep, deep analytics, but you can solve so many use cases with just the overalls from very simple measurement tools that really work well with.
speaker-2: the UNS. Very good. So I think electricity is what I call the primary because you kind of derive compressor from electricity, right? So your compressors run out of electricity, right? Again, think gigajoules per ton is a good metric that I see quite a bit. Back to Zach's point, Zach was saying, hey, I want to see this OE by product, by skill number. That is also a good one. So now I'm saying, hey, When I'm producing a half liter versus a full liter bottle, which one takes more, you know, is it linear, right?
speaker-1: Yep. And actually I want to jump in here too, because this is a very, this is something that a lot of people think is really hard to do, but a lot of facilities, especially if they're already running on BFDs for a lot of their pumps and their motors and their compressors and things like that. This is all data that's usually really easy to get out of existing BFDs in various forms.
speaker-2: As long as they are in the network, right? Right.
speaker-1: Right. But it doesn't usually take, it doesn't always take putting massive, very expensive certified meters on to solve the use cases we're talking about here. There's a lot of things that the big certified meters can do that these can't. And then for things that aren't connected, there's always things like the Opto EMU and other power monitoring solutions as well that are easy just to, you put your clamps around your wires and move on from there. It's a much easier use case than people usually give this credit for.
speaker-2: Yeah, less invasive. On this one here for example when you talk about like sleep mode for conveyors, right? So if the if you don't detect that is the products not coming down But that on sleep mode just like when you go to an event or to pretty sure that Prove it last year when you're not on the escalator the escalator is lows down, right? So now I don't want to get too deep into it, but I've seen the same customer installing new machines without VFDs in the network. So now it becomes a strategy and this alignment. And they don't have VFDs on the network, right? Because it's cheaper, engineering is doing that. And then digital comes in later and decide to add network, which would have been much cheaper if you would have done it from the beginning, right? And then they do the same thing.
speaker-1: It's not even just VFDs either. mean, a lot of your smart overloads or your motor starters can do this as well.
speaker-0: Can a soft start do regen though? This is what blew my mind with water, wastewater is they'd pump, you know, millions of gallons of water up to 100 feet, 200 feet. Like they're storing energy and they're not capturing any of it on the way back down. I'm like, dude, you guys, what the hell?
speaker-1: It depends on the soft start.
speaker-2: That's right.
speaker-1: But a lot of places don't even know that, let alone how to get to it or what the data is. mean, if you know that it's running, you can do a lot by just seeing that data.
speaker-0: It's literally a battery. Like, what are you guys thinking?
speaker-2: Which that one here, if I go a little bit down on this one, this is one that I love, Zach, know, talking about reusing energy, right? If you have stored energy. In this case, you're talking about water as energy. This one for cooling tower is bottling, right? Especially on carbonation. So we have these things that we call the combi lines, which is the blow motor combined with a filler. And then you blow the bottles. And then right after, you need to cool it down. but to cool this bottle down or to wash, I think on the Enterprise B we have washers there as well. So you don't want to wash that bottle with high conductivity water from tap water because it's going to leave a film. So usually you have low conductivity, you already passed through RO or reverse osmosis, and then you're washing bottles at 40 GPM, for example, right? That nice clean water is going straight down to the drain. And then at the same time, you're running cooling tower, which is basically a big pot of water that you could use the low conductivity water to go back there. You know what I mean?
speaker-0: put a heat exchanger too on it like a heat pump or something to capture further energy too.
speaker-2: Yeah, yeah. So we do a lot of that. So once I start learning about Regen, because we do these hot water sets and pasteurization, this is very cool. So you heat it up. So you have a plate heater exchanger, for example. One section is heat, the other one is cool, the other one is Regen, right? Because you can cool down... preheating with the preheated water that already passed it through, you know what I mean? So you can, it's called a regen section of the plate heater exchanger, which I think it's always fascinated because you have like 20 % is heat, know, 30 % cool. And then you have this most of a plate heater exchangers is regen. All right. I think we have time here to go through a few more. I think a comp, sorry. All right.
speaker-1: Plenty of time. Plenty of time.
speaker-2: just wanted to do a time check again, because this is experience that I have as use cases and we can talk and come up with these use cases together and help the customers this way, right? Supply and demand in compressors. So what I see a lot is that I get to a place, compressors are running on manual mode, full throttle, and maybe today just half of the lines are working, right? That's... That can be, you can manage that supply and demand with all the compressors in auto and then you do a good sequence startup and shutdown sequence in there. Only if it was that easy because, and then you learn that with that utility guy, the factory never stops because he knows that two more lines are coming up and he anticipates, he goes there and then he start one or two other compressors before it's time and destroy it. This tribal knowledge, he never transfers to anybody else or let engineering know. And then the day that he takes vacation or he retires, the factory doesn't run as well.
speaker-1: That's again why it's important to be logging out this so at least you know whether or not you know, know, at least you know.
speaker-2: That's right. A good point because if you're logging this data, even if he didn't tell anybody, you could look back at this data and say, wow, you know, it was doing better because these two compressors were coming up in manual right before the...
speaker-1: I mean, if he retires and a month later it's not running well, you do your root cause analysis, why is this machine running worse than it used to, follow down the events and you'll probably get to the fact that he retired and that date is when things started to change, but then you can keep going and in the process of that you'll have seen all the different data points that you use to get there that you can then use to recreate what was done without needing to try to call him back in or try to pay him as a contractor or something to get him back in to teach others. So best case scenario is still having people cross-trained and obviously the culture is the much more important piece here, but even in absence of that, this gives you a little bit of a safety net.
speaker-2: Yeah. And I think you were going to agree with me where I like when I'm doing a project, engineering is leading. I like going down and talking to the operators. Right.
speaker-1: I can't, I know it's common, but I still can't wrap my head around people who don't do that. mean, it's...
speaker-2: Sometimes they kind of don't let you, you know, they say no, no, I can handle everything and then I have to be a little pushy, know, say hey, can I during commissioning, can I stay there with the operators and just hang around? I just, I just want to be there. Yeah. And then I learned a lot.
speaker-1: Yeah, it's more of a requirement on our side, but yeah, it's if I can't talk to the operators, if I can't train them, if I can't, I mean, I'm the unusual, even like when I was doing like raw controls engineering and PLC code and stuff, if I was down there on the floor, I preferred when the operators or the other engineers and maintenance staff was over my shoulder looking at the code and asking questions and why does this work or why does that work or why'd you do this or be able to explain to people kind of how things are happening or why are we making this screen and
speaker-2: Nice.
speaker-1: I mean, without that, you're never going to get the user experience that you would otherwise want. You're not going to get the training. You get a lot more service calls after the projects are done and people are generally just not as happy. mean, even simple things like watching what buttons they press on the HMI and where they're positioned on the screen and just making little subtle changes based on what you see people doing or not doing. The ergonomics are very important. And if you're not watching and you're not talking to, and you're not at least being friendly with and working with the operators. you're never going to know the requirements for that.
speaker-2: Yeah. Yeah. A good one that I had there was ⁓ we had to start a hot water set when we start a pasteurizer, right? And then we were doing commissioning pasteurizer number two. And then I'm there on commissioning. The operator says, well, I'm going to start the hot water set. And I was like, why? know, according to engineering documentation, the way I programmed is going to start automatically. But he says, I just heard the hammering from the receiving. So I know they're gonna call me up because the receiving tank is full. So I'm just getting ahead of the game. I have to spend 15 minutes.
speaker-1: ⁓ yeah, they know what they do. They know why they do it. And as I said before, the data can be a good safety net, but it's not a replacement. The data will tell you what they did, but it won't tell you why. And sometimes you can kind of find why by looking through it. But usually what you can find in eight hours of looking through the data, doing a bunch of machine learning and fancy algorithms, you could get in five minutes of a conversation with the operator who actually made that decision. I mean, it's...
speaker-2: You can actually make changes and get value, which in this case was the trigger now for the hot water set is no longer the HTSC start, but is when the receiving is done. So, receive shut down, the tank is over 80%. I don't remember, it was been a few years. Now, it's consistent. Now everybody does the same. There is no operator that goes there and do something manually, saving a lot of time because they receive that maybe three or four times a day.
speaker-1: There are things that you just can't find through the data too, that no matter how much you want to, you're never going to be able to track. mean, there was one I went to and it was all good with the product they're running. wasn't any problems with recipes or safety or anything like that. But you could tell that this, was a big block press and it was running better sometimes than others. And you could kind of figure out it was this operator or that operator. can kind of, it would work better at different times of day, what correlated to which operator was on staff. And nobody really knew why it was doing this or why it wasn't doing that. And when you walked out to talk to the operators, they would explain when this sensor hits this, it does this, it sometimes gets stuck there, but it doesn't do that. And then go down that rabbit hole that isn't always the best rabbit hole to go down because everybody's going down this big technical rabbit hole of sensors and actuators and timing and all this stuff that they're changing with the, whether it was the hydraulic press or timers and such in the PLC code or adjusting sensors. But the real difference of why it was working better or not was they had a can of cooking spray that they would use to spray down the chute. so some operators used it more than others. And you would never know to track that. You would never be able to track. But when you're there talking to them or even just watching them, you can see they're picking up this can of cooking spray. No, well, Pam will spray in the metal chute under the block press and it would get stuck less.
speaker-2: Right. Now.
speaker-1: And you're never going to get to that Roos Kaz without actually talking to the operators, watching them work, working with them and learning the machines as they know them.
speaker-2: nowadays putting a vision system there with AI and spy on them.
speaker-1: They love that.
speaker-0: You know what want to do?
speaker-2: Yeah
speaker-0: I want to buy an Opto 22 EMU, electrical monitoring unit. want to install it for free at a manufacturer nearby, Salt Lake City. me know if you're spending more than $10,000 a month on electricity or something. And then just collect the data and then let an MCP tool and AI recommend changes for improvements. And then at the end of the year, compare energy savings and then have 20 or 30 % of that energy savings. ⁓ as payment,
speaker-2: Now you're putting, yeah, that's how much you believe.
speaker-0: The ice tarmozy, like the offer is so good you can't refuse it. Why would you not?
speaker-2: Right, right. So you're the risk. That's an entrepreneur there taking the risk. Say, hey, I believe in this, you know, and I'm willing to take the risk and getting part of the reward, huh? I like it.
speaker-0: What are you guys most excited about for Prove It? Let's finish the- Let's end with-
speaker-2: Besides medieval times? Yeah, okay. Yeah, I have a very, very, very few here that I wanted to talk about. And again, these are all real things that I've done, right? So high pressure bleeding into low pressure. So now you're creating energy to waste energy, right? So sometimes they have the compressor for high and a compressor for low is so much more efficient to actually run a low pressure compressor than losing in here. Shut off in idle time. This is a good one. because they wanted to have the blow motor manufacturer accountable. And they were saying, I don't need a shut off valve in there. And we're saying, hey, if we put a shut off valve in there when the machine is in idle, and then you don't have more air leaks. But while you're fighting this battle, I think it's such a low hanging fruit, putting a shut off valve in machines that don't run very well. Again, and others, what I always say is about a lead and lag indicators on everything. I think here's when Unified Namespace, you're talking about real time, real time decisions, real time dashboards. And do you want somebody to let you know how bad you did yesterday or how bad you did last month? Or do you want to see where you are now and what lever can you pull to make improvements? Right. And on this one here specific is a ⁓ cooling tower, right? Maybe for next stream, I have a cool presentation about how to manage a cooling tower. And a big mistake that I see people doing is installing power monitors in there and trying to measure cooling tower efficiency by how much energy it spends. And it's not the metric you should be looking for. It's what's called cycles of concentration is how many blowdowns you have is is the conductivity you're accumulating. So you have specific metrics and you most likely have that data because if you have a blow down valve, it's probably doing blow down by conductivity and all that kind of stuff, all this kind of data you already have, probably you just don't have the metrics, you know, so. I that's all from me. That guy right there is me and I say, don't reinvent the wheel. Talk to an expert, get real value. This is what I'm here for, especially with Enterprise B, which is to what problems you're trying to solve. Let me try to solve the problems you're trying to solve and make you aware of the most common problems in food and beverage that you may not be aware now and you might encounter in the future.
speaker-0: This was like a master class on food in bed. I learned a lot. Thank you, Ricardo.
speaker-1: ⁓ And that's of the many reasons we brought you in.
speaker-2: Good, that's good to know. Nice, nice. Yeah, I tried to cram all into one and make that very first presentation, but usually I have a few more of those when I talk to someone.
speaker-0: Ricardo and I, I followed you a long time. You know, we followed each other, but Prove It 2025 was the first time I actually met you in person. So for me, Prove It is the place where the online Industry 4.0 community can finally connect offline. There's no other event that has that density of experts in manufacturing, Industry 4, IIOT. All of the influencers are there. So for me, that's really exciting. And then I'm also excited about The the casino night that's gonna be pretty snazzy. They say come dress to impress and then um, so I'm gonna have to go shopping Yeah, I don't know just oh Fuse the gap obviously. Oh, yeah
speaker-1: debut. think my biggest thing is just like you, mean, prove it 25 is where I first met Ricardo and some members of his team. And then we've kept in touch ever since. And even back then, we were kind of half joking, but hey, let's do prove it. Let's do prove it. And now to a year later, be here doing it. And what started off as just a yeah, maybe maybe we'll sponsor maybe we'll do it. Let's talk about it. And eventually here we are doing it and proving it in 26. So a sponsor on the stage, big audience, it's going to be a lot but very much most looking forward to all the people there all the different vendors all the different people I know all the end users all the and more importantly just like last year with Ricardo the people I don't yet know so all the connections all the networking all the people there is kind of what makes it what it is for me.
speaker-2: Right. Yeah. For me, the same thing, you know, first time I met you guys, a lot of people from the community, a lot of the influencers I saw, Jeff Winter in there. So that was, that was really cool to see everybody to shake hands of Arlen Nipper, which is like, all right, you want to know who was the co-inventor of MQTT and meet the guy? ⁓ You know, you're going to be right there. And really learning all about the tools that I wanted to work with. And I never had a chance. I feel like coming from a And this is my message that if you are what we call a ⁓ traditional system integrator and you do PLC and HMI, you step into MES databases and you want to get into industry four, this is the place to be to learn about all of this. Right? So this is where you see new tools. You're going to see FUSE, you're going to see Hi-Bite, you're going to see Flow Software and how all of this integrates. Right?
speaker-1: And that's actually Zach, if you could scroll back up a little bit, our stack includes Ignition on top there and Litmus, as well as if you go down to the gold, our stack of our solutions includes Flow, Fuse, HiveMQ, MaintainX and probably Google Cloud. And we'll have different use cases being solved by different tools. But then you've also got all these amazing sponsors down there too on the silver and the bronze side. A lot more integrators this year than there was. A lot more integrators than there was last year. So I'm really excited to see what everybody comes up with. how people are using the same tools in different ways, how things are connecting, the different creativity too. mean, even last year, but even more this year with more vendors, the different ways that people have solved different problems, even what problems people identify in the first place, there's a lot of creativity in the room. ⁓
speaker-2: exchange data with each other, I think we can, right? It is live, right? If you want to do, if you want to do Sesame profile, maybe I'm saying too much in here, but you can check Abelara's topic in there and you can see it firsthand. It just came out yesterday, right? Jonathan Wise from... Yeah, sweet.
speaker-1: So that's the matter of fact.
speaker-0: touching Sesame profiles. So like we're, pushing, we're pushing the, boundary of manufacturing as we speak at Prove It. That's so like, we're pushing the envelope. But also the other thing I wanted to mention just because it hasn't been promoted that much is by getting a conference pass ticket to Prove It 2026, you're getting a free, every single attendee is getting access to Medieval Times night. So they're going to bus everyone to Medieval Times. They've rented out the whole facility. So that's going to be an unforgettable experience, you know, as well as the casino night. I mean, they're really going above and beyond.
speaker-1: back down, that's kind of the new thing for me this year. I loved it when ICC put this into place a couple years back. you scroll down one more, you've got the table talks. Two, I think it was two years ago, maybe three, but two years ago when Ignition put this into ICC. And it was one of my favorite moments of ICC both years now that they've had it have been these table talks, maybe three years now, I think. And the different perspectives, the different conversations, everybody being on a level footing, everybody from People who are brand new to some of the tools of the 4.0 space right next to some experts who have been doing it for 20, 30, 40 years. And everybody's just collaborating and trying to make everybody better and learning from each other, regardless of who they are, where they come from. And the table talks are a really good way to meet people and have some of your problems solved and ask the questions and really learn. having a thing to note there too is end user meaning. You work for a manufacturer, work for a factory, you work for a process facility, you work for the company that's making something. End users are not software vendors, end users are not integrators. They are the people who work for the factories that are actually making.
speaker-2: manufacturing and making the decision, the decision makers, right? I'm just thinking here, me, I was working for a manufacturer, like I said, no problem saying names here, Nestle, right? Huge global, and I'm looking for solutions. If ProveIt existed, that would be the place for me to go out there and look at different vendors and pick what I want, right? And I'm going to learn about Stack. I get a lot of people asking me, hey, what's this unified namespace? Do you have some material? Maybe we're going to have some from Abelara that Zach was reading live, but also what's data ops. People are asking me what's industrial data ops. Go over there, look, right? You're going to understand about your problems will no longer be solved with a monolithic one product that solves it all. But you understand this stack and how you can connect and pick and choose what you want. And everything talks to the single source of truth. When you learn, you go there and you pick your vendors. And now you have more integrators. So you pick the software you want to pick and then also you pick the integration you want to do integration with the integrators. Also you pick people to do your consulting, your roadmap, your blueprint, your architecture. Abelara. So again, if you're an end user, you are in a factory, this is the place to go.
speaker-1: And there are, as we said, a lot of different integrators, lot of different consultants, a lot of different software vendors, a lot of overlap. And you can really get the understanding of who knows what, who's going to be a good fit for you there. There's a lot of very good people there, a lot of really good companies. And this is going to allow you the chance to get in, talk to them, meet the people. A lot of these places even have their founders and their executives there all the way down to their engineers and their programmers. And really getting to see who aligns with you, who knows your industry, who vibes with you, and just kind of what's going to work for you.
speaker-2: You know what I like? Because if I'm a technical person, go to a conference and I want to say, how does this piece of software works? And then they say, well, let me call the engineer. Let me take your number and I will talk to the development. When you go to prove it, you want to see how something works? The developer will be there.
speaker-1: If you want to see how does this work, they're usually going to have their programmer, their integrator is going to have their engineers. They're going to have the people there who wrote the code, who did the thing to make that feature happen. But also a lot of the vendors are also going to have their product. mean, there's going to be some people, a lot of the vendors are going to have their product teams there, their CPO, their CTO, their programmers. And you'll not only be able to see the sales pitch of why would you want to buy it, but also how does it work? Why does it work? Why did they architect their code in this way? You can get really deep, but also on the product side, why did they build this feature? Why haven't they done that yet? And you can have these conversations and because of the people who are there, if you have, if you're somebody who uses their software all the time, it's probably a really good place to go and ask those questions of why, why is this not do this or why is this thing not fixed yet? Or why is that not there? And sorry to all the software vendors who are probably gonna have to field these questions now, but it's a good place for it.
speaker-0: There's a lot of people talking about fuse now and people are asking the question why choose fuse? Yeah, well they're good up on stage fuse is gonna be a gold sponsor presenting to the whole audience so you can learn all about them but a lot of people are coming just to see me Sing and rap on stage, know Z money's act driven. I want my true fans in the Standing on your feet. ⁓ let's see money. Let it be known bill 4.0 plays from the ground to the throne prove it conference
speaker-2: happening again?
speaker-1: Yep, since nobody watches this late in the video anyway, so it's a good place for a teaser, you should probably check out the Fuse Conference if you want to see more Z money.
speaker-0: stage.