Unlock the Power of Data-Driven Manufacturing: Digital Strategies to Accelerate Manufacturing Growth

Hello, everyone. Thank you for joining our webinar, Unlock the Power of Data Digital Strategies to Accelerate Manufacturing Growth, presented by Ronak Makwan and Hennaro Quiroz. Today, they'll be talking about how technologies like CMMS, predictive maintenance, SCA, IoT, remote monitoring, and machine learning can streamline your operations. The role of data and capital planning, asset life cycle management, and how reporting and analytic tools can help you achieve your financial goals with confidence. Before we get started, I wanna just go over some easy housekeeping items. All of your lines are muted during the session. Please submit your questions to the q and a feature at the bottom of your screen at any time during the webinar. We'll answer questions at the end of the presentation. If you don't see the ask a question box, you can open it with the question mark icon at the bottom center of your screen. The presentation is being recorded, and it will be available on demand after the webinar. We'll also send it via email with a link so that you can come back to it or share it with colleagues. It usually gets sent within twenty four to forty eight hours. I'd like to introduce our two speakers for today. With over a decade of hands on experience in both maintenance and engineering roles, Ronak brings a wealth of experience in GMP, HACCP, and BRC programs. Ronak's experience includes lead manufacturing and total quality quality management environments, maintenance management and fast paced environments, and a deep understanding of building automation, PLC systems, and equipment integration product projects. Driven by innovation and a passion for process improvement, Ronak is well versed in non value added elimination principles and has demonstrated successful project management skills in manufacturing plants spanning initiation, execution, monitoring, and controlling, and closing. Ronak is a professional engineer with his red seal journeyman electrician ticket and has a PMP certification. He's based in Toronto. Pinaro Kiraz is a solutions engineer at Stevens specializing in AI and predictive maintenance. He leverages AI to revolutionize traditional maintenance strategies, driving sustainability and innovation. With extensive experience as a controls engineer, industrial technician, and US Navy submarine electrician, Pinnaro understands the critical importance of reliability, efficiency, and continuous technological improvement. He has successfully deployed projects globally, making AI a reality that delivers tangible outcomes across various industries. Now I will pass it over to our two speakers to get started. Awesome. Thank you very much, Stephanie. That was a great, introduction about myself and Genero. So hello, everyone. Good afternoon to those who are joining us from the eastern starter zone, and good morning whoever is joining us from the Pacific time zone and the central time zone. So myself, Ronak Mequon, I come from the manufacturing industry working mostly in the food and bio. I work in maintenance operation as well as the engineering role. And today, I'm I'm thrilled to talk about all the different digital technology, what we have available in the current market that can help manufacturers run their day to day operations and maintenance. So before I start talking about all this technology, let me get into a visual representation what we call as a asset pyramid. So when you look at this asset pyramid, that's the entire visual representation of your entire asset management journey. So if you buy any asset, and I'm just gonna give you a very basic example. Let's say if you buy a brand new boiler or a brand new air compressor, you always start from your the very bottom layer, which is the work management processes. So work management processes, whether it's a part inventory management, whether it's a work order management, whether it's a material management, whether it's a asset registry to start building it from the scratch, those are some of the core fundamental items that every organization, every manufacturers, every maintenance department should and must follow if they want a robust asset strategic asset management program. So that would be the the most foundational item. Then when you look at the second bottom layer, have you envision your asset management journey? Now you have the work management. Now you have the asset registry. Now you have your all your parts in place. Now you start implementing processes to maintain your day to day operations. So what type of process that that would be? So a very basic that that that can go with any companies is RCA, root cause analysis, reliability team, different maintenance strategies such as PM compliance program, work design standards. So those are the processes that will help manufacturer to stay on top of your work order management to detect any reactive failure and act on it and basically work in a proactive mode rather than a reactive mode. So that would be the second bottom layer. After that, now we have started to going into digital mode. Now we are talking about the operational technology. First, you already have your work order management. Now you have implemented the personnel, the manual processes in place to look after those work management processes. Now let's get into the operational technology, and we'll be touching all of them in our next slides. For example, CMMS or IoT remote monitoring or predictive maintenance or FCA, building automation system, or SCADA, which is supervisory control and data acquisition that that links all of your moving existing production equipment into a single screen. So the idea here is you want to have a full vision of your manufacturing industry, of your production environment on a single screen. So if I'm a plant manager, if I'm a senior level executive, if I wanna understand, hey. What's happening as of today, April nine? How's my day shift doing? How what did they do in the next, last six hours? Those operational technology will provide them that information and data. Then we go get into the performance management, and that's where we're talking about the reporting and analytics, which we'll be touching in our next slide as well. But that's where now you already have the defined data available. Now you talk about strategy. Now you talk about how you can leverage that information, that data that can take you to the next level, which is the strategic asset management journey. That will help you and equip you to increase the overall equipment lifespan. And then we will end up into asset investment planning, which most of the manufacturer call capital planning. Right? But in a nutshell, it's it's a synonym for asset investment planning. How are you planning and managing your assets? So this entire pyramid, what we call in, this is very near and dear to my heart because that provides you a very broad level understanding of how you would like to manage your assets. If you are on a intermediate level, if you are just starting your asset management journey, or if you are on a well advanced level, that will tell you where you are and what would be the next step in your journey. So just want to start this conversation from this pyramid so everybody has a clear understanding of what would be an ideal asset management journey look like for any organization. Moving on. So everyone know everything starts from data. Right? So let me give you an example. And as I said, I come from manufacturing. Every morning in my previous work life, as soon as I go, the first question I was asked, hey. How did we do yesterday? What was the equipment downtime? What was the production throughput? What was the overall equipment effectiveness? Because that's what everybody wants to know and everybody is is keen to learn about the numbers, the projections, the performance. Right? So what is the center of everything? It's the data. But let me ask you everyone, and you can ask yourself, just providing you with the numbers and just providing you with the raw data, will that be helpful to make any judgment or any any strategic decision? No. Right? Because you need to convert data, and that's why I have created this slide, which explains exactly what is the difference between data and information. So as I said, data is just a raw information that is coming out from any software. So for an example, if you have a CMS implemented, full quality of raw information about let's say, previous day, you did fifty preventative work orders. But how you leverage that data? So in a in a in a layman term and a laywoman term, when you convert data into actionable insights, when you tie in actions against those numbers, that's when magic magic starts to happen. And that's where you start seeing different trends, different, areas of, what is trending and what is not trending. For example, let me give you an example. If you get a trend of polyoproduction equipment for last one week, now you're comparing that trend into information saying, hey. One of my production filler is constantly going down for at least one hour per shift. So now you have a trend. Now you have the data for last seven days. That equals to almost thirty some hours worth of downtime just on a one single piece of equipment. So now based on that information, now you can start thinking more proactively to go back and talk to your maintenance team, talk to your maintenance supervisor to understand what is the root cause, why we are losing every day one hour on a single piece of equipment. Right? So just wanted to give you a more basic example to understand that only, raw data cannot help. You need to come up with the trends. You need to attach that with your actionable insight, and you need to understand what is your end goal. If you have the data, how would you like the data to interpret to provide your end goal? Is your end goal to improve the equipment efficiency? Is your end goal to improve the quality product that you produce? Is your end goal is to reduce your maintenance cost? So based on that, you can leverage this information, leverage this data that can help you moving forward. Okay. So that based on that information, let's talk about CMS because I already mentioned CMS almost few times already in this session. So CMS, which stands for computerized maintenance management software. Most of you who are already here, they must have either heard this terminology or have used this terminology. There are multiple, I would say, number of different CMMS softwares that are available in the current market right now. There are some which are more basic. There are some which are intermediate. There are some which are well advanced. So that's where if I were you, I would first understand what is my use case. First, you need to ask yourself, is it you need to buy a similar because your maintenance cost is very high and you don't have any any data to support why your maintenance cost is going high. You don't have any information. That's one point. If you are increasing unplanned equipment downtime and you don't know how to fix it, that's another use case. Spare part inventory budget. Like, this was a very, very big issue for me in my previous work life. Spare part inventory management. Right? Because the inventory management is a very, very big people think it's a back burner, but it's not. According to me, I have seen almost close to one million dollar worth of write off every fiscal years because we are not tracking inventory accurately. Maintenance staff, maintenance track trades are not signing out motors, gearboxes, bearings that they use it. They physically use it on the piece of equipment, but they don't follow the process. So that is that your use case? Missed production goals, that goes hands in hands with the equipment efficiency. Your customer confidence. So when you are tracking your customer feel rate, is there your is your customer feel rate is going low? Is or if you are getting more customer complaints, that becomes your another use case. Right? And the last but but not the least, and this is the center of our today's topic, lack lack of data. If you don't have data, and I had I was in that situation in previous work life that I joined the company. The leader of the maintenance department, I was told to start leading the maintenance, but, hey. What? We don't have any historical data because we are using a legacy software we can which can only do minimum task, which does not have a capacity and capability to tell me what happened in this particular facility for past two years. So that can become my big use case. And based on that information, I went ahead and I I install and implement a brand new CMMS system. So how do you define a well designed CMS? Right? So when I look at any system and regardless of whether it's a CMMS, whether it's a capital planning system, there are three main pillars for those systems. Need to understand the technology. The technology, is it meeting your demands? Is it meeting your needs? Second is the people, and that includes management as well as the first hand operators, maintenance trades because they are all gonna use this. Right? So you need to have that understanding that people's skill in mind to see, hey. If I buy this particular software, will that be helpful for my my current workforce? Right? And the third piece is process that you need to understand that whatever the system and software you buy, will that be integrated into your ERP system? Most of the company use SAP, IBM, Maximo, Oracle, JD Edwards. So So there are different ERPs. There are different CMMS, but will they integrate together? And that is very, very common questions every manufacturers has it in mind. So these are some of the use cases and some of the benefits. If you have a well designed system, the first and foremost thing that one of the huge benefit for me in my previous work life was you increase your equipment life expectancy. You produce more quality products. You meet your production goals. So that that's the bread and butter for any manufacturing. You need to complete your production goals. If you are told to finish two million pounds worth, catch up or if you're told to do, two thousand cases of finished product on a weekly basis, that's your target. You need to meet that target. If you always achieve that, that's wonderful, but you need to meet that minimum target levels. You need to make sure that you are attaching and you're meeting all your regulatory compliance needs as well. And most important, you can track all your assets, all the inventory which is attached to your assets. So now that is also helpful from the financial standpoint. Right? So if most of you who have worked in the managerial role, you need to work with finance at every end of the fiscal year to go through the depreciation cycle, the equipment depreciation cycle to understand that, hey. If I have a five year old boiler with this equipment efficiency, how much dollar value it it has right now? So equipment efficiency, depreciation, everything goes hand in hand. So that is what a well designed CMMS can help achieve your goal. So again, you need to go back, rethink what is your use cases and how or which system will help you resolve those use cases. So before we're moving on to the next slide and next topic, which is FCA, I would like to ask everybody that did you guys hear this terminology called FCA, which stands for facility condition assessment framework? Or have you used it? How much are you from there? So you can say yes, no, or not sure. Oh, so I'm I'm seeing, North Shore and most of that. Hundred percent for North Shore. So that's interesting. So let's go and talk about FCA. So FCA, as I said, it stands for facilities condition assessment. Right? So if I give you a brief context about where it all originated, so FCA, facility condition assessment, it firstly originated in early nineteen seventeenth. Right? So I'm talking we are sitting in twenty five. So almost seventy years ago, this concept originated, and the early adopters in North America were universities and health care facilities. The reason behind there that they were getting into more and more complex engineering design within their facilities, their maintenance costs were going really high, and they wanted to have more structural approach when it comes to the capital planning. So that's where they started looking into this concept. And then as the year passed by in early nineteen eighties, it got affiliated. It has a registered benchmark. And in starting early two thousand, two thousand ten decade, now we got into the SaaS model way where we have the digital technology that can help you run this entire process and provide you with the final outcome. So how does it work? So let me explain you in a more basic term. FCA meaning is the company will come and it will tell you. So for an example, if you're running a food and well plant or if you're running an automotive plant and if you have five, I'll I'll talk about the utilities. If you have five boilers and you want to know out of those five boilers, what is their health? As of twenty twenty five, if you have a twenty year old boiler, if you have a ten year old boiler, regardless of their age, what is their life expectancy left? So that's where FCA comes into place. So in my next slide, I want to talk about the three core fundamentals. So how FCA works? First, it's it talks about the condition. So it tells you once the entire process is completed, once the entire inspection is completed, it will the four first condition first thing is they will tell you what is the current state of the condition. If your boiler is twenty years old, if you have already done a corrective refer for almost fifty times, if you have already replaced and rebuild the gearbox, if you already replaced the motor for five times, that will tell you what is the condition. Now it will go back, that condition, they will take it and they will compare it with the OEMs, so original equipment manufacturer and their specifications to tell you what is the expectancy life left after comparing with the OEM. The last, but what is one of the most important is the risk factor. So now let's say after comparing the current condition with the overall health, according to the OEM, now I get to know that, hey, Rona. Out of your five boilers, one of them is at running at eighty percent capacity, one is running at seventy percent, and one is running at only fifty percent. So now based on that, they will provide me a risk factor. So the final outcome I will get is what we call is a facility condition index, and that will be in a percentage score. So let's say if I'm told twenty percent percentage score for one of the boiler, that means that it might only last and stay operational for next one year. So based on that data, based on that most accurate data, I can feed in that information into my capital planning process. So how FCA can help manufacture? First, it can help you in your current day to day operation by telling you the current state of the condition. So that can help you with your CMMS system, day to day operations, allocating budget, monthly budget, allocating spare part in which it that's all fine and done. The second piece is now based on that risk factor, now you can support and and and leverage that information for your capital planning process. So FCA, for for if you ask me, it's it's a integral part of your asset management process because because it can fit you into your CMMS journey and it can feed you in your into your capital planning journey as well. K? So we did talk about FC and how it can help into the capital planning. So the next piece I would like to talk about the entire capital planning process on the what we call ALM, asset life cycle management. Most of you who have who are currently working in maintenance or project or engineering must be familiar with this, design what we call asset asset management life cycle. These are four different stages, planning, acquisition, operational maintenance, and disposal. Right? So data, technology, the historical and real time data, and that's where my colleague, Hanara, will talk about real time data as well. But both historical data and real time data information help you enormously when you are talking and when you are planning for your projects. I can give a really good example. So I was one of my company, I won't tell tell talk about the company name, but that company, I was installing a brand new cooking deck. And that cooking does deck was the one of my major customer, McDonald's. So we the the data and you can understand McDonald customer, they have a very rigorous data, very rigorous r and d and product development team who I was working very closely with. So on that information, I had to go back and forth working with the third party equipment manufacturers, providing them the r and d data, product development data that, hey. This is what I want. This is the texture I want. This is the thickness I want. This is the viscosity I want out of the product. Based on that information, we created and implemented and executed that entire project. So the moral of the story here is whether you are in your planning stage, acquisition phase, operation and maintenance stage, or when you're in at the disposal stage, every stage you require some sort of a data to justify that state. If you're talking operation and maintenance, as I just talked about in my previous slide, you need all your data, whether it's a historical data or real time data. When you want a disposal, you need your finance data, which I talked about the depreciation path. In my previous work life, I had to write off three forklift because we weren't using it for years because it was broken down and the cost to repair that, it wasn't justifiable because the amount of cost it had on my, finance sheet. So what can I do? The finance said, hey. You still have sixteen thousand dollar left for the forklift, but to repair and to use it is forty thousand dollars. So how do you justify? So that's where the data helps to make you a business is where you can take it back to your one of in this case, c level executive to explain them the scenario and come up with the final outcome. Right? So if you're doing any sort of a project, always make sure that you need to go with every stage of this, project stage with data availability. Okay? So the next is data integration into capital planning. So this is a very key note. I think I already touched base on that. But make sure you do the financial analysis. And this terminology and it become very, very popular in last minimum five years, what we call ROI, which stands for return on investment. Right? Before that, I was never asked to to showcase what is the return on investment for any project I did, but right past post COVID, there was a rigorous standard that most of the companies started implementing that talks about ROI. What is the return on investment? So if I'm investing if the company is investing, for example, hundred and fifty thousand dollar in buying a brand new cardboard compactor, what is the ROI? Right? How do you calculate ROI? ROI can be calculated in dollars. So let's say if you increase your production by five percent, that's a easy math. If you're increasing your efficiency, that's an ROI itself. If you're reducing your label, so the new carbon collector might need one person versus the old one which needed two percent. So now I'm elementing one labor. That's an easy math as well. So, again, you need to come up with some sort of an ROI calculation. Most of the company right now feel comfortable if they have less than three years worth of ROI. I have done most of my projects that has less than two years worth of an ROI attached. But just to keep that in mind, whenever you are doing some sort of capital planning process, do some sort of an ROI calculation, whether it's an efficiency, whether it's a time, whether it's a labor, and whether it's a production, enhanced and efficiency. Risk management, always do the risk management. That's where also you need data backup, and that's where you go back, leverage the historical data, leverage the FCA data. And according to that, you can prioritize. So let's say, for example, how you do the priority assessment. The FCA data is telling you you need two boilers in the next two years. One of them is only twenty percent, efficiency left, and one has a fifty percent efficiency left. So that's where you prioritize. The first year, you would replace the one, which is only twenty percent left, and the second year, you will replace the one which still has fifty percent of the life expectancy. So even for the priority assessment, as I said, all you need is a strong, accurate data to justify your business case. Right? Because without data, making guesswork will not take us anywhere else. And based on that, you can allocate your budget accordingly because now you have your priority set. Now you have already done your risk management. The next step is very easy. Allocate your budget based on the priority. Right? So everything goes step by step. This is all three step process which I had followed and which made me successful in all of those projects that I led. Okay. So I think I I I did talk a lot about historical data, CMMS, and capital planning in FCA. But now at this moment, I will pass it on to my colleague, Anaro, and he will take us from here. Hennaro? Hey. Thank you, Ronak. Hey. Good morning, good afternoon, depending on where you guys are located. So I'm gonna talk about predictive maintenance and about how Brightly and Sensei work together to give you that closed loop maintenance solution. Right? It's it's it's all about streamlining our data. As Ronak mentioned, data is cool. Information is better. So it's it's what you do with that with that data that you have because everybody here knows that you have pieces of equipment that have very important pieces of data within them. And, you know, with with the rise of ignition, with SCADA systems, MES, cake systems out there, you're starting to collect that data and you're starting to play with it. You're starting to formulate OEM OEM dashboards, but there's there's so much more potential there. And it's about utilizing that data from your assets, right, feeding them into AI, and actually using AI in a real world scenario, taking that real time data, getting predictions, getting forecasts on failures. And then the closed loop maintenance part is utilizing brightly work orders, utilizing that all that data that's stored from your maintenance people and filling in that that that gap. Right? We're experiencing a gap in the market now where in an all industrial, manufacturing operations, you're you're having issues with operators. You're having issues with your maintenance guys. Your your senior people are exiting the workforce, and and there's a gap. So it's it's about capturing that information, capturing the knowledge and expertise, and having it all feedback into a tool that you could exploit. Right? So that's that's what that's what actually utilizing you, AI in a real words world scenario means. So when you look at traditional maintenance events, you're talking about being notified of an issue when an operator, notices that there's, an abnormal noise or the machine stopped. You have a fault on the HMI screen, so a maintenance operator is called over. Right? Your maintenance engineers, manufacturing engineers, your line techs, all of these people show up with very limited information to the event, and now they have to troubleshoot. And not only do they have to troubleshoot, but they have to do it in a high stress environment. Right? We all know how critical it is, to get equipment back up and running. I'm talking from experience. Leaving the Navy, I started working as an industrial electrician in the automotive vertical. So I know that my number one priority was restoring that equipment back to, operability and getting getting the operation back up and running. So it's it's it's safety involved in that. It's did I get it right, not verifying, trying multiple things around the machine to get it up and running. And if it starts great, you watch it run five, ten minutes, and it takes off and it runs for the rest of your shift. You're okay. But it's that part of verification, that missing link of getting that closed loop audit, right, of actually knowing that your maintenance action was efficient that's missing in a lot of our a lot of our our teams and a lot of our cultures, that goes into developing a larger complex problem. Something that maybe needed lubrication or alignment turns into ten small downtime events and then generates a big downtime event that has mechanical and electrical failure. Then now you have a major downtime created with all these other little stops that go along the way with that. So how does AI actually help you? Help your maintenance teams, help your engineers, help your quality guys in real world scenarios. Well, we developed Sensei to help you in three very specific categories that actually streamline that entire event. So number one, your team is going to be notified by an attention engine. Right? And it's that, ability to create notifications when you need them and eliminate distractions. If I have a system that's just generating alarms for your guys, eventually, they're gonna stop looking at them. Right? They're gonna ignore them. You're gonna be like, ah, whatever. It's all false positives. Who cares anyways? Right? So we understand that getting notifications, getting the right notifications, getting a high level of fidelity when it comes to accuracy, and then eliminating false positives is very critical when we talk about predictive maintenance and an alert notification system for your maintainers to actually keep them engaged and focus on the equipment that they actually need to perform actions on. So this dynamic AI powered notification system allows us to do just that. Now this notification system is powered by another AI engine that runs in the background, which is all ML analytics. It's the way we go about creating models. Now why is this important? What differentiates Sensi from anything else out there? You know, these black box systems or these, other predictive maintenance solutions where they're they're solely model based based on data scientists creating very specific sets of data and data algorithms for your equipment. All those things are nice. Right? They're they're they're good. I'm not saying they don't work. They work in very specific scenarios. But when you alter that scenario, you change the speed of your mode or you change the recipe, you change the product that you're running, then you might as well throw all that out the window and start back over again. Right? What we what we notice is having and leveraging ML, to automatically create models from the bottoms up approach, kinda like Ronak's pyramid there. We're we're gonna start with the measures and creating models around measures individually. Then we're gonna create a measure, or a model that encapsulates all of those smaller models within itself, and that's going to be your asset. That's going to be your motor. That's going to be your pump. That's going to be your robot. And then based on that, we're gonna create different models that mirror the equipment, operation, its function. Right? If you're running an auto, I'm gonna filter out all the data. I don't care. If I am running in recipe a or recipe b, product x, product y, I'm going to change the model, and I'm going to create models based on what you're actually running. So I'm going to have very, very, very high fidelity and accurate models built because I have an understanding of what your equipment is actually doing. Now now we have that portion covered. Right? We have our data. We transform the data into information, into information in the way of notifications for your maintenance teams. Now your maintenance team is notified notified that there's a potential issue, a forecasted issue into a piece of equipment, whether it's a robot, whether it's a welder, whether it's a pump, a conveyor. They're gonna have a notification to go do something. They're going to see a measure, and any other solution out there is going to give you an alert. It's probably a threshold based system. You get a one a zero turning into a one, and that would require an action. But you still don't know what to do. Right? And that's where that gap comes into play, that knowledge and expertise gap. We included Gen AI into our tool within the last six months to help bridge that gap and gather all the information, you know, from Brightly, from your work orders, gather the information from your maintenance teams, gather information from the ML analytics, gather information from my notification process, my notes. Everything that comes together from this UI will actually give me a very prescriptive portion. Right? That prescriptive maintenance culture where I actually have a dialogue that guides me through what is going on with my piece of equipment, why the application has notified me that there's an issue, what that issue potentially means, and then what the most important thing. How do I go about fixing it? And then once I do take an action, all that closed system based solution that I just talked about is having a work order done, having that data from that work order automatically fed back into this application, the Copilot learning from your action and saying, okay. This specific event meant this, right, meant this particular maintenance action, and then having the audit. Right? Having the action take place and then all the measures return back to its normal level of operation, which is the piece that most of us are missing. And understanding that what I did was actually be effective and did solve the problem at its root cause. Now Sensei itself works just like the pyramid that Bronek showed when he started this presentation. Right? I mentioned that we're going to have a bottoms up approach. The very base layer of that is going to be your factory networks. It's gonna be your assets. Right? The data sources themselves. Now I mentioned how brightly comes into the picture, how CMMS ties in, and the work orders are very valuable pieces of data. All of that is categorized into your maintenance data. Your preventive, your corrective, your reconfigurations, setting changes. Anything that you do to a piece of equipment that allows me to filter out information, give the system a better understanding, I'm going to take. The operational data that comes from your PLCs, that comes from your controllers, your DCS systems, that lets me know if the system is running auto, whether it's running manual, your teach pendants. If I throw that in teach, then I don't really care what my disturbance torque on my axes are for my robots. I know that you're running it in manual and that is not running its normal operation. Right? So I'm gonna filter that data out. I'm going to focus when the machine is running and then what product or what recipe, what weight, what speed that particular process is running at. The condition monitoring data comes from your assets, from your motor, from your drives, from your sensors, from your PLC, from your DCS. All that gets fed into the core platform of Sensei, which is three different AI engines running. They're all layered together to give you the detection diagnostics and forecasts tools that you see, as anomalies, trends, and step changes in diagnostics. And all of that is controlled through the attention engine, which actually filters out false positives and is dynamically changing the status of an asset as I see it. Essentially, health status as I see it as a maintainer on the UI, so I only focus on what is truly in an anomaly state and what is truly degrading. Right? So I'm gonna jump in, and I'm gonna show you guys the tool itself. Right? I think that's the best part of this. I think you guys are gonna get a better understanding from the tool itself than from me trying to break down the tool within slides. So I'm gonna jump into that and, give me a second here. Alright. Perfect. So you guys should be able to see, this this the tool itself. This is the UI. It's web based. So that means you could deploy this on any web capable device. So I have clients that deploy this on tablets. I have clients that deploy this up on screens outside their maintenance shop. All these accounts, there's no limited or we don't apply a limitation to how much users utilize this. Right? It's all about having this data accessibility and then transforming that into information that's valuable to your quality guys, your production guys, your maintenance guys. So it's far more than just getting notifications on failure. It's understanding your equipment and then how your equipment operated during certain circumstances and then utilizing that information to go about planning your maintenance actions, understanding where some quality defects came from because you have the status of your equipment. So what we have up here is a navigational level. This layer allows me to create a customized way of navigating through the solution. Right? This could be a global map that gives me a representation of all my plants across the globe and gives me a status indicator and an insight into those things. And as it pops up here, you guys will see that. Right? You'll see a map of several industries here, and I could jump into those. But as you see those, every single layer that I could jump into has a health indicator. That health indicator is going to show me the highest level of attention case that's within that specific sector. And as I navigate through those, you'll see all the information that's down below the list of cases, and then the information on those cases be filtered to the specific level of navigation that I'm in. I'm gonna jump back into the the, level that I was showing previously, which is a level that we created for some of our shows. So we're back. Right? We're looking at this very customized imagery, and we have different sections here, all all with a condition indicator on their own. But I have one in that's in red. Right? I wanna focus on red. So if this was my area of responsibility, I'm gonna jump in here. Now I could see that it's further broken down into this very specific area. Now it's limited to the three assets with three cases associated to the area that I'm in. I'm gonna hide the map because we're gonna focus in on the information and then how this tool actually works for your maintainer and for your quality guys, for your manufacturing engineers, for the people that are using this, not data scientists. Right? Easily could be used. We have things where you could exploit algorithms using derived measures within the application, but the tool is really designed on the front end for the specific purpose of solving problems. So let's go let's let's go ahead and do that. So I have a pump. I have an anomaly here. It says it was opened twelve days ago, so I get a date stamp. And then as I navigate through these, you'll see that the information on the right hand side, my right window is gonna change depending on what case I have open. So I'm gonna focus on the critical one. The Copilot is the newest addition here, and as I mentioned, is going to give me the prescriptive portion of this tool. So it's gonna let me know what's happening. It's gonna let me know that I have a rising trend, that I have a correlated behavior on these other measures. And as you see, they're all velocity. I have RMS. I have peak to peak, which lets me know that these are coming from vibration sensors. Right? But I don't need to be an expert. I don't have to know that to go into actually have some impact within my maintenance team. So as I navigate down, right, I see the historical information, the data, the anomalies, the evidence that these that the application is trying to show me. But the best part about this is the possible explanations. It's going to let me know, hey. It's very likely that you're running cavitation or that you're dry running your pump, that there's an issue with misalignment or wear or that it's normal behavior due to operational cycles as its last possible reason why I'm having a experiencing this rising trend in my measures. Now the information doesn't just stop there. Right? It it I have the ability as a user to interact directly with the Copilot. I could ask it what actions should I take to resolve. And the application is gonna give me the reasons, right, as it previously stated in the asset window in the case window. I'm sorry. But it's gonna break it down a little bit further. And then it's gonna let me know where that documentation is coming from, and this is key. This is making a system that has the ability to ingest data, making it smarter, making it smarter via the manuals that your OEMs provide when they ship you your systems. OEM manuals from your pumps, OEM manuals for your motors, your VFD, your drives. All of that information being stored here so it's easily accessible for the people go doing and planning, maintenance actions. Actions. So I could see I have a machine passport template. I could see I have asset data. I have failure characteristics, and I have contextual information that allows the system to understand where this data comes from, what it means to the system, and then what failure characteristics the pump can experience based on the data that it's collecting. Right? So it has an understanding of where these things are coming from. Now it's not limited to just that. This is where CMMS, where Brightly come into the picture. If I had issues so that have, with other pumps or with this pump a year ago or two years ago and something similar is happening now, if it's happened before, the system's gonna tell me. Right? It's gonna reference old work orders that let me know, hey. I've seen this before. It means a lubrication. I've seen this before. It means alignment. Right? So I'm gonna ask it just what documented work do you have on damaged seals. So it's gonna let me know that it does have a maintenance log entry, dated October twenty fourth. It's gonna let me where that document is located, who the technician what type of work was, took place, the cause of damage, and then the, the it's gonna give me the link to the actual work order itself. Right? And it's not limited to just understanding text. We have a state of the art VLM ingestion system that breaks down imagery. Right? So you could put flowcharts, procedures, yes and no decision points, and all of that be stored within the system to be available for your maintainers, for the people that need it most when they need it most. I know we were running out of time here, but, I wanna leave a little bit of room for for some questions here at the end and wanted, hopefully, you guys to understand, you know, that closed loop maintenance that comes within predictive maintenance in Sensei and Brightly. I am gonna talk about, sorry, a little bit here, about what this actually does for your equipment. The robot reducer replacement, this is a success case where, this is what the what the system actually looks like. For for the automotive people that are that are here, you'll understand, the complexity of doing taking an action on joint two on on that reducer, especially if it's a heavy, big robot. Having a counterbalance system and then having to hoist and remove all the weight off of it, take the gearbox off, take the motor off, replace that, getting it all back up together, and then hopefully, you don't have to mass master your axes and then verifying every single point. You're talking taking a four hour maintenance event for a very, very, very experienced group of people. Now if you don't have that within your facility, you're talking about eight plus more hours to get all of that done. Right? It's it's a long period of time to to be without your piece of equipment, especially a robot that shuts down your line. Right? We we know the type of impact that has. So when you look at disturbance and you look at torque on your axes, you could have up to a five month lead time where you have a clear view into what's going on with your equipment and plan that ahead of time. Know that that gearbox is gonna fail. And instead of doing it on the fly while production is running and trying to move your operators around because because, you know, they're done for the day. Planning it on a downtime event. Planning on a a PDM on a on a preventive maintenance, predictive maintenance, Saturday or Sunday, and having this work done on your time and on your schedule by having the equipment, planted for you. And, yeah, that's it, guys. That's it on me. So, hopefully, you guys enjoyed the presentation, and I'm excited to hear what you quest what your questions. Thank you both that was fantastic we do have some questions here that have come in. The first question that came in here is for Ronak and it has to do the person is asking can you explain what would be the first step in explaining to a manager, why or I'm sorry, how to make the case for an asset management system? Yes. So for for the first case, so depending on the maturity level, and this is this is how I see it, and this is how I pursued it in my previous career. So the first thing you do need to have some sort of a business case, and you have to time that business case with the production and with the overall quality. Like, by implementing asset management, you will have let's say, I'm just throwing the number five percent more production by end of twenty twenty five, Or you can increase your facility conditions by fifteen percent or reduce maintenance cost by a hundred thousand dollars on a monthly average. So something as any project, you need to tie in the numbers. You need to tie in the data that can reflect. So if I'm a plant manager, if you come to me and if I'm a CEO and if you tell me, hey. I just need the asset management without having any context behind it, it's not gonna make any sense. So only and unless they understand the impact. You need to make sure your business case has an impact on your company's overall organization. Right? So as I said, go with your production numbers, go with your maintenance numbers, or go with your overall expansion number, something which is more visionary or more strategic that can align with your c level executive mindset. Thank you. Similar question for you, Genero. This person is asking, when talking about sensors and AI, how can we make the case to leadership that this will improve the production process and improve output? Yeah. Absolutely. So we conduct assessments. Right? ROI has already been mentioned here a couple of times. And ROI is a big, big, big building block to proving this to leadership on how a application like Sensei could have an impact within your specific environment. So we assess downtime that you're currently experiencing. Right? And that goes not just breaking down hour wise how much product you're losing, but goes into labor. It goes into all the things that actually impact a downtime event. And then we calculate the type of, efficiencies and the increase in productivity and sustainability within your equipment by reducing that downtime. Just to give you an average, our clients experience a thirty percent reduction in downtime within the first year, and within the second year, they experience up to fifty percent. Right? And that's huge. Great. Here's another question for Hennaro. Are the first two steps of data gathering from the day of Sensai implementation, or does your team pull in the history that we already have? No. That's a very good question. So we do both. We we get to pull in historical data, to kind of train the system and get it ready and look at deviations, that you've experienced previously. But we're mostly going to start the attention engine, which is what actually runs the application day one. Right? So from live data. And the reason why we do that is because all of those AI layers are interlocked with each other. So when there is an anomaly, the system expects result in actions. And without those actions in your historical data, it skews the way the attention engine perceives, like step changes or, peaks within your data. It perceives those as normal operation without any active feedback. Fantastic. And we have one last question that appears to be for the both of you. And it says, knowing what you have explained in this presentation, how would it have changed both of your manufacturing experience when you were working in manufacturing? So I I can go first and I can talk about it forever. Right? Because as I said, I worked in operations, I worked in maintenance, and I worked in engineering. So those are three different pillars of the company. But when I talk about these asset management and all these manufacturing practices, before, when when I was and I wouldn't talk about the company, but we didn't have any data. And that's where we implement this SCADA system. We implemented the brand new CMMS system, and we started tracking the data. The very first month were chaotic because first change management, trades, maintenance, unionized companies, they don't like it. Right? So first thing first, the change management was a hard piece, and that's where we had to get to work with the people skill. But then the second piece, when people start seeing the impact of the data onto your day to day or into their day to day operations and their work life, that's where the magic starts happening. So let's say if I'm a first line, operator, if I say that, no. Now Donut's team is maintaining this piece of conveyors very well, so I don't have to go and take that cardboard away every five minutes. It's taking away a lot of my headache. Now I'm seeing that that, improvement slowly but steadily, it's coming out. So as I say, it's a journey, and that's what I have it on the screen right now. It's a journey. As I mentioned, it's not like a flicking the switch. You start it's a baby steps, but as soon as people start seeing the difference, people start seeing the improvements around that, they will start putting more trust and they will start getting more engaged, especially around working around the unionized company work working around different diverse group of culture. That's what I would like to say. Yeah. From my end, unfortunately, when I came up, working as an industrial electrician, these tools weren't around. Right? Copilot wasn't a thing. I didn't have access to a GenAI capable system that could walk me through some of the issues and walk my teams through some of the issues that, you know, our experience in industrial manufacturing. But the the way this would have changed if I were in that environment now, I mean, it's it's it's a huge impact. Number one, like Ronick said, it's a journey. Right? Predictive maintenance is not magic. It's not something that I'm gonna be able to just walk in your facility, open up a box, and then you're gonna start getting predictions left and right, and the system telling you that something's gonna fail at three PM on a on a Friday. It is it is definitely exploiting data, understanding where the gaps are, doing assessments, and then taking the data that matters, putting it into a system to give you tangible results. And then those results having the type of impact where me as a maintainer, utilizing this tool not only affects, my level of attention, it reduces my stress. I'm able to focus on the piece of equipments that need it. I'm able to plan things in the head, which is the biggest thing. Right? Is understanding whether I have the components at hand to make a certain repair, talking to my production guys and being like, hey. I need thirty minutes here. I need an hour over here. When can I do these things? And then having those things be planned out without a system running, which affects safety. Right? It affects my troubleshooting, doing things under power. And everybody here knows lockout, tagout. Right? But realistically, like, are your maintainers following that type of procedure, you know, every time there's a downtime event? And it's it's taking all those things into consideration when you're talking about a journey of this of this magnitude. Awesome. Thank you both for answering those questions and for the fantastic presentation. This concludes our webinar. Thank you so much to everyone for attending. As a reminder, the recording, will be available soon. You'll receive it in your email with information on how to, view that recording and how to share it with others. And as always, if you have any questions, you can feel free to reach out and we look forward to seeing you at our next webinar. Those are always available on our LinkedIn page and you'll get more information via email about our upcoming webinars. Thank you so much and enjoy the rest of your day. Bye everyone.