Demystifying MES: Understanding Manufacturing Execution Systems

Reggie Valin (00:27):
Good morning, everyone. Hopefully you can hear me. Mark, let me know if there’s any audio issues. Welcome to Sepasoft’s Demystifying MES webinar. We really appreciate everyone on the call, taking the time out of your busy day to learn a little bit more about MES and our processes and how we go to market. I’m Reggie Valin, I’m Director of Sales with Sepasoft. So, sales contact, I run the sales efforts for Sepasoft as a company, and we look forward to working with a lot of you after this call on really helping identify opportunities where we can help.

(01:06):
I’ll be your host today and then we also have Mark French, our Director of Design Consultation. He’ll be taking you through the bulk of our presentation and then we also have Kent Melville, Sales Engineering Manager with Inductive Automation, to talk a little bit about the SCADA/HMI side. So, I wanted to start the presentation just as a brief intro around our partnership with Inductive Automation. So, you can think of us as brother-sister companies. So, Inductive Automation on the SCADA/IoT side and then Sepasoft, we are their MES solution of choice. So, a lot of our customers we talk to, they’re really looking for the flexibility that Inductive Automation’s Ignition brings to the table as a platform and then MES plugs right in from our Sepasoft perspective.

(02:03):
So, a little bit about our company. Interesting fact, the founder of Inductive Automation, the founder of Sepasoft, were actually system integrators in the field running into similar challenges, whether it was a functionality or lack thereof, budget issues, licensing issues. So, both platforms were built based on actual customer needs. We’re a fast-growing company, as well as Inductive Automation, growing so fast. Our founder and CTO decided to onboard a new CEO late last year to take on the business side of things so he can focus on the product and customer enhancements UI/UX. So, really exciting time here at Sepasoft. We’ve been around since 2010, over 1,000 implementations, small, medium, large businesses. We’re a global company, and one of the biggest differentiators is we’re 100% focused on MES. That’s all we do here at Sepasoft.

(03:03):
So, the agenda for today, we’ll talk a little bit about why is it important? Why is it important to demystify MES, some of the paradigms, some practical examples with the software, and then opportunities that we see within our customer base and then we’ll end it with Q&A. With that, I’ll pass it to Kent to talk a little bit about why this is important.

Kent Melville (03:29):
Yeah, thanks, Reggie. So, why is MES and demystifying MES important is a great question and a great place to start. I think that at Inductive Automation, we have really understood the value of MES for a long time and that’s why we’ve wanted to have such a strong relationship with Sepasoft so that customers could really enhance their systems. Raw data and raw control is not enough, right? We need to add context to that. So, why is this important? Four reasons I wanted to list here. The first is that standing still is not an option. You’ve heard the old adage that if you aren’t moving forward, you’re falling behind and in the competitive environment of modern manufacturing, competition from other vendors is constantly requiring you to improve your processes and improve your products.

(04:24):
So, if you’re looking to gain or even just maintain an advantage, you’ve got to be implementing new things, trying new things, evolving your process and really the advantages that you can get often come in the form of reduction in cost, improvement in quality or growth of market share. Also, a big part of what an MES system can help with is improving quality and productivity and these things do not just improve on their own over time. In fact, they don’t even stay the same. If you just leave your system as it is, it slowly will degrade over time.

(05:08):
So, without dedicated effort and dedicated systems to these things, you’re over time going to be in a world of hurt. Misunderstanding is expensive. When we talk about projects and deployments and things like that, especially of new systems, there can be a lot of misunderstandings that come into play. Those can be misunderstanding the engineering requirements. It can be misunderstanding your system and how it works. It can be misunderstanding what the outcomes are that you’re looking for and how what you’re doing today is going to affect the outcomes that you want, and we’ve found that this solution with Sepasoft and Ignition can really help reduce the number of misunderstandings at play, and then also we want to talk about some concepts today that can really help you get a firm grasp of what you’re looking to get out of your system.

(06:09):
Next one, massive improvement awaits. We really do, think, and have seen a lot of massive improvements as people adopt MES to improve their manufacturing processes. So, we’ve seen customers experience single to double to even triple-digit percentage increases in productivity and then also continuous improvement and effectively making quality goods with this data-driven approach. It can really help be a massive opportunity for improvement. Lastly on here is that failure to progress will result in regression, which ties into that standing still is not an option and essentially it’s just we’re in a world of friction, forgetfulness and failures and not moving forward is just going to really cause regression within your systems and your neglected infrastructure will not be as efficient in the future as it is today. So, the time is now. The time is now to start improving your systems and we think we’ve got some great solutions to help you do that.

(07:18):
Moving on to MES paradigms here. So, there are multiple paradigms for MES, and today we’re really going to be looking at two of them. One will be a structured model approach informed by ISA-95, and the other is a newer data independence-driven approach, often called a unified namespace. So, we want to talk about these two concepts, how they fit together, are they mutually exclusive, all that kind of stuff, but to start, I’ll turn it over to Mark.

Mark French (07:57):
Thanks, Kent. All right, let’s keep it moving. So, here on screen, you can see at left the classic Purdue model. We want the Purdue model to be first here because they came second this week. Anyway, so we can see this kind of classic manufacturing IT stack. This is also captured in ISA-95. So, for those of you that might be newer to the space, talking about OT and IT systems fitting together, let’s just do a quick review here. So, you can see we start at the plant floor at the bottom with the machine or process layer. So, we’re typically talking about PLCs, sensors, drives, all of that sort of stuff. What we talk about is operational technology or OT.

(08:49):
As we move up this stack, we’ll notice some trends. We notice we’ll move away from that cyber-physical relationship there at the process level. We’ll move physically away from it. The data will become more complex. The rule sets will become more complex and the time scale will change as well. So, at the machine layer, we’re physically touching the process. We’re controlling the process, and we’re running at maybe single digit millisecond as a rule of thumb. Of course, we have slower and faster, but that’s a good rule of thumb and the data’s pretty simple, right? Turn this on, turn this off, and so forth, and we’ve all written PLC code that’s really fancy, but for the most part, that’s the standard.

(09:42):
And as we move up into HMIs, human machine interfaces, we have 250 milliseconds approximately as a good rule of thumb for time scale. We’re starting to bring in more data to influence or determine how we present information to the user, might start to introduce alarms and things like this. There’s control aspects, so more complex and a step away from the physical process. Of course, SCADA, supervisory control and data acquisition, might be looking at entire plant or multiple plants. So, we’ve taken a big physical step back.

(10:19):
Now, the data is yet more complex in terms of scale, in terms of applying different rules for various conditions, alarming rules, historian rules, things like that, and the time scale is around one second. As a rule of thumb, can you run it faster? Of course, can you run it slower? Of course. That’s a good benchmark, and we see when we get to the manufacturing execution system, MES, or manufacturing operations management, excuse me, we’re now dealing with more complex subjects. How do we schedule? How do we control the production activity and more complex reporting and analysis requirements?

(11:07):
Here, we typically have a one-minute time scale, and again, we’re further removed from the plant floor. Lastly, we have our enterprise resource planning representative of business systems, which may not even know what the plant floor looks like. So, here again, time scale increases. Maybe it’s a shift, day or week, and the rule sets there. I don’t really want to get into it. It’s more complex. What are we doing with taxes and payments and all kinds of stuff like that, which is very, very far from the plant floor, right? So, ISA-95, when it looks at MES, is concerned about these different application layers, the functionality required to fulfill these of tasks.

(11:56):
So, the standards, whether ISA-95 or 88 or others, give us robust data models to affect what we need to affect at each layer. Furthermore, it’s concerned with interfaces between these layers. When we have these different applications and different functionalities, how do we make sure that they talk to each other effectively and share necessary information across? So, this is very structured, very robust, very proven approach to MES. Now, one thing that we’re doing that’s a little bit interesting here with Sepasoft and Inductive Automation is what we see in the stack at right.

(12:40):
We’re combining three layers, removing unnecessary interfaces because we’re all on that Ignition platform. Ignition being best in class, talking to the machine layer and OT Systems. Ignition, again, being best in class in connecting to IT systems as well. So, we have great interfaces on each side as far as protocols and lots of options and then we have robust data models within the standards that apply across a great many industries there. So, this is the kind of standards-based approach, if you will, to MES adapted on the Ignition platform. To explore this a little bit more and zoom in on MES, I would be remiss if we spent an hour talking about demystifying MES and we didn’t get into some nitty-gritty here.

(13:34):
So, you can see this graphic explodes that MES layer, and we really see three kind of categories come to the forefront when we’re talking about MES work. We see definition, control, and tracking. As we take raw materials and transform them to finished goods, we see those three categories come to the forefront and what does that mean practically? Well, for definition, that means bills of material, bills of process, workflows, material master data, work instruction data, quality sampling and quality inspection type information.

(14:18):
When it comes to control, here, again, we have workflows and recipes during execution. We have the actual data around plant floor resources. What’s the current status of this equipment? What materials? Do we have line side available for use and things like that? And of course, data collection is critical during control so that we can analyze it later so that we can report on it in real-time and affect change before we make mistakes or expensive rework, and then that last category there of the three is tracking, live monitoring, and analysis.

(15:02):
So, there, whether it’s productivity and continuous improvement focused, quality-focused, material genealogy and tracking focused, those things are really critical. Of course, we need to integrate with alarm systems as well. Fortunately, Ignition has best-in-class alarm system as well. So, we can see how the MES layer here focused on those three categories with robust data models there can provide definition control and tracking and on either side of that, we see a lot of emphasis in this graphic on connectivity and optionality in sharing data as far as interfaces are concerned. So, I think hopefully a helpful graphic to situate this standards robust model approach to MES. With that, I’d like to have Kent come back on and speak for a minute about what is emerging and claimed by many to be a competitor to MES, unified namespace. Kent.

Kent Melville (16:17):
Yeah, that always makes me laugh when we talk about this as a competitor because there are some things that are just true and things that are just needed, and I think we see this in all kinds of parts of the industry. One is we had the concept of SCADA, which Inductive Automation started as a SCADA company. That’s still a huge part of who we are is we do SCADA, which is supervisory control and data acquisition. There’s all these disparate systems. We need to collect the data and bring it in and have one unified system and then we went to IIoT, Industrial Internet of Things, where we have a bunch of disparate devices and systems and we need to collect the data and bring it into one system and now we’ve got a unified namespace. What is that? Well, it’s a way to connect to all your different systems and bring the data in. So, you have one system.

(17:13):
I think really what these are is they’re not new ideas. They’re refinements on ideas, and what we want you to get out of this is that we’re not throwing out everything that you’ve ever learned, and this is a whole new thing to learn, but this should build on where you’re at now and where you’re coming from. So, what is a unified namespace? A unified namespace is the concept that as we look to the future, we need to be able to have a single pane of glass into all of your data and your structures so that you can then use that in meaningful ways, and we’ll talk about what some of the uses are for that single pane of glass here in a minute, but essentially, a unified namespace is collecting data from all over your enterprise and providing structure for what that data should look like so that it can be queried easily and then you get apples-to-apples comparison of your data throughout your enterprise instead of it being in completely different formats throughout your enterprise.

(18:18):
So, it’s about the connectivity, the connections between things. It’s about the structure of the data and also the naming schemes of that data. So, moving on to the next slide here, what are the benefits of having that unified namespace of that single pane of glass? Well, it makes your data more accessible. So, being able to get access to your data easily is going to enable you to make decisions on that data in the future. Secondly, it’s a more cost-effective way to gather your data. It seems like everybody under the sun has got a new interface for how to store your data and query your data and if you have all these disparate systems, especially as you start looking at the cloud, it gets really pricey to access your own data. So, this is supposed to help alleviate some of those costs.

(19:11):
Also, you’re able to get that data together so that you can predict when things are going to go wrong. So, part of knowing if your system is healthy is knowing what your system looks like anyway. So, getting that data into a single pane of glass helps you know the status of your system, the health of your system, provides better traceability because now your data is not in all these different places. It’s in one place to see it, and you can track it better and also from a scalability standpoint, as you add more stuff, you can fit it into that single structure and it’s easier to add stuff because you’ve already got a blueprint for what that’s supposed to look like.

(19:51):
Coming on for a couple additional benefits here, we talk about a single source of truth. When you do have disparate systems, it can be… I remember when I first took statistics in school, I was a huge fan of math, which is why I went into engineering and all this stuff, but statistics never sat well with me because you can make the numbers say whatever you want them to say, especially when they’re in disparate systems. So, having a single source of truth to try to get a single right way to analyze your data. So, that across your enterprise, you have the same mindset of, say, what the health of your system is, can be really powerful and one place to query that data.

(20:33):
Also, by bringing this into one system, you have less systems to maintain. So, less manpower, less engineering effort. I will say that by saying less manpower and less engineering effort, that means once you have your UNS defined, UNSes don’t happen overnight. So, there is still some effort to get up and running, but once you’ve got it established, it really makes your life easier and helps you understand your whole supply chain better. It also boosts efficiency overall, especially when paired with great tools and then allows for better decision-making.

(21:07):
Now, this can be a human making those decisions because they have the tooling where it shows them that data in a meaningful way, but as I go to my next slide here, you’ll see that people are also looking to have that decision-making offloaded and be done by AI and ML and PM and all that kind of stuff. So, that’s really what’s been driving this UNS in a big way is that people are finding that they’re not making good decisions within their organizations. Things are falling apart, and they’re being very reactive to these issues.

(21:50):
When they come up, they’re reacting and they think it would be great to have all the data in front of a person for a person to notice when there’s going to be a problem, but things happen so fast. That’s not always feasible or whatever else. So, they’re saying, “Can we offload this to machines? Can machines figure out when there’s going to be a problem, how I can optimize my system?” All that kind of stuff, and to be honest, as smart as we think these machines are, they are really dumb when it comes to understanding different structures of data and finding… You give them data that’s apples and oranges and they’re supposed to figure out an apples-to-apples comparison, they’re not good at that. Machines are really bad at that.

(22:32):
So, people are driving to a UNS because if you can get your data into a consistent structure, then you’re giving the machines a fighting chance at actually delivering you valuable insights into that data. So, a UNS is really a first step if you’re looking at doing AI and ML, but yeah, also here, large enterprises are seeing that they have problems. They’ve got silos of data. How do they fix that and then how do they do all of this in a cost-effective way? It’s tricky. So, a UNS is really not a defined standard in like, “Hey, this is exactly how you should structure your data. It is really just an acceptance of the fact that the data needs to be structured,” and also the way that you structure your data won’t necessarily be the same way that somebody else is going to structure their data.

(23:26):
So, a UNS is going to be customized to you. So, in some ways, people go to tackle UNSes and they find that it’s maybe prompted more questions that answers because now the question of how do I actually connect my systems? How do I bring it all in? What is the right structure for my data? Those are things that you figure out along your journey to a UNS. The UNS doesn’t necessarily provide you all the answers out of the box for those things. So, with that, we said that it doesn’t have all the answers. You’ve got structure and non-structure, and how do we marry that together? So, Mark, let’s talk about that a little bit.

Mark French (24:06):
Yeah, I think this is probably the slide that I was looking forward to the most in the presentation, just with the state of the industry right now, some of the things that you drew out there that I think folks want to hop on this UNS bandwagon and think that it’s going to solve maybe some fundamental problems upstream that they need to fix maybe with MES, but I’m totally partial. So, I’m borrowing this kind of phrasing of the problem from my colleague, Doug Brandl. So, yeah, are these systems in competition or are they complimentary? We see at left here this very rigid structure from the equipment model of ISA-95 standing in for our standards, and at left, we’ve got this fluid, “bring everybody together” unified namespace, and really, I think to view these as a competition, I think maybe misunderstands some of the purpose of MES and overstates what a UNS can do, I think practically, by bringing data together.

(25:26):
We still need an integration point. We still need a place to apply things and inform operators, organize data payloads to equipment. I think we need that integration point, and as I was thinking about this, I was like, “What’s an apt metaphor for this?” Kent, you sing, you rap. You perform musically. So, I was thinking, if you’re singing in a band, we all take it for granted that you’re singing from the same sheet of music. So, it’s like, “Well, if I take a step back from that, the musical notes are a data encoding,” right? There’s literally a scale, a standard from which to organize that, and you need someone also, either a bandleader or a conductor, to keep everybody in time, and then also when the critical moment arises, that band leader, that conductor draws out the most important element.

(26:36):
It’s the trumpet solo, “Hey, I need it a little louder,” or it’s literally a horn on the plant floor because we’re going out of control on our quality measures and we’re about to start creating scrap, so we need to intervene now. So, I think that orchestration example shows where MES is necessary and then can contribute that structured data that serves as an integration point for plant floor activity can serve that up into the unified namespace where it can then be leveraged for machine learning, AI, and many, many other applications. What are your thoughts on competition versus complimentary?

Kent Melville (27:28):
Yeah, no, I like your metaphor. You’re speaking my language with the music here, but I think sometimes people see the structure of ISA-95 as limiting and structure in general as limiting for a project that slows down the project or whatever. They can’t represent their real world scenario perfectly because they’re trying to fit into this standard. Same thing with music. Maybe people feel there’s too much structure. They want more jazz, so to speak, but jazz actually has a ton of rules because the way that people can be creative is by the other people doing what’s expected. For some people to do a few things that are unexpected and revolutionary, you need a really strong backbone in order for that to work.

(28:24):
So, with Ignition, what are the features that people ask for from our platform all the time? They want more templates. They want more UDT structures to structure their data. They want reusability of code and how they do it. They want inheritable projects. In order for projects to scale and to solve big problems, you have to have a lot of structure, and I think that ISA-95 does a good job of defining structure that’s going to be a pretty good fit for most people. Now, I do think the industry has learned a few things in the last 20 years, and I think UNS is accepting the fact that not every system you’re going to connect to, not all sources of data are going to have that same structure.

(29:19):
So, it can’t be just this top-down approach of everything has to fit this structure and also the ISA-95 structure, you may have a few exceptions, and is that so bad to have a few exceptions to the 95 structure if there’s a legitimate reason for you to do it, right?

Mark French (29:37):
Right.

Kent Melville (29:38):
I don’t know that it is. So, I think that you can take the starting point of ISA-95 and then add the ounce of common sense of, “Okay. Well, there’s actually other data. There is things in other structures. I may have other structures in my organization,” but it doesn’t have to be all or nothing, right? I don’t have to say, “I can only do ISA-95 or I can only have other structures that I completely invent.” You say, “I take and I figure out a rigid structure for me, and then I can do jazz on top of that.” You know what I mean? That’s where it starts getting to the magical side.

Mark French (30:20):
Well, I hope everybody’s hanging in there with us. I really want to have this discussion with you. I think we could go for another hour, but I think we do need to get into some practical examples. So, just to wrap this up, I think we have here an understanding of structure and an open capability that can be leveraged together to facilitate real-time data, better analysis, better flexibility moving forward in the industrial environment. So, thank you, Kent.

Mark French (31:01):
All right. Well, let’s get into some practical examples. For the sake of internet speeds, I’m going to pause my camera for a minute, and we’re going to look at some examples that highlight definition, control, and live monitoring and analysis or tracking, that kind of three-part categorization of MES functionality. Again, when we think about definition, we’re talking about materials, we’re talking about equipment, quality, workflows, work instructions, and across the product suite at Inductive and Sepasoft, we can leverage these things. We have dedicated models again to support holding that information and applying it on the plant floor.

(31:51):
So, this example is… I’m actually going to play this twice. So, here, we can see our document management module where we can see work instructions in the prospective client session and we’re actually accessing this from a configuration management perspective where someone could affect what instruction a user sees on the plant floor during a certain process step. Obviously, the big draw here is the 3D model that is present in the work instruction. Well, here, let me restart that. Sorry. There’s a little bit of leadup. So, yeah, we’ve got the sports car in there representing the 3D part. Maybe we need to look at… flip it around and see the face or see some measurements on it. Where do I place the calipers? That sort of thing.

(32:56):
Or maybe we need to take a look at it in this 3D rendering to properly do the inspection or something like that. So, when it comes to definition, whether work instructions or master data, we’re leveraging the standards to capture that and then we can access that even within the client session for an engineer to make sure that the right work instruction is displayed to the user. We’re actually going to see some more work instructions, excuse me, let me go to the next slide, in some of these later examples as well.

(33:42):
Apologies. Don’t mean to cough in your ear here. Moving on to process control. So, real quick, we touched on definition, specifically, work instructions. Now moving into process control. So, we need to allocate resources as people, material equipment, and more. We also need to make sure that we’re dispatching and executing production properly, making sure that the correct procedures are followed, whether we’re talking about batching processes, discrete assembly processes or continuous processes where we’re doing more monitoring, but maybe we have a particular procedure when we start and stop around some discrete event.

(34:33):
So, here in the process control example, going to show a food-kitting scenario. So, whether we’re talking about kitting for line-side, automotive applications, or in this case, kitting for a batch of Italian sausage, fundamentally, the data is very, very similar. So, you can see I went ahead and started my procedure here, and I’m going to start kitting dry ingredients for this particular product. So, you can see the top table shows me my bill of materials and my target weights. The bottom table shows me my available dry materials at WIP inventory that I can select from. So, I’m scanning in those materials, and I can see my bill of materials and fulfillment up top as I go.

(35:30):
So, serial number level tracking of material lots provided here. Of course, we’re tracking quantities and we’re also tracking bins of material as well with expiration dates, and so forth. So, all of that is being presented in the user and the user being guided through that process with an interface here in the perspective module. So, of course, if you’re new to Ignition and Perspective, basically, dream it, do it here, what you want the screens to look like that’s available to you.

(36:11):
So, here again, we are continuing with kitting. So, our dry ingredients have been kitted into bins and now we’re kitting the wet ingredients into a wet kitting bin. Again, capturing quantities, weights. Obviously, there’s a lot of clicking on screen in this demo and that’s rather arbitrary. If we have a scale that is connected to Ignition, that’s very, very common. We could get these weights automatically if we have a fixed scanner hooked up to A PLC and we scan the barcode automatically off of conveyance. That’s very normal as well.

(36:56):
All right, so now we go to our charge and mixing station. You’ll notice we can see which batch we’re running, the formula, and the recipe. We understand when the material was kitted and the remaining bins that we need to process. We’ve got our work instructions displayed to the operator, and then we can start scanning in our bins, charging those to the industrial mixer, and that’s being recorded. What material was in that bin that we just added? Updating the bin so we know that the bin needs to be washed before next use, and we’re also keeping track on a macro level, where are we in this process? The wet-kitting bins have been charged and now the dry bins have been charged.

(37:42):
Once we’ve completed that with our workflow, we’re into mixing. So, the typical SCADA/HMI functionality that we take for granted on Ignition, well, we can leverage that, and in the same screen, single pane of glass where we’re leveraging the MES functionality. So, we’ve discharged the bin. Let’s go ahead and take a look at some reporting then, and apologies, if I’m moving too fast, but we’ve got a lot of ground to cover. So, one thing that I’m interested in right away is my material genealogy. I’m interested in what materials went where as I made this product.

(38:26):
So, here, you can see I selected the bin and I can see the wet-kitting process, what materials came in, what materials came out displayed on the trace graph. At left, I can see details about every material, its serial number, what material it was, where it came from, the timestamp, the expiration dates and quantities, all of that captured and available for integration with the reporting module as well. Now, at right, I have the trace graph and I can see the individual lots of material. In blue, the blue top nodes, let’s say I’m interested in that pecorino romano cheese. Where did it go? Maybe I have a problem with it. Maybe my taste tester said, “Hey, there’s a problem with that.”

(39:19):
Well, by clicking on it, I can instantly see everywhere that that raw material went, and then I can understand what intermediate goods need to be put on hold before I do more value-add, which finished goods need to be put on hold and need to be inspected. So, this can be searched from raw materials side, from the intermediate or finished goods side throughout. So, that’s some of that material genealogy and traceability. Looks like I’m skipping the time to kit screen. I’m not sure why. Oh, well, we’re going to keep moving. Giveaway, that’s everybody wants to reduce their material waste and also in many food and beverage applications, you need to make sure that you fulfill the minimum, but you don’t want to give away any more than you have to.

(40:12):
So, we can see here a quick screen to see giveaway. I do have some places where it looks like someone didn’t fulfill the material requirements of the recipe, so that’s problematic, but again, if we can see that information, then we can act upon it and improve that process. Here, I’m looking at some material expiration, so I can see particular material lots, material types. When did we bring them into inventory and when did they expire? So, lots of process control there, making sure that the correct procedure is being followed and that’s whether an operator is running things, the piece of equipment is running things, or it’s a hybrid of the two.

(41:04):
All right, doing a quick time check here. We’re a little behind, so I’m going to skip ahead here. We’re going to skip the motorcycle assembly today. I hope you guys don’t mind. All right. Quick demo on tracking and analysis. Here, I’m going to skip ahead for a second. So, here again, we want to see real-time our productivity scores. We want to see real-time costing us money. So, here we can see what we’re packaging on the floor. We’re calculating necessary KPIs, whether that’s OEE, Mean time between failure or others, and you can see in the downtime occurrence, card there in the middle. I can see what’s costing me by duration. I can also see occurrence count for those downtime events.

(42:01):
I’ve also got process capability and performance shown top right. Center bottom here, I have a time chart showing me the equipment status and how that affects other equipment on the plant floor. I had a stoppage that cascaded through the line that’s really important to understand so that I can address that and get that equipment back up and running and prevent that in the future. Now, because we’re bringing in all this data through Ignition and putting it in proper context so that we can execute on the floor, we can also do analysis on that.

(42:41):
So, here, I’m going to select a week’s worth of downtime data and start to get it into some charts. So, we have dashboard building tools and report building tools out of the box that you can leverage with perspective and the reporting module and give a little sample here of some of the data items. Notice it’s multidomain, right? So, if I want to interrogate my workflow and see information in a table or chart on when I’m signing off on electronic signatures, I can do that. Similarly, if I need to access any downtime or OEE-related data, I can do that, and also here, I’m going to bring up the SPC data so you get the idea. We want to put the engineer in the driver’s seat so that they can build whatever reports or dashboards are necessary for their users and you can see here, if we’re collecting quality data, we want to apply statistics, control limits, control rules, et cetera.

(43:48):
Now, all of this data isn’t just for reports and dashboards, it can also be published to Ignition tags and then therefore leveraged alongside the alarm system and all of those features as well. In the interest of time, I’m going to have to cut this little demo section short, and all right. Sorry, everyone, but if that wets your appetite, and you’d like to go into greater detail on any of those things, we’ll talk about that in just a few minutes. I think I’d like to hand it off to Reggie to talk about some of the opportunity that is represented by applying MES. Reggie.

Reggie Valin (44:39):
Yeah, thank you, Mark. Thank you, Kent. Really good information, and to echo Mark’s point, we’ll include our contact information to everyone. So, definitely, we love to continue these demonstrations, conversations with you on the one-to-one level. Before I jump into it though, there was one thing I did forget in the intro. I wanted to take a second to introduce our new CEO, Tony Nevshemal, if you could come on and say a couple words to the group.

Tony Nevshemal (45:08):
Yeah, sure. Thank you, Reggie. I wanted to turn on the camera so people can put a face to the name and thank each of you for joining us. We hope that the technical session was insightful and that you have a better understanding of manufacturing execution systems, but before I turn it back to Reggie, I want to say that this is part of a semi-regular cadence of webinars. So, stay tuned to our website and our social media for future webinars and events at Sepasoft. So, with that, I’ll just turn it back to you, Reggie.

Reggie Valin (45:38):
Thank you, Tony. Appreciate that. So, yeah, I know we talked about a lot today and I just wanted to spend a little bit of time talking around the value proposition. When I see successful projects and we talk to customers every day, it’s typically when all groups are involved, C-level, the business side, IT, OT, engineering, plant floor leadership, to really put together all of the requirements and necessary on that digital transformation journey. So, we understand that a lot of these efficiencies and the production, it comes to putting your teams all on the same page and having that real-time data to execute actionable items that can go for improvement.

(46:23):
We understand really if a customer’s talking about an MES solution, the end game is an increase in output, right? The bottom line, that’s what the C-level’s asking. How can we adjust that bottom line and we give you visibility to make those decisions? So, from our interactions with customers, when we’re talking to C-level, they’re looking for ROI. How can we lower our total cost of ownership? How can we achieve these levels of visibility to hold accountable departments spread across the United States? I want to view where I can see production and quality over 15 different locations.

(47:03):
I want to be able to make high-level decisions in real-time based on production and we give that visibility. IT, the CIO, what are they looking for? They want security. That typically starts with security. How will this system integrate with my existing systems? Is IT-friendly? What are some of the standards that we have to look for? What does the upgrade path look like, disaster recovery, and we cover all of those points with our customers. Engineering maintenance, one configuration platform. I talked to two customers this week that are coming to us from Aviva because Ignition and Sepasoft is just more flexible, right? It gives them that flexibility to have that visual interface that they can tailor.

(47:56):
I tell customers this all the time. Our goal is to get you 80% there, but you still have that 20% to tailor that to your company’s specific needs at all levels. Plant operations. How does this improve efficiency? Keep a pulse, target achievement, establish KPIs. The ability to see what the plant operators are executing on in real-time and make those adjustments and make them quickly. Unified scheduling, production control, quality management recipe. Just managing that workflow is really what we help customers accomplish, and again, we try to tailor that to your specific needs.

(48:42):
So, advance analytics, and I know this is something you all are having conversations about today. Machine learning and AI, how can we help with that today? Provide clean contextual data, serve as a feedback loop, provide the right data in the right time and the right place. Be open and available for what’s next. That’s really the key. What are the next steps to improve? So, we run in tandem with machine learning and AI along with analytics. Just an example here, one of our customers of Inductive Automation, as well as Sepasoft, SugarCreek, food and beverage, and they had similar challenges that a lot of you might have today. Day-old data, no information on downtime occurrences, no alarm notifications, controls on the equipment, and they went with our OEE solution for performance, SPC for quality web services to bring down that ERP data, and it really gave them that real-time visibility across multiple plants.

(49:47):
I believe six to eight plants gave them that unlimited licensing, TCO, more visibility, more control at the system level. They didn’t have to hire any employees, add any machines, add any lines, minimal overhead. They just had visibility. Now, they looked at the data every day. What can we fix today? Do it. What can we can’t fix today, fix by the end of the week. What we can’t fix by the end of the week, do it by the end of the month. First year, exponential growth just from visibility and making those changes. So, this is a public case study that’s on our website as well as Inductive Automation, but it’s really powerful around… Our goal is to really help you do more with the same.

(50:32):
So, now, I think we’ll open it up for questions and answers. Thanks for coming back, Mark. There may be some answers I can’t answer, so you can help me out here.

Mark French (50:45):
Yeah, I’ll do my best with the answers. One question that I see here is how do I get started with this and is Kent on as well? I wanted Kent on the Q&A as well. So, there he is. Thanks, Kent.

Kent Melville (51:09):
Yeah.

Mark French (51:11):
Well, Kent, I have my answer for that, but I want to hear your answer for that. How does somebody get started with MES, with digital transformation here?

Kent Melville (51:24):
Yeah, so Ignition as a platform is a good platform. So, what it has? It has connectivity, meaning we’ve got drivers to talk to different devices and start bringing in that data and be able to define alarms and history and all the kind of base functionality to get you some raw data. So, we like starting with that and then with the Sepasoft modules, we come in to start actually taking that raw data, those raw tags and everything and start putting that into a structure like the ISA-95 structure so that you can then go and start actually doing all the stuff that Mark was demoing.

(52:04):
So, where do you get started? I think you find a need, right? You go and you find a system that is… You need more visibility into it so that you can start improving the process and then you work on getting the connectivity, and then you start building out your system from there. So, I think it starts with identifying the system you want to tackle and probably not your whole system all at once. You probably find a line or something that you’re trying to tackle and then use Ignition to get that connectivity going and take it from there, but yeah, Mark, I’m interested in your answer as well.

Mark French (52:38):
Well, it’s really easy. From my perspective, some of my team here, Tom, our CTO, wrote a blog post like 14 steps to a successful MES implementation. So, he outlined a path to that, which starts with identifying the need, identifying the team. This is available on our website and will be part of the follow-up material, but I think you’re absolutely right. You’ve got to identify those requirements and you have to connect to systems. We have the garbage in, garbage out rule. For computing, it applies in this space as well. So, I love that answer.

Kent Melville (53:16):
And one other thing to add to it, I think, is you need to have the outcomes that you want in mind, right?

Mark French (53:23):
Yeah.

Kent Melville (53:23):
I’m trying to increase efficiency. I’m trying to be able to better track downtime and prevent downtime, but if you just say, “My goal is to have an MES system,” you need a better goal. You know what I mean? Because I think those outcomes are going to be important because the modules you’re going to use and the strategies you’re going to use are going to be dependent on the outcomes you’re looking to get.

Mark French (53:48):
I definitely agree with that success criteria. Another question that caught my attention here is what is your experience with life science industry? I’ll start first and then I’ll throw it to Kent. So, we actually have a validated systems guide on the website. So, if you’re in pharma or life sciences, strongly recommend you point a browser to sepasoft.com and it’s right there under Products. So, check out our validated systems guide and we have… Usually, customers in pharma and life science are the NDA types, but we can start the conversation from there. Kent, do you want to talk about life sciences?

Kent Melville (54:34):
Yeah, I agree that that’s a growing industry for us. I think that Inductive Automation and Sepasoft have grown to a point where 10 years ago, we weren’t in life sciences because we were not mature enough as organizations and the products were not mature enough and it’s been great over the last five years or so as the products have really hit their stride and become mature products and feature-rich products that we’ve seen life sciences really gravitate towards this because it’s this good balance of providing the structure that you need in life sciences, but also the flexibility that you need to be competitive and to stay modern and take advantage of these things.

(55:13):
So, some existing solutions out there, especially batching systems and things like that, are very old school and if you want to use them, you do them in the way that they are designed, and it’s been great to see some of the new releases from Sepasoft, like the Batch Procedure Module, and just the refinements of our platforms together over time. It’s really provided a great solution for life sciences. So, we even just started going to some conferences, some shows specifically around life sciences. We had a good time with them last year. We’re going to do some more this year and that led to the guides and stuff like that that Mark is referring to. So, yeah, life science is definitely a growing industry for us, and it really is around the marriage of the structure and the flexibility together.

Mark French (56:02):
Absolutely. I don’t want to cut the Q&A off yet, but-

Reggie Valin (56:07):
Mark, a couple questions to that I wanted to address real quick. First one is, yes, everyone on the call, you’ll get this recording sent to you via email here shortly thereafter, and then there was a question around training and certifications, and yeah, we take that pretty serious. So, we have instructor-led training by Mark and his team throughout the year that’s listed on our website under Resources, and then we also have certifications and a certification path. My Global Partner Manager, Vince Ares, he can walk you through that. So, if you email us here at sales@sepasoft.com, we can help you get down that path.

(56:45):
The other piece we also offer our customers is our Quick Start Program, which is Mark and his team, once you buy the license, we give you as many hours as necessary to help get that project started in partnership with you and the systems integrator. So, yeah, we take training and certifications very serious because we want our customers to have successful projects. So, I know we’re running low on time. Any questions we did not get to, we’ll get back to you via email or a call, but yeah, thank you, Mark. Thank you, Kent for answering those questions. I want to let everyone on the call know, all attendees, reach out to us at sales@sepasoft.com or Inductive Automation at sales@inductiveautomation.com to schedule a demo, if you need a quote, if you just want to have a conversation to see a little bit more around what we’ve shown today.

(57:35):
From a Sepasoft perspective, we also want to offer a 10% discount to any attendee of this webinar on a purchase if done within Q2 of this year. So, again, please reach out to us, follow us on LinkedIn. Like Tony said earlier, we’re going to have more of these, and we’re going to be as consistent as we can be to get this information in front of as many customers as possible. So, we look forward to interacting with you. Thank you for taking the hour out of your day, and everyone, enjoy the rest of your day and the rest of your week. Thank you all.

Mark French (58:10):
Thanks, everybody.

Kent Melville (58:10):
Thanks, everybody.