As you can probably tell, I like talking about OEE. I’ve talked about it in several previous blogs about manufacturing efficiency and find that it is a subject close to the hearts of some of the smartest minds in lean manufacturing.

Recently, I noticed something peculiar when searching for the latest OEE articles. There are tons of articles and experts out there talking about how to measure OEE and what you should do to increase OEE. However, there aren’t any experts out there really talking about what you should do if you actually do increase OEE.

The smart guys in manufacturing know that the answer is never, “let’s do nothing about it”, so what’s the answer? I’ll try my best to share what some of the most intelligent manufacturers are doing when OEE actually goes up.

increased oee

The basics of OEE

Most of us already know that OEE stands for Overall Equipment Effectiveness – it identifies the percentage of manufacturing time that is spent on pure productivity. A perfect OEE score is 100% (but, world-class OEE is a myth, more on that here); which means you are manufacturing only quality parts, as efficiently as possible, with zero stop time. Those are 3 of the main tenants of OEE: availability, quality, and performance.

OEE is something more and more lean manufacturers are trying to measure more effectively and manage with greater ability. OEE is considered by many as the single best metric for identifying losses, benchmarking progress, and improving the productivity of manufacturing equipment (i.e., eliminating waste). However, baselines must first be established and the data being measured must be accurate and consistent.

You can’t improve what you can’t measure

One of the most important parts about understanding what to do once you’ve increased OEE is establishing – in a credible fashion – that you actually have increased it. Duh, right? Not so fast. There are a lot of manufacturers out there that think they are reliably collecting machine data but the data is neither accurate or consistent.

Step 1 in determining if you are working from a consistent number should start by asking yourself, “how do I know for sure my data is right”? If you aren’t confident in your answer, consider what could be done about it. Tools like SensrTrx make it very easy to view all of your machine data without complex setups, etc. Many manufacturers are looking in this direction to double-check their data.

Once you can confirm the reliability of the data, you can identify key metrics that you can benchmark and measure against. This is the whole point of OEE. It’s the ability to establish an efficiency metric and see if the initiatives you put in place or improvements you make in processes can improve those things (but, OEE can be flawed, so be careful). Your data is good, you can usually identify issues and do something to improve them.

Increased OEE for the win

So, you’ve done it. You’ve collected accurate data, identified inefficiencies, made improvements, and measurably increased OEE. What else is there to do?

Well, first you should consider updating your ERP information. Many manufacturers use ERP for forecasting, planning, and determining a large array of enterprise functions. Smart manufacturers are using their OEE metrics inside of ERP to predict demand, set up schedules, and more. If you increase OEE but never update your ERP, you essentially handcraft the perfect cake and never eat it. You want to ensure your organization is reaping the benefits and planning properly in light of your improvements.

Another thing to look closely at is exactly why you were able to increase OEE. If you fully understand the mechanisms behind the increase you can determine if improvements are permanent, seasonal, or part of a larger, more complex process, etc. Knowing this information could help you make improvements in other parts of your business.

Finalizing changes

Fully understanding why your OEE goes up can be just as important as understanding why it goes down. Both impact all parts of your business and planning, and if systems of record don’t reflect performance, accuracy, and quality improvements you may never truly reap the benefits of marked improvements.