Why Overall Equipment Effectiveness (OEE) should be banned immediately

What is wrong with Overall Equipment Effectiveness (OEE)? In the basis nothing, but in reality everything. The idea of OEE was developed in the ’60s and it has become one of the most common used metrics for operations. OEE certainly has brought many good things over the recent years. It provides a structure to how you measure, common definitions and typically data for the various loss categories. So far, so good.

So, why should it be banned? Because you will hit the limitations of what OEE can do for you. Here is why. First of all, OEE is an aggregated measurement that disconnects you from the physical reality. It is a calculation, multiplying Rate x Losses x Quality. When I was running plants, the operators did not recognise the calculated numbers. As an operations director, I had to get into the nitty-gritty details to understand the numbers. Yes, OEE certainly is a reference number, but that’s basically it.

Overall Equipment Effectiveness

Secondly, we like our metrics to be good -not necessarily to be real- with the option to learn from it. However, an OEE measurement is subject to manipulation. All too often during our detailed data driven assessments, we notice that OEE categories are used incorrectly. For example, the category is machine unscheduled but in practice maintenance activities are being executed. Or operational losses are shifted into commercial downtime. The examples are endless. Sometimes, it is done to manage the numbers, sometimes because of historical behaviours. How this works in practice becomes clear when we push for internal benchmarking across plants. The argument we hear frequently from plant managers sounds like “Yes, we all measure OEE but all differently”.

Variance based

Thirdly, OEE is not variance based. Since OEE numbers typically get calculated on a daily basis, you loose sight of what has actually happened. Suppose you have a line rate norm on which the line should run. Running faster than the norm will help your OEE since it compensates for other losses or lost rate. This is what we call catch-up behaviour and it is simply bad. It might (and probably will) destabilize your line performance even further.

Plant behaviour 

Finally, OEE doesn’t drive the right plant behaviour. We believe operations should do one thing and one thing only: hit your production standards, for example on changeovers, speeds, cleaning, maintenance, et cetera. And if you meet your standards, improve them. With OEE you can’t drive this behaviour. It actually recognizes day or week records of which is proven that they further destabilize your plant leading to less output on average. Another limiting aspect is that OEE does not measure business impact. Furthermore it is a lagging and not a leading indicator. And it doesn’t drive cross-functional improvements. We will review these OEE specific aspects another time.

StableOps™ as a superior alternative for OEE

To summarize: what was good in the past is not necessarily good in the future. Why would you let OEE limit your possibilities to find the next level of productivity opportunity? At R&G Global Consultants we prefer to work with superior StableOps™. We will get into the advantages of this method this in a next blog. Until then, feel free to contact me at adewolf@rnggc.com.

Aart Willem de Wolf is Managing Partner at R&G Global Consultants in The Netherlands.

Stable Ops™