Preventive & Predictive Maintenance

Wade Anderson:

Hey, manufacturing world. Welcome to another episode of Shop Matters, sponsored by Okuma America. The podcast is designed to talk about all things manufacturing related. Today, joining me in the studio I've got Mike Hampton from Okuma America and Andy Henderson from Praemo. Welcome guys.

Mike Hampton:

Yeah, thanks for having me.

Andy Henderson:

Thanks a lot. Happy to be here.

Wade Anderson:

All right, so let's do a quick introduction. Mike, tell us a little bit about yourself and what do you do?

Mike Hampton:

Yeah, absolutely. So I've been in manufacturing since 2010. I was actually on the customer facing side. Worked with a number of brands of machine tools, being lathes and mills of some of our competitors. And since 2018, I've been with the Okuma America and I'm strictly focused on business development for the aftermarket parts and service side. So, that's what I do on a day to day basis and the topics we are talking about today, I have an extreme passion for.

Wade Anderson:

2018, it seems like you've been with us like six months.

Mike Hampton:

I know, time flies when you're having fun, Wade.

Wade Anderson:

That's the way to say it. Jim King might listen to this, so-

Mike Hampton:

Perfect.

Wade Anderson:

Andy, tell us a little bit about you, what do you do?

Andy Henderson:

I'm Andy Henderson. I'm the CTO for a company called Praemo. Praemo does advanced data analytics. We utilize technologies like AI to help us make better use of data faster. And we focus primarily in the industrial operation space. My background, I'm an engineer, specifically a PhD in automotive engineering. My research, while working on my PhD, was an updated cutting force model when milling nickel-based super alloys. Somebody at Lockheed said, "Yeah, those are the really angry metals." They are, because they tend to, in most forms of machining, we tend to talk about things in terms of parts per tool, with nickel-based super alloys we talk about things in terms of tools, per part.

Wade Anderson:

Yep. Tools per part or minutes per tool.

Andy Henderson:

Yeah. And so, my research a number of years ago, it was around understanding cutting forces for the purposes of trying to understand tool wear and optimizing consumable cost and cycle times when machining those alloys. Then that was really interesting to GE power, which is where I went after that.

Wade Anderson:

I was going to ask, how did you make a transition from automotive engineering and studying nickel based alloy machining to AI type conversations?

Andy Henderson:

My entire career has in some way or another revolved around data acquisition and analysis. So my first role out of my undergrad, so I took a non-traditional path to my PhD. I finished a bachelor's in mechanical engineering and I went to work for Caterpillar doing diesel engine testing and development. There, it was all around designing tests- working with pre-production diesel engines in order to put them through their paces, do validation testing, and identify failures that may occur, root causes, and then working with the design teams to try and adjust to the design of that product to mitigate those issues before they're actually released to the customer.

Andy Henderson:

So it was a lot of specking sensors, designing test, putting sensors on engines, acquisition of that data, getting it into something that we could then go take and do further analysis on in order to link it back to, like I said, root causes or other failure modes that occur. Then going into the PhD, a lot of what I was doing during my research was designing test, acquiring data for the purposes of building a model to understand what was happening in the midst of the machining process. And then at GE, it was all around improving GE's, internal manufacturing operations. So working with folks on the shop floor to understand how are you doing things today and are there other technologies that may exist?

Andy Henderson:

Maybe it's a new coding on a cutting tool that extends the cutting tool's life that you can make use of in this operation, in this machining operation to a lengthen tool life, or run the process faster to cut cycle times down. So a lot of that was gathering data about what was being done today, so that we could establish those baselines and know if any changes that we made actually did make an improvement to the process. It was still a lot of data acquisition and analysis. But it's very manual.

Andy Henderson:

And so the AI piece was around, I was doing some online courses about AI and thinking, Hey, these algorithms are built to try and make sense out of that data and have the machine learn those relationships between-

Wade Anderson:

And kind of meld the two together.

Andy Henderson:

Yeah. And so it should be able to help us, because my manager at GE would often say, "Hey, another great set of projects, way to go." But how do we take this team of, I forget the numbers. I think we were in the teens as far as the number of people supporting internal operations. And so there couldn't be an Andy at every single facility that GE had around the world. So how do we take what I was doing and scale it? So it was a lot of, well, we can definitely automate that data acquisition, which is where I got to. I had already known Okuma for a long time because of my research, even I was using an Okuma in my research, but it was through that of how do we get to the data and pull it in, in an automated way? It was where I got even more heavily engaged with you guys because the OSP control really makes that easy, right?

Andy Henderson:

There are various things that you guys have done over the years that makes it easy to get a lot of really valuable data off the machine itself without having to go spec out and put new sensors on and put expensive data acquisition systems in place.

Wade Anderson:

One of our control engineers used to say that we had over 9,000 sets of data points that you've got access to through the OSP control. Most people can't scratch the surface on that, but there's so much under utilized information there.

Andy Henderson:

Yeah. And so that's AI, right? That's where AI comes in. Is how do you make use of stuff like that? Because the traditional approaches are going to be, you make some graphs that show 10 of those 9,000 points trending over time, and somebody sits while they're sipping coffee, and that doesn't look quite right. Let me go figure out what's happening there. But that's where, we talk a lot about trend detection and anomalies, and that's a foundation of what you start from as you're looking at that data, looking for trends and anomalies, because that's the person is sipping their coffee is doing too. They're looking saying, "Hey, that line that's been going in the wrong direction." Or, "That was a weird spike in that data item. I got to go figure out what that is."

Andy Henderson:

So, but he can't do it on all 9,000 data items for one machine, right? Because he's got 20 machines maybe, as in the case of GE, one facility had 500 different assets that we were trying to capture data from. So multiply that by 9,000 and that's an unwieldy data set. So you apply these tools that are doing some aspect of what that person is doing. I keep saying guy, it could be a gal, right? What that person's doing as they're reviewing the data. So, letting the analytics engine run on that data and bring those out so that those people can be doing their normal jobs. Be out on the floor trying to execute on projects, and then the engine says, Hey, there's something weird related to these particular data streams on this particular asset. Maybe you want to go take a look.

Wade Anderson:

Yeah. Very cool. So Mike, you come from a background of plant operation and production. Tell me a little bit about the importance of predictive maintenance and preventative maintenance. What's the difference between those two?

Mike Hampton:

Yeah, absolutely. I think the ultimate goal, whether you're looking in any manufacturing facility is to be focused on uptime. And both of these have to do with uptime. But that the real difference between the two is preventive maintenance is performing maintenance that is scheduled, controlled, and measured to restrict the possibility of having unforeseen downtime. So it's saying, "I'm going to bring this machine down on X date, I'm going to change these components, I'm going to focus on these areas, and then bring it back up so the machine is running when I want it to run."

Mike Hampton:

Predictive maintenance is slightly different. But it's using data to predict when items are getting near the point of failure. So whether that's focusing on vibration of a spindle or different wear items or the wear of a ball screw, it's using data to say, this component is getting close to its failure point, let's go ahead and schedule the machine to come down. Let's replace that component so in the future, the machine will stay up. So when our maintenance crew isn't here in the middle of the night, it's not going to fail.

Mike Hampton:

So both of these really, preventive and predictive maintenance are both going to help the machine run. We really practice preventive maintenance in everyday life. I like to compare it to household items. So why would you perform preventive maintenance? There's a lot of reasons, but first and foremost, machine tools are very expensive and you want to protect your asset. A good comparison is your car. I mean, cars are very expensive too. Most individuals have them. Why do you change the oil in your car? You change the oil in your car so the motor continues to run. If you don't change the oil, you're going to lock up the piston and the block. You're going to have a huge expense to overcome and it's going to be extremely frustrating. So you schedule your maintenance, that's going to protect your investment.

Mike Hampton:

Another example is let's just say your HVAC system in your house. Everybody changes the filters to their air handling system every three months. Why do you do that? You do that because you want your system to perform as good as it can and it's also going to let the air flow through at the proper rate, making that air handler last longer. So another example I like to point to is a fire alarm in your house. So individuals change the batteries when it starts to beep. You don't do this after it goes completely dead, you do this when it's beeping, because you know, it's getting close to the point where it needs to be changed. Why do you do this? Because it's a safety factor.

Mike Hampton:

So in everyday life, people practice preventive maintenance and it doesn't change with the machine tool, you need to apply it to a machine tool as well for all of those reasons.

Wade Anderson:

Okay. So the efficiency side of it, that's such an important factor from manufacturing. For people to be the most profitable and most competitive they can be in the marketplace, their process has to be as efficient possible. So the maintenance aspect of it tends to be one that is kind of low hanging fruit. I think a lot of people overlook it or they put it off. I'm on a school board and we're dealing with maintenance on some school buildings right now. And some of the facilities that we're dealing with, what we're dealing with is the fact that because of budget restraints and just different issues, well, let's forgo it. Let's push that off another two years and then let's reevaluate the roof. Well, okay, let's push it off another two years and then we'll reevaluate it.

Wade Anderson:

Meanwhile, that roof's still leaking. There's still issues going on. That's creating other processes down the road that we're having to deal with from a maintenance standpoint. So at some point, a guy used to tell me, pay me now or pay me later, right? At some point you've got to take care of it. So what tools do you guys have in place that gives tools to customers to be able to handle preventative maintenance in a smooth transition?

Mike Hampton:

Yeah, absolutely. So one of the things that we've launched recently is the CARE preventive maintenance kit. And what CARE stands for, you might know or you might not. But it's an acronym. It stands for Constantly Available Resource Experts. And that's been around since IMTS 1993.

Wade Anderson:

I don't know if that's a Japanese thing or Okuma in general, we've got acronyms for acronyms, don't we?

Mike Hampton:

We do. But this acronym, it's a good one. And not only do we care, but we want to express that we are experts. No matter what situation arises, whether it's a field service issue, an application engineering, a service parts, break fix issue, a mechanical or electrical exchange issue, whatever the issue is in the field, we want to supply a solution. And that's what Okuma CARE is. That's why we launched the Okuma CARE preventive maintenance kit. What this kit is, is it comes with all the necessary components to do a full preventive maintenance, a robust PM on your machine tool, and it's one box. So you get one box you have, if it's a lathe, it's going to come with all the necessary filters, the way cover wipers, the saddle wipers. All of the necessary wipers that need changing along with other components.

Mike Hampton:

If it's a machine tool, it's going to have all of the filters, the O-rings, the Springs needed for the through pin and the through pin is going to come in the kit as well, because ultimately that's going to protect the heartbeat of your machine, which is the spindle. So this kit is delivered to our customers. They open up the box, they can perform the PM, it's going to protect their mechanical, electrical, their lubrication, their coolant systems.

Mike Hampton:

At that point in time, you'll do a full machine inspection, just like if you were to take your automobile to the dealership. They're going to change the oil filter, they're going to change the oil. They're going to change the air filter. They're going to do a full inspection. And they're going to say, "Hey, you might need new wiper blades." Our service technicians are going to do the exact same thing. So when that car pulls out of the dealership, it's in like new condition and so will your machine tool. It's the exact same thing.

Wade Anderson:

How the preventative maintenance scheduled? Is there a list of timeframes for each machine that is predetermined or is that something based on the customer utilization?

Mike Hampton:

Yeah, it's based off of utilization, but there's obviously recommendations that come in your operation maintenance manual, that's provided with the machine and it's also tracked in the control. So we have great functionality in our new controls that know exactly how many hours have been run and we can set periodical maintenance interval suggestions, and it'll actually set an alarm on the control, telling you when you should perform your PM.

Wade Anderson:

Okay. So if a customer has a unique manufacturing process and the preventative maintenance that he's wanting to do, maybe there's a few items beyond what comes in the predetermined kit, is that something he can incorporate and work into his personal preventative care kit or preventative care process?

Mike Hampton:

Yeah, absolutely. You can tailor it in any form or fashion that you want to. So you have the ability to go into ECO suite on the Okuma control. There is a maintenance monitor piece of software. You can go through the whole list of pre suggested periodical maintenance, and you can add to that.

Wade Anderson: Okay. So Andy, how does the anomalies in the AI side, how can that tie in from a maintenance standpoint in determining the health of the machine and the longevity of the process that you're working on?

Andy Henderson:

Yeah. I'll come to the question, but I wanted to talk about, there's always a question of, "Data's great, what do we do with it?" One of the things that I've seen and you touched on it, right? Or both of you guys touched on it, right? There's the calendar base where you can say, every three months I've got to do something or every six months I've got to do something. Well, a lot of machines are tracking hours. So you can be more, I don't know, deterministic is the right word, but where you can say, now, instead of just going and doing maintenance every three months, let's do it based off of the number of hours.

Andy Henderson:

A lot of people, when they're thinking about trying to get at data, for some reason, they say, well, you know, AI is the big buzz word. That's the new thing that we're going to go after. Well, and I think that should be part of the strategy in the meantime, while you're working towards that strategy. Going and doing things like collecting all of those hours and using that to trigger maintenance work orders in a CMMS system can be a very, what's the word I'm looking? Of value. It can add a lot of value to a maintenance program just having that. I've seen it, right?

Andy Henderson:

I've seen what people are doing is they're looking, at first, they have a calendar of all the assets and when they need to go touch them and then they go and they find out this particular asset actually hasn't done anything in the last three months, but we're going to do the PM anyway, because that's what the calendar says we should do. And then they move to a system where there's a person walking around with a clipboard and they're checking all the controls saying, "Well, what's the runtime hours on that?" They write it down and they use that to make their decisions.

Andy Henderson:

Then you can say, well, let's string a line or let's figure out how to get that data off of that control and store it somewhere centrally. What that does, right, is now it opens up all of those 9,000 data points that we can start turning on and processing. It makes them available to us. So you can use that as a value proposition to expand into this bigger opportunity that you may have by applying advanced analytics on top of it.

Andy Henderson:

But your question specifically was how does AI or how does an advanced analytics engine help in the maintenance piece? How does it help in the preventative maintenance side of things? A simple example is like, I was talking about trending, right? Trend and pattern detection. So, if you're going to put an accelerometer on a motor casing near where the bearings are, you're going to have to monitor that in some way. And what are you looking for? You're looking for a trend in some direction or another. That trend is what you're going to then base your understanding of the health of that bearing off of.

Andy Henderson:

So, what am I talking about? So, the accelerometers on the motor casing, you're monitoring the vibration of that motor around the bearing. Probably in that case, you're going to be doing something called a 48 transform. You're going to be looking at the frequency content and you know that the bearing is supposed to have a vibration signature at some frequency. Then over time, you're going to see some other frequencies start to creep up. And it's tracking the trend of those other frequencies showing up on that that 48 transform. It's the power spectrum.

Andy Henderson:

When they start to show up, then you can trace how much, the magnitude of it. What you would want to know is, okay, what's the magnitude of that when this thing is reaching the end of its life, when it needs to be replaced? So an AI engine can take that data and learn those patterns and learn when in the past, action has been taken to replace that thing, or maybe it's something like you're talking about a moment ago with lubricants changing the lubricants out. Because you see the same exact thing. There's going to be a trend to some point, a lubricant is going to be changed. The filters are going to be changed. There's going to be a reset, and it's going to keep on trending. We see this in tool life, right?

Andy Henderson:

It's well known that the power, in most cases, power increases as a tool wears out and you see the same exact pattern there. So the AI engine is looking for those patterns in the data, looking for when those resets happen. And it's saying, I know, based on the trajectory of this trend or I've learned based on the trajectory of this trend, and when resets have happened in the past, when action needs to be taken. So it can be that thing that says, okay, at this point, let's trigger an alarm. Let's make a notification that says, actions should be taken to go replace that cutting tool, should be taken to go change that filter, should be whatever, right?

Andy Henderson:

So it all makes sense and people say, well, we do that anyway, right? What do we need an AI system to tell us that for? Well, the AI system is taking into account all of the other variables as well. Because in the case of a cutting tool, it might process 10 different products and each one of those has a slightly different impact on the life of that cutting tool. So it's not always a set threshold, a single alarm, but it's a combination of a whole bunch of factors that go into that component's life and so the AI tool learns to understand those and incorporate those into its projections as well.

Andy Henderson:

I feel like I may have gone off the deep end there a bit. Hopefully I brought it back. Did I answer your question?

Wade Anderson:

Yeah. All interesting stuff. It's neat to see how the technology changes and can be incorporated to a manufacturing process, to make you more efficient, make you leaner on how you attack parts and the type of work that you're doing. Again, always say the more efficient you can be, the more profitable you can be, or the more productive you can be.

Wade Anderson:

So what are some working examples that you've seen from a machine tool standpoint, where you've utilized Praemo and changed the way a manufacturing process is?

Andy Henderson:

We have a customer that we like to talk about, and this was simple. What we were doing relatively simple, or it's simple from a high-level view. It was learning what typical cycle times of the machine. We often do this, right? Where a software vendor comes in and somebody in operations says, "Well, just tell me when my machine is sitting idle. That's what I really want to know when it's sitting idle for too long." Software vendor says, "Okay, well, what does idle for too long mean?" "I don't know, 10 minutes. If it sits longer than 10 minutes, let me know." "Okay, we can do that."

Andy Henderson:

We set an alarm, machine sits idle for longer than 10 minutes. It sends a text message. Great. Well, then they get flooded with text messages because 10 minutes was too conservative on one side and then something escapes and they don't get notified about it because it was too long on the other side, right? So, some parts of their operation, 10 minutes is fair, right? It takes 10 minutes to do a change. In some other parts, if it takes 10 minutes, it's disastrous. And so what the engine learns in that case, right, is it learns to recognize the nuance between those two and set limits accordingly. All on its own without having to have the person say, "I don't know, 10 minutes."

Andy Henderson:

So providing this to one of our customers, they were getting notifications about cycle times that were too long and this launches an investigation. It says, okay, well, that's odd. The cycle times are longer than expected. We wouldn't have necessarily known it otherwise because it would've gotten hidden in the noise of everything else. But cycle times are too long, let's go figure it out. It launches this investigation that raises a few things. One, we should adjust the size of our kanbans. We need to tweak how much is sitting there ready to go, or that's available at any point for the machine to consume and use.

Andy Henderson:

The other thing is, well, there was one of those changes happened over the duration of a month. Well, actually during the duration of that month is when we had different operators flowing through. So it's not necessarily a skillset thing, but maybe we just need to keep people cycling around to keep them refreshed in different parts of the operation, so that we can keep parts moving effectively. So does that answer your question, is that what you were looking for?

Wade Anderson:

Yeah. I was really looking for what are some customer examples where you guys have installed Praemo, you're doing some AI analysis and what does that look like for a typical manufacturer that has never utilized it? So I know you've got some installs on the East coast here on some Okuma equipment, what's kind of the low-hanging fruit that you see that customers can realize the ROI on that investment and incorporating that into their production system?

Andy Henderson:

So, so it's things like what I was just talking about. That's probably one of the easier ways to do it, right? To get those notifications. Because it doesn't tell you exactly what the problem is immediately, but it tells you the area to go look to say, "Oh, that's weird. I need to go address it." When you do address it, so that particular customer takes... They support MRO businesses. So they get phone calls. I need something in 24 hours and if they can't deliver in 24 hours, they lose that sale. So they're really focused on taking the time from order comes in, to the time it leaves a shipping dock down from, it was five to seven days, down to 24 to 48 hours. By addressing those things like I was describing, it enables that.

Andy Henderson:

Now, I can't say that Praemo solved that entire problem, but it was a contributor, right? It was part of a holistic approach that really helps you drive that overall, order comes in, leaves a shipping dock. It helps you condense that down. So that's an example. From the customer standpoint, our philosophy is it should be super easy, right? So, when we engage with customers, it's all about, give us a dataset and don't worry about trying to clean it up. Don't worry about trying to harmonize the data, have everything make sense. We want it as raw as possible.

Andy Henderson:

In fact, lots of times when people try to clean up the data for us, it ends up adding some amount of human bias on top of it that we have to then unweave, right? We have to go back and figure out what they were doing. So, raw as possible. Therefore spend as little time as possible just getting the data out and getting it over to us. We've built razor to just churn and process and find these interesting points in the data and the time span of the data that we can then go back and say, "Here's something that Razor is identifying as being interesting. Is it truly useful to you in this case?" Yes or no? If it's no, then we keep tuning and say, ignore that. If it's yes, then keep tuning to say, keep finding those things for sure, now let's go find some other stuff that is useful.

Wade Anderson:

Now that's something that I realized I don't think we've touched on is what is Razor? So I've been talking about Praemo a lot, but Razor is your actual product, that we would typically see on a machine tool, correct?

Andy Henderson:

Yeah, Razor is our analytics engine. We've got a bunch of data scientists and software application folks that are building that engine. It would sit, you mentioned edge and cloud at some point, all of our customers to date, use Razor in the cloud.

Wade Anderson:

Okay. Are there any connectivity concerns that you see from utilizing the cloud? I know a lot of your higher end aerospace shops, for an example, won't let some of their machines be connected to go out to the cloud, it's got to be internal. Do you guys run into situations like that?

Andy Henderson:

Yeah, there's that concern which stems from security. In fact, I have some experience from a prior life with ITAR. And a lot of it revolves around ITAR or the, the U.S. government, what is ITAR, what's that acronym for? Everybody's got their own acronyms, but I forget, right? It's making sure that foreign citizens don't have access to that data. Is what it kind of boils down to. And so that's a thing that drives it.

Andy Henderson:

Another thing that people talk about is latency. So I've got this high-speed accelerometer data, I want to be able to process it as fast as I can sending it up to the cloud, letting something in the cloud process it and send results back down. That takes too long. So there's two different reasons why somebody may ask us for an edge approach versus a cloud approach. Edge being closer to the floor, to the shop floor.

Andy Henderson:

Do we run into those? Yeah, we run into them, but it's a lot about finding where value can be addressed today. So, can we work in all of those situations and work around and figure out ways to provide value if security is a concern, if speed is a concern? Yes, we can. But do we need to start there right now? Is there something else that we can do where there's a lower barrier of entry that we don't have to solve all those problems upfront? Yeah. We can provide value based off of those too.

Andy Henderson:

So, I guess the messaging would be, don't get hung up on those challenges. Let's go-

Wade Anderson:

Explore it?

Andy Henderson:

Yeah, let's go explore it and let's go realize the value that's available to us today. And we can lay out the roadmap to be able to put all those pieces in. A lot of our opportunities to date have been in that, right? Where we've been able to say, "Yeah, we can deal with security, but what about the stuff that's not ITAR controlled? Or what about things that don't require high speed data transfer and analysis?"

Wade Anderson:

Okay. Mike, how does Connect Plan tie into the predictive, preventative maintenance, understanding anomalies, as companies like Praemo are able to detect them? What would we do from a connective standpoint on the control?

Mike Hampton:

Yeah, absolutely. So I'm just going to back up a couple steps. Connect Plan may not be widely known to all. So what is Connect Plan? We started this conversation by talking about preventive and predictive maintenance. We talked about the CARE kits, which are the physical parts or the wear items, the components to perform a robust PM. Andy got into the data section, and then he was talking about walking around a manufacturing facility physically with a clipboard to collect data off each individual machine. So what is Connect Plan?

Mike Hampton:

Well, Connect Plan is a smart factory. It is a machine monitoring software that Okuma has produced. And what you can do is you can remotely see your whole factory in a visual layout. So you can open up this great visual layout that shows all of your machines. So you're physically looking at your facility. And you can be fishing at the Wando river in Charleston, and you know exactly what's going on inside your manufacturing facility. So, you know every time that every machine is running, you know every time that it's not running. You know when a tool change is taking place, you know when a part is being unloaded or reloaded, you know the maintenance that has been performed on your machine, you know every alarm that has gone off.

Mike Hampton:

So it just gives you all of these different data points to help you increase utilization in your manufacturing facility, which is just a great tool.

Wade Anderson:

It seems to me like there'd be a good harmony between Connect Plan and then Praemo. From Connect Plan, giving you the overall the big picture view of the health of your manufacturing process, what's taking place at any given time. And then when you needed to really deep dive and understand and find these anomalies, and as we talk about data, always think about the talk about the 787. Is it a terabyte of data that it pulls off every flight? But one of the engineers that we talked to said, "While that's neat to talk about, we pull all this amount of data, we're only looking for a small set of data that we're actually going to make an executable plan off of." So I kind of see these two pieces really tying in well together.

Mike Hampton:

Andy talked about wanting the raw data. And Connect Plan, what it's doing is it's pulling these data points and sending it to a server. We can remember up to two years of data. So with that said, we could export it, provide Praemo with the raw data from two years and that's a great starting point.

Andy Henderson:

And we love those. Because when we talk about what's our core competency, it's all about the analysis. So we tend to look for platforms, I don't know if you guys won't use that word, but platforms like that, that do the data collection putting it all in one place, because that's a single point that we go to as opposed to try and create adapters for all the disparate systems that may exist on the shop floor. So, we love to do those types of partnerships because we just don't consider the actual extraction collection of data at the very edge to be part of our core competencies.

Wade Anderson:

All right. Pretty good. What have we missed? Any key topics that we wanted to hit that we didn't?

Mike Hampton:

Well, I'll just say we mentioned Connect Plan and how it's a machine monitoring software. One thing I would just like to let everybody know is this can be used with non Okuma machines as well. So we can still connect to a Mazak or Maurey or whatever type of machine it is and collect data off of that control in addition to obviously Okuma machines.

Wade Anderson:

That's an important piece because very few shops only have one brand of machine tool, right? Every place you go, there's multiple controls, multiple styles of machines. So that's a very important piece. I would assume you're the same Andy?

Andy Henderson:

Oh yeah, for sure. And that's key to what I was saying a moment ago as well. That's why we like working with those types of platforms, just so we don't have to know about all those different nuances. I kind of smirk because I remember having some conversations with some Japanese counterparts a few years ago around having lots of different types of machines. Okuma does some great things for making data available. Part of our challenge is that there's a lot of other machines out there too, that we have to tap into. There's value from looking at a particular asset. But then that value is probably multiplied when you start looking at chains of assets and a whole facility of assets and merging that data together to tell you a more holistic story is extremely valuable.

Wade Anderson:

So Mike, somebody wants to learn more about Okuma CARE kits, anything like that, what's the best way to get in touch?

Mike Hampton:

Yeah, absolutely. Contact your nearest distributor and, or go to okuma.com. If you go to okuma.com under the Customer Care tab on the website, you can see the navigational point that goes straight to Okuma CARE kits. Or you can go under Technology, and Machine Monitoring, and that'll take you directly to the landing page for Connect Plan and you can actually fill out a call to action or a CTA requesting any additional information. It'll be sent to distribution in Okuma America and we'll get back to the customer as soon as possible and answer their request.

Wade Anderson:

All right. Andy, how about you? Somebody wants to know more about how they can utilize AI for their processes, how do they find you?

Andy Henderson:

So there's our website for sure. Praemo, P-R-A-E-M-O.com. We also have a company LinkedIn page, a Twitter account. I have my own personal stuff out there as well that that people can find if they're looking through those social media platforms. I think those are the better ones.

Wade Anderson:

All right. Guys, appreciate your time today. Everybody out there listening, we appreciate you taking the time to listen to us. If you got any thoughts or questions, please reach out to us. You can find us on the Okuma website and I'm on LinkedIn and all the famous social media sites as well. So until next time, thanks again.