Calculating OEE Performance Measurement for a CNC Plasma Cutter

[marcelbakker7]

I will see it this way ...
OEE is the time the machine is actually performing its task taken over 24x7. The premise is that unlike men, machine can work 24x7. Now we have two other things, the setup time and the down time. Once you have the OEE calculated over a spread time, then to improve upon it several steps can be initiated, which primarily aims at eliminating the various causes of the down time, reducing the setup time and then optimizing the working time. It is in this optimizing exercise that you apply the principles of manufacturing technology.

Thank you for your reply. Fortunatly most of the problems of the machine i'm going to measure are in the availability. Ofcourse i could leave the whole performance rate out of the measurement but that wouldn't be according to the 'rules' of the OEE principle

 

[prototyper]

Thank you for your reply. Fortunatly most of the problems of the machine i'm going to measure are in the availability. Ofcourse i could leave the whole performance rate out of the measurement but that wouldn't be according to the 'rules' of the OEE principle

Sometimes you can't tick all of the boxes and comply with the "rules".

How can you come up with meaningful measureable data on your process?

Look at what you can measure. You can record all downtime, make sure you capture all of the main reasons such as waiting for parts from the previous operation, waiting for the next operation to finish, breakdown of the tables, breakdown of the laser, etc.
From this you can calculate the machine uptime or availability.

You can monitor the in cell efficiency as a seperate issue i.e. the optimum laser cutting speeds, etc.

 

[marcelbakker7]

Thought i'd give you guys an update on the problem, we did find a way to calculate the performance. Every draw file has 3 distances wich the cutter travels: Printing distance, Cutting distance an moving distance. For every 3 theres a theoretical maximumspeed so you can measure the performance. Unfortunatly this turned out to be too much work for the operators and ourself and we also didn't think there would be much problems with that piece of the plasma cutter.

So we came up with another sollution: Move some of the time measured with availability to performance. example:

Availability = (Total time - failure time) / total time

Performance = ((total time - failure time) - performance loss) / (total time - failure time)

under performance loss we calculate the times that there is no material, no crane to load, operator not ready to cut etc. So every time loss when the machine is able to operate.

Maybe not the best sollution, but it gives the numbers we need.

 

[prototyper]

It sounds like you are on the right track.

 

[Stijloor]

Thought i'd give you guys an update on the problem, we did find a way to calculate the performance. Every draw file has 3 distances wich the cutter travels: Printing distance, Cutting distance an moving distance. For every 3 theres a theoretical maximumspeed so you can measure the performance. Unfortunatly this turned out to be too much work for the operators and ourself and we also didn't think there would be much problems with that piece of the plasma cutter.

So we came up with another sollution: Move some of the time measured with availability to performance. example:

Availability = (Total time - failure time) / total time

Performance = ((total time - failure time) - performance loss) / (total time - failure time)


under performance loss we calculate the times that there is no material, no crane to load, operator not ready to cut etc. So every time loss when the machine is able to operate.

Maybe not the best sollution, but it gives the numbers we need.

Thank you Marcel!

Could you keep us up to date on your progress?
Some may be curious about what you will do with the actual data and subsequent actions.

Stijloor.

 

[marcelbakker7]

Thank you Marcel!

Could you keep us up to date on your progress?
Some may be curious about what you will do with the actual data and subsequent actions.

Stijloor.

Sure, no problem. As soon as i've got some usefull information i'll post it

 

[phil1234]

Hi Marcel
I am the Production Coordinator in a factory which has 5 cnc lasers and 1 cnc plasma cutter. We have manipulated the calculation of OEE to suit our plant - it is nowhere set in stone - make it work for your situation! We basically exclude performance/efficiency and only concentrate on machine utilisation and quality. We do random checks throughout all shifts on the feedrates the machines are running at and compare these to budgets; the figures are kept on a spreadsheet for further analysis and future budget setting. We obviously keep track of the Dollar amount each machine produces daily - you can substitute this for your performance/efficiency component if you want.
PS. See my other posts re OEE and machine breakdowns.
Cheers.

 

[marcelbakker7]

Hi Marcel
I am the Production Coordinator in a factory which has 5 cnc lasers and 1 cnc plasma cutter. We have manipulated the calculation of OEE to suit our plant - it is nowhere set in stone - make it work for your situation! We basically exclude performance/efficiency and only concentrate on machine utilisation and quality. We do random checks throughout all shifts on the feedrates the machines are running at and compare these to budgets; the figures are kept on a spreadsheet for further analysis and future budget setting. We obviously keep track of the Dollar amount each machine produces daily - you can substitute this for your performance/efficiency component if you want.
PS. See my other posts re OEE and machine breakdowns.
Cheers.

Thank you very much for your reply. We thought too much in OEE standards too and thats why we also decided to manipulate OEE in such a way that it brings the results we want. basicly its the opposite of what you are doing, we're concentrating on performance/efficiency by measuring all the reasons why the machine isn't cutting. As mentioned earlier I will post some results as soon as we've got a good method to measure it. Were now measuring with just a piece of paper and a pen to see what reasons of inefficiency there are now so we can come up with a good measuring tool for the long term later.

 

[Mikael]

Hi marcelbakker7, any news on this subject?

 

[Mikael]

I think in OEE terms every productcategory is "another machine"...?