Select Page
9. How to measure OEE for different products & different cycle times…

Q: We have lathes and machining centers that we use where we have multiple set ups and many different parts with multiple cycle times run on those machines. We will be applying OEE to these machines and I am curious what others have done when the cycle time for the parts always changes. When a process has a constant cycle time, Performance measures also remain constant. Averaging cycle times will not provide accurate information.

Arno Koch •   This type of equipment indeed challenges a proper OEE implementation…
I Agree,  averaging cycle times is not the way to go. What then? First of all it helps when the variation is not too wide.

## Different products with different speeds in OEE

The performance rate is set to 100% when the machine runs at theoretical maximum speed. But what to do when the machine produces different products with different maximum speeds?

## Determining the theoretical maximum speed for different products

To determine the theoretical maximum speed, the maximum cutting speed multiplied with the amount of material to cut has to be determined. Different types of raw material (Alu, V2A, Bronze, Nylon etc) have different maximum cutting speeds, and the amount of material to be removed can be different per product. (Two physical limited parameters)
(The complexity of the product is not a ‘maximum speed inhibitor’ since theoretically this can be overcome).

## Optimization

Probably the 80-20 rule is hidden somewhere! In other words, usually it is possible to divide the product portfolio into a limited amount of ‘maximum speed groups’;

Now the operator can go through a simple ‘questionnaire’ or ‘decision matrix’ to determine the maximum speed of his next product and uses this as the ‘standard’ against what the performance rate is calculated. (Example: Vertical axis the material determining the cutting speed,  horizontal axis the class/group of amount of cutting needed).

The goal is not to have a 100% correct calculation. The goal should be to give the operator an indicator, showing him where improvement potential is hidden. This is most likely not in maximizing the actual (cutting) speed. And if it is, it is enough to know this is the place to look for optimization.

I am curious if there are other solutions being used in similar situations… Please let us know!