078. Idea - Understanding Overall Equipment Effectiveness (OEE)

Mike: 00:00

We have to have a more efficient plant and we need to be competitive. So we need to constantly measure the efficiency of our plant and keep reinventing what we're doing and then hold that intellectual property close with the proper cyber security so that we maintain a very competitive manufacturing environment on the global scale. 

Chris: 00:20

Welcome to EECO Ask Why. A podcast that dives into industrial manufacturing topics, spotlights to heroes to keep America running. I'm your host, Chris Grainger, and on this podcast, we do not cover the latest features of benefits on products that come to market. Instead, we focused on advice and insight from the top of minds of industry because people and ideas will be how America remains number one in manufacturing in the world. 

Welcome to EECO Asks Why. Today we have an idea episode where we're going to be talking about OEE and to break that down for us, we have Mike Caccia, who is a Product Manager with Electrical Equipment Company. How you doing, Mike? 

Mike: 01:01

I'm doing fine. Thanks. 

Chris: 01:03

Good man. I'm looking forward to walking through this a fun topic. So maybe get us started off, man. What is OEE? 

Mike: 01:10

Well in its simplest terms, it's really a ratio of good parts produced versus what could have been produced given the ideal conditions. OEE stands for overall equipment effectiveness. And it's a simple equation: availability x performance x quality.

So this calculation takes into account the uptime and downtime of the machine, which is availability. Performance, which is, you might say, the throughput of the machine or the speed of the machine. And then also quality, which it relates to does the product meet specifications for the customer. It's kind of like a three-legged stool analogy where the three legs of the stool are performance, quality and availability. If you address one leg of the stool without addressing the other two, you're stool is not going to stand very well. 

Chris: 02:07

Yeah, you're right. You're all over it, man. That's a great analogy. And it's funny back when we had the shops, man, we had a version of this. We called O T E. We call the overall technician effectiveness and we factored in a couple of the same factors performance availability.

How from a task standpoint we're doing on a job to just, it was kind of cool to make a metric somewhat like this, that we could just measure and put out there and let everyone compete against. So, but OEE. Very important. Great explanation. Thank you for breaking that down. So why is the understanding that OEE and the factors associated with and important for manufacturing? 

Mike: 02:48

It's really important, particularly for manufacturers in the USA because in the U S labor is a little more expensive than maybe a lot more expensive depending on which country you compare us to.

And yet we're competing oftentimes in a global market. So places like Asia, where the labor costs say in China might be less expensive, the main way that we can compete in the U S with the Chinese manufacturing is to have maybe a higher labor cost, but more efficient manufacturing. That's the crux of it.

And the efficiency of the plant is really a juggling act on the part of the plant manager who really manages all the resources in the plant. If you look at performance, quality and availability and kinda dissect who's responsible for them, they span across departments in the plant. And that's where it becomes the juggling act.

For example, performance is normally influenced by the operators and production. So to put it in a simple term, they're always trying to crank up the speed without necessarily knowing the ramifications of the other two legs of the stool. And then quality is really related to basically sales because if your quality doesn't meet specifications, you're going to impact your sales.

And in a plant there's two main things we worry about for profitability: lowering costs and improving sales. And a plant doesn't have a whole lot of the impact on sales because they don't have sales force, but they do have impact on the quality. So it becomes really important.

And then the last leg really is availability, which is normally the focus of the engineering in the plant, in the maintenance department. So when you're the plant manager, you're sitting there juggling production versus availability of equipment. And then when your customers complain about product not meeting spec, that's the quality aspect.

Chris: 04:48

Right. Very good, man. That's great. Now from a plant management standpoint, I know you're in a lot of plants out there, Mike, on the road, supporting end users. Do you have any examples, real life examples where you've seen OEE factor in? 

Mike: 05:02

Absolutely. I'll give you a simple example. If you take efficiency of a plant and you increase that efficiency by 1% in a $50 million per year plant running at 50% OEE, that's $1 million per year in increased production. That's making money.

The resources that you're using are typically counterproductive if production speeds up the line it could be at the expense of the, for example, the equipment breaking down sooner. The gearbox may be it isn't designed to run at that RPM. You crank the speed up. You're able to make product for some time, but then prematurely, the machine breaks down and now you have an unplanned shutdown.

So you may have shortly improved production performance, but at the expense of availability. So that's that two legs of the stool competing against each other. The other aspect of is if you speed up the line, you might start affecting quality. And then you're not meeting spec and you may get rejects from your customer.

If you're lucky you detect the rejects before they leave the dock door. And your customer doesn't see that and it doesn't affect your sales. But if you do detect that defect and well now you're going to start effecting your rejects and your scrap rate, which costs money. 

Chris: 06:26

Right. Each area is so dependent and important on the .Other striking that balance is is key. So thank you for that. And when plants are talking about, there are a lot of KPIs out there and there are a lot of things that people measure from a performance standpoint, anything stands out when you're talking about production data associated to OEE, that you hear plants and management talk about? 

Mike: 06:52

There's three main areas that come to mind. There's production data. There's also event data. And lastly there's fault data. And when you think about production data, it's more about the actual product and how many counts you make. Then maybe the quality or the specifications or the properties of the product. Event data is like looking at the lifespan of the machine during an eight hour shift and trying to figure out, well, how many start stops were there on the line and how many times during the course of the day was the equipment available and ready to run, but not actually running. 

And then fault data is an area where more often than not it's engineering and maintenance that care about the fault data, because the machine stopped for some reason, and there's a red light somewhere, or an alarm buffer that comes up and says, here's why I stopped with some sort of text or something.

Chris: 07:48

Right. Right. Now on the event data. From those reports how could a plant or industrial end-user use that data there on the event report standpoint to make a an improvement we'll make let's go ahead and say a significant improvement to efficiency? 

Mike: 08:07

Yeah. Event data reports are extremely good at illustrating a timeline or the life of the machine during a shift. And it's really important to look at runtime and stop time and to quantify the reasons for the stop. Was it an operator stopping it because it was the end of the shift. Or was it stopped time because of some other reason. That's really important to figure out so that you can study it and make that line more efficient.

If you have a good OEE system that's measuring holistically all day the three legs of the stool on your plant. You can really use this to capture all these start stop conditions and even report that automatically. And then study that later to keep tweaking the efficiency of the overall line.

The key thing that you want out of there is probably a reason code to be quantified with every time it stops. That way later, you can really discern why the machine's not running. Reason codes could be anything from an operator deciding to stop the machine or it could be, you ran out of raw materials coming into the machine.

It could be the machine was starved of product because it's in a main line where one machine feeds another. Every reason code could point you to another area to improve. If you ran out of raw materials, it could be just an inventory problem on your raw materials. And that that would be addressed by the plant manager in one department.

If the machine stopped because a motor tripped on overload, he's more than likely going to pick up the phone and call maintenance. So you can see where he can really help you find a root cause. And then address it, into the right department because the plant managers, again, responsible for operations, the labs where they test the quality, and maintenance and engineering. And it's a juggling act. 

Chris: 10:05

So here's kind of what I heard, man. It sounds like those reason codes when used correctly could potentially be, an analogy could be like a treasure map. You want to get better. You want to find those areas to go lead you to make more money as an operation.

That's your guide. Learn from that data. And then invest time and resources in the areas that are holding you back to get better. So, but if you don't measure it, you can't manage it. So it's great stuff, man. And you also talked about fault codes and fault data. Why do we need to care about that?

Mike: 10:42

So fault data is the machine's ability to really capture a machinery issue as to why the machine stopped. There's all sorts of fault codes like PLC faults, there's drive faults, network faults, and then there's also process faults. But the fault codes give you an example, if you have a motor overload trip on let's say a fan or a pump, some people, if they see a red light will quickly go to the red light and replace that component. And you start the machine back up and it goes right back to the same fault. Well, that's not doing very well on your first time fixed rate in a maintenance department, you really want to find root cause and fix it the first time.

So really it's important to capture all the faults associated with the motor and have an understanding of the interaction of everything. You could be, for example, it could be the operation of the machine where you're trying to run the pump with too much back pressure causing the motor trip on overload.

It's really not a problem with the motor. It's really not a problem with the overload relay cause overload relays just protecting the motor. Could be just an operations thing. On the other hand, it could be that the motor is actually overheating because of some reason like the motor needs cleaning or something.

All of these things you can see where it troubleshooting, something could span again, operations or it could be an equipment failure. Discerning which one is the root cause, it can be difficult if you're only looking at one leg of the stool. So in the case of what I just mentioned understanding the operation or the production side and the maintenance side of the equipment is really important to look at both at the same time. And OEE and fault data is really helpful at finding that root cause quickly. And enabling the maintenance team to fix things the first time, because they have all the right data at their fingertips. 

Chris: 12:40

Right, right. Very good, man. That's great. So let's go into plant for a minute. Great, great high-level discussion on OEE. What it is and how to use some of the data to make better decision. But from a tactical standpoint, you have a industrial end-user out there right now and you're listening and you don't have anything in place to measure OEE, where is a us a natural entry point to get us going and who would own that? 

Mike: 13:07

Yeah, that's a really good question. I dare say most of the plants are actually measuring OEE and a crude form in some shape or fashion and they probably just don't even know it.

Most people care about efficiency. And have their own methods of measuring things. Oftentimes it could be as crude as a form on a clipboard that is filled out by hand collecting data as to why things stopped or what production recipes were and what the outcome of those recipes were for that batch.

Or it could be that that data is collected on an Excel spreadsheet. So that collection is probably going on in some crude way or another, but it's up to the discipline of the personnel. And sometimes it goes down to the individual's capabilities. And what we want to do is want to turn that good practice into more of a scientific way and kind of a robotic way so that we're collecting the data 24 seven without the need, or we lessen the need for human intervention as must as possible. So it's a scientific gathering of data and it gives us better data to base our root cause analysis on. That's the key. 

But where are we with start is I think nowadays we're in the year 2020, I think we can assume there's a PLC and most of the machines doing some automation on the machine and is probably an HMI or SCADA on the machine. I dare say most of the PLCs have the ability to collect all the start-stop reasons and all the fault codes that we're looking for for the OEE system.

It's just a question of getting that data up into a software package that then can report on this across not just machines but across the production line. So I would say turn to your PLC and your software vendors. And find out how you can leverage the existing automation you have in your machine and get better data out of it for the sake of measuring efficiency and then making smarter decisions to run your plant more efficiently.

Chris: 15:10

Right. So, I mean, inside of a plant, are the process engineers typically who were involved with the OEE  from an ownership standpoint or is it management? Where do you see typically owns the process, if you will, of establishing the OEE parameters and things like that?

Mike: 15:27

Yeah, that's a really good question. Everyone is measured differently in the plant, depending on the department. I think we all use the common term KPIs. We're all measured at the end of the year on our department, whether it be production, quality, or maintenance or engineering. We're all measured on that.

And each one of us do it in our own way with our own tool sets. OEE helps us to kind of provide a holistic viewpoint across departments for the plant manager, but then serve each department individually. So that the short answer is everyone needs to care about OEE. But there probably have more effect on one leg of the stool versus the other.

Chris: 16:08

Okay. Well, thank you for that, Mike. And we hear a lot of buzz words out there. Smart manufacturing. Industry, 4.0. The IIOT. All these new devices so much data's moving. So it's an exciting time in manufacturing. So how has OEE being impacted by some of this? 

Mike: 16:28

So I'm sure everybody's heard of IOT and you're looking at the field devices and the PLCs and HMI that are in your machine. Most of the field devices are becoming more and more intelligence. With more and more data available. And it's not a new thing over the last 10 years, 20 years, we've been collecting an awful lot of data in manufacturing.

The real challenge is though is getting that data in some form factor or in a data pull, if you will, and using tools to actually utilize the data to make decisions. That's been the challenge over the years. But with industry 4.0, smart manufacturing all the new things coming out with the open standards, it really is becoming possible that all of this data, as we get more and more of it down to the field device level, we're actually able to get that up into some very sophisticated software tools and make really good decisions quickly. 

OEE, and it's been measured for quite a while, but what you find is depending on what vintage of OEE that has been deployed in the market, your data is months old, maybe weeks old, maybe days old. That's oftentimes the case. With industry 4.0 smart manufacturing though, the prospect is that your OEE measurement will be in real-time meaning it's minutes old.

And that means you're going to measure three legs of the stool and understand as you're running the machine instantly the effect that you're having, when you make your changes or you address issues, and that makes it so that you can really fine tune and make the best efficiency overall quickly. Minimizing your losses. Improving your profits. That's where we're headed with smart manufacturing. 

Chris: 18:18

Yeah, man. It's exciting times isnt it in manufacturing out there, that's for sure, right? 

Mike: 18:23

Yeah. It really isn't. The other thing that is not to be underestimated as you've heard, maybe the buzzword digital twin. For some plants, that's kind of a pipe dream, but there's actually like I'll give you an example, like an automotive line or an air base line, or some of the leaders in technology and in manufacturing, digital twin is actually a thing that they're actually deploying right now.

And what it means is you're digitally creating a simulated machine and production line and able to run the machine. Before you actually run the real machine. You design it. Try to get most of the bugs out of the system. And OEE is a part of that. So you target a certain efficiency of the overall line.

You can simulate the OEE in that digital twin, but then when you make the real McCoy. The real machine and run it. You put an OEE system on the real McCoy and that data that comes on the real machine gets fed back into the digital twin. And now you can make your digital twin model much more accurate and actually make changes in the digital model before you actually make the changes in the real machine.

So the whole thing just gets accelerated. And I'll give you an analogy about this, that everyone will understand. When you get your first iPhone. And it's rev level one or zero coming out, right. You get your new phone and you start using it. And your phone typically has bugs in it. The software, the new firmware has bugs in it.

Well, the whole time that you're using your iPhone on rev one, Apple is collecting data on those bugs and they're writing software before they even ship the next rev level. So they're collecting data on the machine and rewriting software. The next patch oftentimes is available before you even buy your phone.

You'll get a phone that was manufactured six months ago. You'll pull that out of the box and you'll fire it up and it'll automatically go pull patches down that were already written to fix bugs. And that's retail kind of example of how the data collection cycle and modification of the digital twin becomes reality. 

Chris: 20:34

Right. That's great examples, Mike, and digital twin that's exciting stuff. It's definitely fun talking about that. And when you get to see it in manufacturing and see it at these end users you can see the power behind it. So you've unpacked a ton of information for us around OEE today.

This has been great. We always like to wrap up EECO Asks Why with the why, where we get down to the purpose. So the plant managers that may be listening, Mike, or someone in a facility themselves, if you had to put the why behind OEE and why it's important, what would it be? 

Mike: 21:09

It really is so that here in the U S and manufacturing we can compete on a global scale because skilled labor is a real challenge, but in the us, we have some of the best educated people in the world, really from our universities. We need to leverage that labor. They're going to get paid more because we have a higher standard of living, but we have to have a more efficient plant and we need to be competitive.

And so we need to constantly measure the efficiency of our plant and keep reinventing what we're doing and then hold that intellectual property close with the proper cyber security so that other bad guys are not stealing our stuff. But we really need to think about it hard so that we maintain a very competitive manufacturing environment on the global scale.

Chris: 21:58

No doubt, very important stuff. And you don't measure it, you can't manage it. So, I mean, this is great stuff, Mike. You've done a wonderful job of unpacking OEE for our listeners. And we thank you for your time today on EECO Asks Why. 

Mike: 22:11

Thank you. 

Chris: 22:13

Thank you for listening to EECO Ask Why. This show is supported ad free by Electrical Equipment Company. EECO is redefining the expectations of an electrical distributor by placing people and ideas before products. Please subscribe and share with your colleagues and friends. Also leave comments, feedback, and any new topics that you would like to hear. To learn more or to share your insights, visit EECOAsksWhy.com. That's E E C O A S K S W H Y.com.