OEE – Not Just for TPM Programs!

If someone were to force me into a corner and only allow me to use one metric in my plant I would have to choose Overall Equipment Effectiveness (OEE). The reason I would pick it is simple… it is really 3 metrics in one! Here is a quick summary of how this powerful metric works.

The 3 Pieces of OEE

OEE looks at 3 areas:

  1. Performance: Here we compare the actual output with what that particular piece of equipment should be producing. We are really focusing in on what we call “speed losses.” For example, due to the hardness of a particular material the lathe may need to run at a slower speed than normal. So, in a given day a machine should produce 100 widgets but at the end of the day it only produces 85 widgets so we have a performance of 85% (85/100).
  2. Availability: Here we look at how much time is available to run the machine (taking out breaks, etc.) and how much we actually ran the machine. With availability we are primarily looking at our losses associated with downtime. Things like setup time, machine breakdowns, and startup losses are common culprits of availability losses. So, if there were 7 hours available to run the machine and we actually ran 6 hours our performance is 86% (6/7).
  3. Quality: Lastly OEE also looks at quality. If 80 out of the 85 widgets were acceptable (5 bad) we have a quality score of 94% (80/85).

Once we have our three percentages we simply multiply them together to get our OEE. In our example, we have an OEE of 69% (85% x 86% x 94%).

Depending on which book you read or consultant you talk to you will hear various opinions of what “world class” OEE is. I work with the thought that if you have an OEE of 80% or more you are doing well. Really well actually!

Not Just for TPM Programs!

The one thing that really irritates me is how so many people think OEE is just a maintenance thing. If we go back to the aforementioned books and consultants we mostly hear about OEE in the TPM (Total Productive Maintenance) chapter or module. This is fine as OEE and TPM definitely go together.  But let me just get on my soapbox for a bit. OEE is arguable to most holistic productivity metric known to mankind.

So if you think the maintenance guys are the only people that should be driving OEE with their TPM program you are dead wrong.  Anyone and everyone that cares how many widgets are produced in the plant should care about OEE.

The operator should care (and be intimately involved in capturing the data). The line supervisor should care. The quality manager should care. The plant manager should definitely care. The maintenance team should care. Even the controller, who wants to see all the cash flowing on the financial statements, should care. Heck, your significant other, who wants to see you employed for a long time, should care about OEE! It’s just that important.

Here is free video and OEE calculator you can download.  And yes, I know, it is in the TPM section of this website.  C’est la vie.

Subscribe to LSS Academy

If you enjoyed this article please consider subscribing to our full feed RSS. You can also subscribe by email and have new articles sent directly to your inbox.

Tagged in:,

4 Comments

  1. Jim Barnard

    December 22, 2007 - 12:46 pm

    What happens if you don’t have orders? Won’t you take a hit with this metric? I ask since over production seems to be more critical than poor OEE. Am I missing something?

  2. JWDT

    December 22, 2007 - 7:41 pm

    OEE is a great indicator and exposer of problems that need to be addressed. In my opinion, to drive OEE to as high as possible requires the base assumption that a) you are over capacity & b) each machine completes an end product from start to finish. OEE is great if you are looking at one machine, but if you look at the overall (assuming an unbalanced line) it may actually make you overproduce.

  3. Ron Pereira

    December 22, 2007 - 11:07 pm

    Hi Jim and JWDT,

    Great question and comments.

    Theoretically, OEE could lead one to over produce if your sole mission was to maximize the availability portion of the formula.

    My approach is to simply allow the operator the downtime code of “no demand” which still counts as downtime and technically dings you on OEE. And it should.

    However, the problem (and it is a problem) here is simple. The constraint is something other than this asset. It’s the market or the supply chain or whatever. Whatever it is we need to know about it and do something about it.

    The key to OEE is that it is a holistic measure of productivity… so it’s not just about the machine(s). It’s about the whole system.

    Remember, we would love to maximize the use of our machines assuming they were producing product someone wanted. That is how you maximize things like RONA. But, of course, we never ever want to over produce for the sake of OEE.

    The other option is to play around with the available time in the formula. But in my opinion this is a slippery slope and I would stay away from doing this.

    Finally, I would definitely focus my OEE analysis on constraint assets first and foremost. This also helps with the potential over production issue.

    Thanks for the comments guys.