Monday, August 24, 2015

10 Ways to Ensure You Don't Screw Up the OEE

By Kathy Barthelt, member of MESA International

I have to be honest with you... Most OEE projects fail before they’re implemented and it's up to YOU to make sure that doesn't happen! Not everyone has the same priorities and what’s important today may not be tomorrow. So how do you put together a project that will address the desires of all the stakeholders? What’s the best way to navigate the political landscape that’s been laid out in front of you?
Is it possible to NOT screw up the project? 

Does your company need to increase OEE and the project has been dropped on your plate? Ah yes, the OEE approach. You remember what that is right? The Overall Equipment Effectiveness metric as developed in Japan during restoration made famous in crafting some serious efficiency. It’s your lucky day and management has decided that THIS is the way to make the world better, and you are the person to make it happen.

There will be pitfalls along the way:

·         Somebody is going to tell you that the best way to start is collecting data manually so that you can understand the process better.
·         Somebody else is going to say that your OEE is fine, but the real problem is _____________ (fill in the blank from the following: SPC, lot tracking, scrap, downtime, indirect time, maintenance, line scheduling, set-up reductions, etc.)
·         Somebody big (The VP Ops) is going old school and needs proof that an OEE system will work for their people.

This puts you in a tight spot. An OEE project requires a lot of people and they’ll all have their own ideas. Try to satisfy them all and you’ll have a six-figure project with 18 months to implement. Even if it gets approved, good luck getting it implemented successfully. This could get ugly if you’re not careful.

Is the key to happiness a good OEE?

Not everyone has the same priorities. Importance is relative in the big picture. And what’s important today may not be tomorrow. So how do you put together a project that will address the desires of all the stakeholders? What’s the best way to navigate the political landscape that’s been laid out in front of you without getting off course? I have to be honest with you. Most of these projects fail before they’re implemented and never get off the ground. Now it’s up to YOU to make sure that doesn’t happen.
What if you had a plan that could deal with all of this? Your life will be easier, you’ll get the recognition you deserve, and you may become known as that “someone who can get things done.”

Here are 10 ways to succeed at an OEE project where 90% fail at trying:

1.   Collect the Requirements. Learn with the intent of developing a phased approach to implementing on your shop floor with OEE being Phase 1.
2.   Create Your List. Capture the required functions, taking into account what the output of the system will be. What does the plant manager need to see in real-time? What KPI’s does each line need displayed in real-time? What reports are required?
3.   Insist Upon Real-time. In the moment data for the right OEE is the right approach. If it’s possible, collect the data automatically. Remember that real-time feedback to line operators results in an automatic increase in OEE.
4.   Evaluate Your Lines. Focus where production counts can be monitored automatically. If the data is in your PLC’s, can you get it out? OPC communication is the right way to go here. If not, the approach is to install a new dedicated PLC with sensors installed on each line.
5.   Find Your Data Points. If automatic production monitoring is not applicable, what will be your collection points and how will you collect the data?
6.   Calculate the Load. Determine how to load the job you’re reporting into the OEE system. This will typically be the order/operation or the product from the ERP.
7.   Recognize Great Data. Do not accept manual collection of data as a viable approach because it produces false results and is labor-intensive.
8.   Be Tough. Evaluate systems based on OEE specificity to start and expandability to future phase functions as determined by your requirements. Plan to justify the OEE purchase on its own merits.
9.   Go Easy. Make sure the system is easy to implement. Software installation and configuration should take no more than 2 weeks.
10.    Be Simple. Put together a detailed but simple project plan indicating who will do what, how long it will take, and how you will monitor progress.

Execution is Everything

So now you have a well-tested plan that works, so your chances for success are good. Don’t worry, you may not be an expert in OEE at this point but you will be shortly. And best of all, you’ll not only look like you know what you’re doing, you actually will.

Ok, let’s go get that data and turn it into real information that can be used to improve OEE, not just monitor it.


Kathy Barthelt is the vice president of Crossroads RMC, which helps optimize manufacturing systems. She cares deeply about bringing the human side into manufacturing. One can often find her writing and speaking about stress and shop floor workers, job skill, or employee motivation and production improvement.

No comments: