MES Matters with Matt Barber

David (00:00)
Welcome, you’re listening to the IT/OT insider podcast. I’m your host David. Subscribe to get the latest insights shaping the world of industry 4.0 and smart manufacturing. Today I’m joined by Matt Barber. Matt has 15 years experience implementing and designing MES for different companies all over the world. He currently is the VP responsible for the MES business at Infor. Now what triggered me the most to reach out to Matt is because he’s on a mission to

teach the world about MES via his website, mesmatters.com. And that’s exactly the stuff we like to talk about here at the IT/OT Insider. So Matt, thank you for joining me.

Matt Barber (00:41)
You’re welcome. Thank you for having me.

David (00:43)
Let’s kick things off maybe again. As always, short introduction. Where do you work? A bit about your career.

Matt Barber (00:52)
Yeah, absolutely. So I currently work at a company called Infor. Infor are the cloud ERP company, big on innovation, big on technology. And yeah, I run the MES business at Infor. So we were acquired by Infor in late 2021. And before that, we were a company called Lighthouse, which some people may have heard of with a product called Shop For Online, which has now been rebranded to Infor MES. And I’ve worked there for my entire career, which has been about 15 years.

David (01:22)
So that means that the only thing you’ve seen in your professional life is MES.

Matt Barber (01:26)
Absolutely. Although

I don’t want to get us off topic, but what is MES? you know, the big discussion around MES versus MOM, you know, we are technically speaking more of a MOM platform, but the world doesn’t know what that means well enough to market that way.

David (01:35)
Mm-hmm.

That’s a perfect first question to start it. I’m gonna bounce back the ball to you, Matt. So what is an MES and what is a mom? And I would say how do they complement each other?

Matt Barber (01:59)
Yeah, good question. and one which I get asked very regularly for me, an MES is a, it’s a specific type of application. That application is one that focuses on, let’s say the production related activities. So if you’re being very strict about the definition, it’s things like starting and stopping production orders, tracking downtime on your lines, production counts, scrap counts.

getting to the point where you’ve got an OEE figure, those kinds of things, that’s really what we call an MES. So you’re connecting to the shop for equipment to pull the data automatically if those signals are available, maybe some labor management. An MOM, Manufacturing Operations Management, it’s not really an application specifically, it’s more of a concept, but it basically expands out MES, which was really production.

into quality, so doing quality checks, maybe taking samples from the line, SPC, those sort of things. Inventory, so tracking materials that are consumed into the process, creating materials at the end of the process, managing your semi-finished goods, managing all of your WIP and looking at traceability. And then maintenance, and maintenance is kind of like asset care, maybe managing assets like tooling and gauges and containers on the shop floor.

So it’s much broader and MES is really a part of that. A lot of MES applications on the market do more than just production, but not many of them do production quality inventory maintenance.

David (03:40)
Yeah, Given the fact that MORM is more about the concept, the way I would say you operate your plant, I’m sure I would say that tightly, that links tightly to how a company is organized or how they want to be organized or how do you actually start such a discussion where a company they…

They send you an email or they ring your doorbell and they go like, I think it’s time for us to start talking about MES or MOM.

Matt Barber (04:15)
You know what, different people start in different places. You would get a lot of people that contact us. They just want to do something small. They just want to do OEE. It’s very, very simple. And that’s great. And I think you can get a lot of value by doing something small. Get it implemented quickly, learn a lot of lessons, build the relationships with the customer, understand their processes. And then once you’ve delivered some value,

Okay, now we want to expand, we want to do some quality or we want to do some maintenance or we want to do manage our tooling, you know, or we want to do energy monitoring on the shop floor. And you can layer that in on top as, as appropriate. And one of the things we’re very keen to make sure is that our platform is modular and composable and you can take as much or as little as you need because these applications can be quite a big investment. And there’s no point in paying for everything if you only need a little bit.

And we really want customers to get a good return on investment as quickly as possible and get value from the system. So yeah, that’s one way of doing it. I call that the step-by-step approach, phased approach. But increasingly, a lot of people are coming to us saying, you know what, we want to look at enterprise MES across all of our sites. We want to standardize. We want to make sure that we’ve got all of the systems in place to manage all of our operations. And I think that

You can take either approach or you could maybe take somewhere in the middle.

But there are several good reasons to do the second one. So the second one is if you think about being an operator on the shop floor, are at the machine and you are responsible for a lot of things. You need to know what the production schedule is. You need to make sure that you’re collecting the right data from your assets. Maybe that’s coming in automatically. Maybe some of that you’re happy to do manually. You need to make sure that you are

signing on to and off of the line. You need to be aware of your OEE. But also, you need to do quality checks. You you do some of them. Maybe you have to take some samples from the line and you have to send them to a lab. You have to consume material and check that the right you’ve scanned the right barcodes. You have to print the labels at the end and put those in the pallets. You have to communicate with logistics about what materials are available at the line to be consumed. And you need visibility of that so you can do your job correctly. And

You need visibility of the maintenance schedule so that you can see when the maintenance team are planning to repair your assets. You can see whether people have already requested for some work to be done because there’s a problem. Otherwise you log the same issue more than once and increasingly operators are being asked to do autonomous maintenance checks. So cleaning, lubrication, safety checks, those sorts of things. So what you don’t want is an operator using

three, four, five different applications to do that. You want to use one and you want everyone in the factory to be working off the same set of data for there to be no ambiguity for the data to be contextualized. When you do a quality check on a machine, it’s the same machine that’s starting and stopping. It’s the same machine that’s producing material. It’s the same machine that needs to be repaired by the maintenance department. If you’re doing that in different systems, then you silo all of that data or you have to build complex integrations.

And that’s really not a good outcome. So yeah, there’s a really good reason to do it all at once. And if you’re not going to do it all at once, then certainly there should be an end in mind, right? You should be working towards that to get this kind of coherent, rich set of data across operations.

David (07:57)
I noted down a couple of interesting follow-up questions. The composable MES thing will definitely come back to that because I think that’s one of the, I would say, the buzzwords around MES we’re seeing right now in the market. But first of all, you mentioned OEE a couple of times. So for those who are not too familiar with OEE, maybe just a, I would say, a brief explanation.

on the concept and I immediately have a follow-up question. Do I need an MES system to calculate an OEE or is it more kind of an optional thing? So your take on this.

Matt Barber (08:38)
Yeah. So OEE, the simplest way that I think I can put it is OEE is a measure of how much good stuff you make in the time you have available compared to the amount you could theoretically have made. So things that damage that number are you could have been running the machines, but you weren’t. We call that downtime, but in the OEE calculation, it’s actually the inverse of downtime, which is availability. How much time with your machines available for.

Then we talk about the count of things that were made. So how much did we actually make in that time? And then out of the things that we made, how many of those were actually good? Because if you had to scrap some of those, then, okay, you made it great, but you had to throw it away or you had to rework it or whatever. So all of those things have an impact on OEE. And can you use it in, can you do it in different systems? Yeah, you can record OEE on paper if you want.

You can do. MES is the natural application to do that, especially if you’ve got automated data capture, because you’re putting those things in. There’s a lot of applications on the market now, which are kind of IoT platforms. Those are typically collecting data from sensors and other equipment on the shop floor. A lot of them can do OEE calculations. Yeah, it really depends what you’re looking for.

David (09:39)
You have seen that.

Matt Barber (10:08)
Some ERPs calculate OEE.

David (10:08)
Yeah. Yeah,

probably also depends on the complexity of your process. If it’s very complex to, would say, to have these counts or know exactly what is being produced, et cetera. If you have multiple lines or you’re running in parallel or probably, I would say, I’ve seen OEE calculations in Icon from the chemical industry. So in continuous chemical plants where we definitely didn’t need an MES system.

to the calculation because the thing just runs. It’s just chemicals flowing through pipes. Easy to count that or to flow meters or extract the data from those flow meters. But I can assume that if you go into the more, I would say, if you’re talking about individual parts, maybe towards better or discrete industry, it might become a bit more complex.

Matt Barber (11:06)
Yeah, I mean, and quite often, especially on machines that kind of do cycles rather than make individual, if you’ve got tooling, for example, you might stamp your, might do the cycle once and depending on what tool you might make one piece or five pieces or 10 pieces. And so you need to factor that in as well.

David (11:25)
yeah. That’s a good point. That’s a good point. also comes down a bit to, I would say, how linear is your process? Do you really go from one station to another station? It’s really one step after another step versus does the… Is there a rip? Is there… cetera, et cetera.

Matt Barber (11:47)
Yeah, I think one of the yeah, a lot of people I mean, just take downtime as one portion of OEE, a lot of people track downtime on the bottleneck, which is good. I mean, you have to track down some of the bottleneck in order to do an OEE calculation. But actually, the most interesting thing is when you can capture downtime at every step of the process as it’s moving through the line, because you can then start to see, okay, the bottleneck was stopped. But was it stopped because there’s a problem with the bottleneck?

David (11:58)
Yep.

Matt Barber (12:16)
Or was it stopped because there’s a problem upstream or downstream and the bottleneck is either starved of material or blocked so that it can’t move material further ahead?

David (12:26)
So one of the topics we love to write about at the IT/OT Insider is not just technology, but it’s people, process, change.

even a bit of marketing from time to time because sometimes you just need to do some internal marketing to convince people how great the new tool is. But one of the, I would say, inherent complexities of MES is that you really touch the work process. An operator or a technician or whoever is using the tool is following. And what I see is in many cases a big difference where between the

the process, how it should be followed on the shop floor and how work is actually done by operators. And then once you come in with your big fancy MES project, they tend to become very complex because either maybe you want to adapt to all these very, I would say, the nonstandard things operators are doing or yeah, it comes in as some kind of yeah.

a process where operators go like, now I just need to stop working because this is not the truth. It’s been pushed top down or something like that. I’m sure you have questions, sorry, not questions, but you have stories about that as well.

Matt Barber (13:53)
Yeah, it’s a huge challenge. It’s a huge challenge in every project. think it’s really important to, you’ve got to tackle it from both angles. There needs to be the vision from the top. There needs to be strong change management in place. It needs to be sponsored by senior leadership. And it needs to be given credibility inside the organization.

But it also needs to be thoughtful about how operators are using the system. And it needs some of the strong voices from operations to make sure that they’re chipping in with their ideas and that they don’t just get given a system at the end of it. They get included on the journey. If you don’t do that, then you can meet a lot of resistance.

Ultimately, everyone wants a system that’s adopted, right? If you have the best system in the world and no one uses it, then it’s not the best system in the world anymore. It’s got to be used. And that’s something that we’re very, very conscious of when we’re running projects, trying to get the right people in the room, trying to assemble the right team. And it’s not just about assembling a team at the site. It’s about assembling a global team that can orchestrate and manage

David (14:57)
Yeah, no. Yeah.

Matt Barber (15:19)
centrally and be the center of excellence and then teams at each individual site to carry that forward and be the super users at each site. So you end up with quite a quite large team because you’ve got the central one and then you’ve got one at each site as well and they should all be working together and feeding into the processes. I think the good thing about MES is that implemented correctly, you should be

validating the processes. You define what the processes are and then the system should help you do the right thing. So many times we go in and we talk to customers or whatever and okay, this guy who’s worked here 25 years, he does the right thing. He makes the right decisions. When there’s a quality issue, he isolates the last five pallets. But is that what’s happening with the person who turned up?

David (16:13)
Yeah.

Matt Barber (16:17)
two months ago, probably not, because they’ve not been trained, is that what happens on the night shift, even if the night shift have been at 25 years, probably not, because they’re always go a bit rogue. But the system should make sure that people are following the right processes, doing the right thing, not making life harder, making life easier, automating things. And if you can get the system to work in that way, then that’s a real benefit. And that’s what we try to do.

David (16:46)
And that’s maybe a bit about indeed people and processes. Let’s jump into technology. The MES market on the one hand is a, I would say, rather mature market. We’ve been implementing MES systems for many, many, many years.

But I also see it’s currently a market which is changing quite rapidly. You already mentioned the Composable MES. We also see this, I would say, split or balance between on-prem and cloud first systems, et cetera. So maybe a couple of insights in how you see the current MES market, how you see it changing, where you think it’s going to end up.

in the next years.

Matt Barber (17:43)
Yeah. So the, the MES market has been around for a while. I think that the MES market historically and still quite a lot today, to be honest with you is let’s go and code an MES. know, vendors might have a toolkit, for example, and you know, let’s go and run a big project and code it. One of the roles that I did at Lighthouse before we were acquired. And then it was the role that I actually transitioned into Inforwith was

David (17:59)
Yeah.

Matt Barber (18:12)
heading up the product team. I came from a services background, running projects and implementing projects for customers. And I got quite frustrated because although we had a fantastic toolkit, so much of the project was just doing the same thing that I’d done in other projects. And especially when it was in the same industry or the same manufacturing processes, what did we do for customer X? Well, we did that. well, probably should we do that for customer Y because they’re in a similar process and you go off and you build it all in it.

takes a long time and that’s not a very good approach. So about probably eight years ago now when I took over that role, or perhaps slightly just before, I wanted to make sure that we had a really big push for moving as close as we can to an out of the box configurable system. And I’m very pleased to say we made a really, really good inroads on that journey. And you can, we actually presented to a…

couple of analysts actually, and we actually built a plan in front of their face in two hours. And I’m not saying it was all bells and whistles, but you had something, you had a dashboard that you could do quality checks on, you could start and stop jobs, you could consume material, you could do an autonomous maintenance check, you could see all of those things. You can view SBC. And that’s kind of the beauty of an out-of-the-box configurable system, especially in

David (19:25)
Mm-hmm.

Matt Barber (19:40)
of common processes. Now, of course, there’s more to that. You’ve to do integrations. You’ve got to do business logic. You’ve got to do workflows and other things. But the premise of getting something quick in front of the customer is actually really valuable. And the way that we run projects is we actually sit down in a workshop with customers. And the first thing we do is we build a prototype with them. So we get as how far can we get on your process just through configuring the system? And we might get

90%, like 80 % might get 95%, you know, depending on the process, but they’ve got something really quickly. And then it becomes a case of not what do you want, but this is what the application does. And where are the gaps? How can we, how can we fill those gaps? So I think the market is going towards configurable. There’s another part of the market that is going the other way. And I think that’s the lower end of the market, which is going build it yourself. So there’s kind of this, this split.

David (20:30)
Yeah. Yeah.

Matt Barber (20:37)
which is fine for us because we’re targeting enterprise customers. And I think the lower end of the market has always been there, right? We often go in and we’re replacing homegrown systems that have been built over the last 10, 15, 20 years. And that’s fine because that’s what that customer needed at that point in time. And there are plenty of manufacturers out there right now who need that right now at this point in time. I actually went to Walker Reynolds’ Prove It conference in Texas a few weeks back.

And that was aimed at that lower end of the market where you’re building it yourself. I learned a lot from that. I thought it was really interesting perspective from a lot of, a lot of different vendors and a lot of different manufacturers. But I think you’re always going to have that divergence, right? You’ve got the top end of the market, which is wanting standardization, global master data and enterprise solution, consistency out of the box, configurability, modern platform innovation, those words. And then you’re going to want another.

part the market, the lower end, is the kind of agility, being able to do it yourself, stitch different things together, pull data, put it into a common platform and kind of grow with your own needs over time without relying on a vendor. And they’re both fine, but they’re just different approaches.

David (21:57)
That’s a very interesting insight. I didn’t look at it, I would say, that way, but it’s indeed, I would say, you can’t expect the smaller companies to indeed invest in a bigger standardized system. So that’s a very interesting take. And what was then the role of Composable, I mean, yes, in this discussion.

Matt Barber (22:20)
Yeah, composable. That’s a big word. People talk about composable a lot. I think for us, the way we market composable is we have one application with all of the functionality in it, but you can just pay for as much as you need. So you take the bits you need and you don’t take the bits you don’t need. And if at some point in future you want to layer on something else, then you pay for it, you turn it on and you get the benefit. I think there’s another element to

composable, which is where you are using multiple applications together. And that really is, you you’ve got your landscape and your landscape is, it’s an ERP, it’s an MES, maybe it’s an APS for planning and scheduling. Maybe it’s PLM. And there’s all of these different applications, WMS, workforce management, you know, and they all come together and they all play their own part in the broader ecosystem. When we talk about smart manufacturing,

I fundamentally believe that MES or MOM or whatever you want to call it is, it’s one of the, if not one of, then the key pillar of smart manufacturing. You know, it’s the central hub for all of your operations. And it then reaches out to all of the other systems to get the information you need. You can’t have a smart factory without a MezMom type application, but you can’t have a smart factory with just a MezMom type application. It needs…

David (23:32)
Yeah, yeah.

Matt Barber (23:44)
It needs the other applications in the ecosystem to work together with it. It needs the information about production orders and deliveries and other things from ERP, the material mass data, the routings, the bombs. It needs information from PLM about how to make the product and documentation. And that might come directly to the MES or it might go through the ERP depending on how it’s configured. It needs a plan and a schedule, which again might come from ERP or it might come from an APS. But my point really is.

it, you can’t do on its own. So yeah, it’s, it’s, it’s lots of applications working together in a composable way, because you’re just picking a best of breed for everything and, and, and making them part of the same landscape.

David (24:27)
So that makes things a more clear for me regarding Composable MES. We’re often also talking about industrial data platforms. There’s also something we write about on the blog.

From my perspective, a data platform includes MES data because you want to build contextualized information and that requires MES data as well. We see or I see the role of data platforms evolving quite rapidly. They are shifting more and more towards the IT domain. Data needs to be consumed by…

by the data bricks and the snowflakes, et cetera, et cetera, of the world. Where on the other hand, MES still is really OT stuff. From my perspective at least, what’s your perspective? How do you see the IT OT conversions thing taking place in the MES domain?

Matt Barber (25:29)
Yeah. So I think historically the MES data was captured, viewed in the MES aggregated, sent to an ERP. That was the typical way things worked. And that’s still how things work, but there is an extra layer on top now, which is data lakes, data fabric. And that’s really been driven a lot by AI. You you want all of your data in one place that you could do analysis across all of the different applications.

David (25:44)
Yeah, yeah.

Matt Barber (25:59)
That’s one of the, I genuinely, I mentioned to you before this started, when we were acquired by Infor, we weren’t really sure what to expect. A lot of us were a bit miffed, being acquired by a big company, ugh. But actually it’s been brilliant and it’s been challenging for sure. There have been a lot of difficult bits, but the culture’s fantastic. The innovation, the R &D is fantastic.

And one of the things I actually like the most about Infor is they have put a huge amount of investment into what they call Infor OS. Infor OS is like the technology platform that underpins all of their applications. So it does data fabric, it does data lake, it does all of the RPA, which is robotic process automation. It does all of the AI, ML, generative AI, all of those things come together in that platform.

They do process mining. I could probably listen to the five things, but it’s all of that contextual, it’s all of that kind of advanced technology. And you can’t do that unless you have the contextualized data from all of the applications. Infor offers a lot of applications. offers ERPs, it offers the MES, it offers WS, WFM, CPQ. There’s so many different applications that it offers. You can’t move all the data everywhere. That has to go into a simple repository.

David (27:21)
Yeah.

Matt Barber (27:24)
And that’s something that they do really well. And it’s something we would never have done had we not been acquired. And it’s something that I think has to be done to remain relevant in this market, because that is the expectation of manufacturers now. They want all of their data to be accessible and to be able to be interrogated, know, to be able to run useful insights on it and to be able to do AI, machine learning, generative AI, et cetera. So yeah, it’s been a huge benefit of us being acquired, to be honest.

David (27:37)
Yeah.

Okay, so with all this information in mind, I’m sure there will be listeners and they will ask themselves the question. I’m interested. I want an MES. Where do I start?

Ha

Matt Barber (28:12)
One of the things that’s interesting about the MES market is that a lot of the MES applications are specific for particular industries. The first place to start is to try and get an understanding of the applications on the market for your industry, because it won’t be all of them.

I could start listing them. won’t list them just in the interest. There are particular companies who are very good at particular things. What you don’t want to do is go and knock on the door of a particular vendor because they’re scored high on a particular analyst report, for example, and then find out that you’re in food and bed when they do aerospace and defense, because it’s not the same. So you’ve got to pick the right ones. You still look at the reports. Of course, those reports are really useful.

David (29:02)
Yeah.

Matt Barber (29:07)
go and look at the IDC one, because that’s where we’ve just recently been rated as a leader. But you need to just be aware that just because it’s got 20 or 30 vendors on there, there might only be two or three that are relevant for you. So you’ve got to be really cautious about that. I think it’s really hard to find education on these things. That’s why I started writing the blog. It’s why I post on LinkedIn. And since I started, there have been a lot

of other people that have been doing that as well, either that we’re already doing it and I’ve now met or have started doing it since. So the community is bigger than I thought it was, which is great. And there’s a lot of good people to follow on LinkedIn. If you want more information, there isn’t really one central place to learn about the kind of broader manufacturing operations management place. That’s something I would like to do at time permitting in future. Maybe as a bit of a collaboration with some other people who are interested in it as well, because

time is tight and it’s a really big topic. But yeah, I think that there’s a lot to learn. There are lots of people you can talk to, anyone can reach out to me for a chat, no problem. There are specialists in the market who are partners that you can work with that will help you choose an application to work with. The only thing to be cautious thereof is that they’re

David (30:07)
Yeah.

Matt Barber (30:34)
it makes sure that they’re impartial and independent and they’re going to look at the right thing for you, not the thing they get paid the most on. Slightly controversial, but you’ve always got to be wary of those sort of things. But yeah, it’s hard. It’s really hard to know where to start. I that’s the, I’d probably leave it there. It’s hard, but it’s worth it.

David (30:42)
Yes.

It’s hard, it’s worth it. It’s probably figuring out what part of the process or what line or what warehouse or wherever you can, there is some low hanging fruit. Maybe also that part of the line where your level of automation is already on a high enough, it’s high enough. I would say it’s rather…

I would never say easy, at least bolting on an MES doesn’t require you to install a lot of additional sensors, something like that.

Matt Barber (31:34)
Yeah, if you’ve got automated lines already, that’s great. You can plug into those and you can get more real-time visibility and control and understanding of what’s going on in your process quite quickly. If you haven’t, it doesn’t mean you can’t start. You can start collecting data manually on some or all of your lines. And as you automate, you can switch it over to be automatic.

you can still get some of the benefits and you can still, that’s particularly useful if you want to do things like quality and inventory and traceability as well, but you still want to record your production even though it’s not automated. That works particularly well. So yeah, not having things automated is not a reason to not start, but it is something you’d want to look at relatively quickly to get the maximum benefit.

David (32:23)
Matt, those were very valuable insights you shared. think that’s a wrap for this episode of the IT/OT Insider podcast. So thank you, Matt, for joining us and also thank you to our listeners for tuning in. If you enjoyed the conversation, don’t forget to subscribe at itotinsider.com or go to mesmatters.com, Matt’s website, and leave us a rating because it really helps.

and see you next time for more insights on bridging IT and OT. Until then, take care. Bye bye.