Food and beverage processors use a number of different metrics to measure and improve upon performance, and one of the most powerful is Overall Equipment Effectiveness (OEE). However, it has to be measured and applied correctly to reap the full benefit.

In the July episode of the Food Engineering Podcast, I spoke with Anthony Doss, VP of Engineering, Poultry, for Tyson Foods. Anthony shared his thoughts on how to most effectively measure, apply and learn from OEE to ensure you’re hitting your performance goals.

This partial transcript has been edited for length and clarity. To hear the full podcast, click the play button below.

Casey Laughman, FE (CL): We’re going to be discussing OEE and how it fits into a manufacturing strategy. So to start out with, could you give us a little bit of an overview of the philosophy behind OEE and how it can be applied to a manufacturing operation?

Anthony Doss, Tyson (AD): Sure. OEE stands for Overall Equipment Effectiveness and it was really created to be a standard for measuring manufacturing productivity. It tries to identify the percentage of manufacturing time that is truly productive, and so an OEE score of 100% means that you are manufacturing everything perfectly—only good parts or products as fast as that process will allow with no stopping.

CL: And with that in mind, is it useful as a tool for improving your manufacturing capability, or is it more just measuring what you’re doing?

AD: It is a tool. It can be useful as a tool if it’s defined properly. So the variables have to be defined properly. The variances that come off that have to be clearly visible and action must be taken to correct the variations at the immediate time when they come up or are seen, or that would make it non-valuable.

So just being a scorecard that you record and put it in a book for overall time is not effective for improving your process. But if you define it properly, you make it visible when you react to it, it is a useful tool in improving your manufacturing capability.

CL: If you want to set up and apply OEE to track a production line, what are the best practices for doing that? How do you go from, “OK, we aren’t really tracking anything,” to “here’s how we’re going to have a good, accurate OEE measurement that we can use to improve our performance over time”?

AD: The first key thing is to define each parameter at the product level or SKU basis for it to have meaning. So for the availability, performance and quality, you need to define each one of those per SKU that you’re going to be running on that line.

Whether it’s a batch process or a continuous process, that doesn’t matter as long as you’re defining those parameters correctly for that particular product. Also determining the right amount of planned runtime in the availability section to include scheduled maintenance periods.

So you basically could say, well, if we were to run an eight-hour shift and we had a 30-minute break, that’s seven-and-a-half hours of runtime. But if you have some planned clean-up or stops that you do consistently, anything that you do on a consistent basis, you need to take that out because then you’re just really watering down your calculation. So just make sure you define what you’re going to call in planned line stoppage, and then take that out.

And then, also accurately tracking the amount of unplanned downtime is critical so it’s reported correctly. So let’s say something breaks on the line or you have to stop it and somebody forgets to start the timer; they wait five minutes before they start the timer. Well, if it’s not tracked correctly, you’re not going to get the right measurements that you’re anticipating.

So making sure that you have a method or protocol in place, tools and people, and procedures, to get that unplanned downtime accurately reported and tracked.

CL: Obviously OEE is a pretty powerful metric in and of itself, but what are some of the other key performance indicators that are commonly used with OEE or really should be used with OEE?

AD: There are some key things that OEE alone really leaves out that you really need to also be aware of, and one of those things is labor utilization. So, the staffing that you need to run that particular product. Are you using more staff than you need or not enough, because if you’re staffed and you have people doing a particular task, let’s say that half of them don’t show up that day. Well, that actually may be in your OEE calculation based on that number of people.

Well, that’s really nowhere in there. You’re going to get a bad performance number because your cycle time is not going to be as good because you don’t have the people. And so tracking staffing, and then staffing number as well as staffing performance is key.

CL: So when we look at this from a big picture view, and we want to start applying OEE to a process. You’ve touched on some different things to watch out for, but what are those main challenges that people need to be aware of if they want to start using this as a metric?

AD: Well, I think the biggest thing is if you’re going to use it, again, is if you don’t take action when parameters get out of control, compared to what you define they need to be, then you have to do something. If you don’t do that—that’s the primary thing, is non-action.

Sometimes it means you may have to stop the line. And that’s one thing that I think there’s a little bit of a fear in people is they don’t want to stop the line, or they don’t feel empowered to stop the line. So if you do that, you’ve just got to make sure that you know that you get one chance to make things right, and you can’t wait. So when you see they’re out of control, you need to get things back in control at that moment. FE