OEE (Overall Equipment Efficiency) - One question

[Umang Vidyarthi]

For OEE measurement, i saw some sites use the planned work days as the total availiable run time (TART), e.g. 20 days a month (instead of 30/31 days a month), especially during the encominic recession period or a new facility. Is this correct (or say popular) for measuring the OEE??

You are not right in your assumption. OEE (Overall Equipment Effectiveness) breaks the performance of an equipment into three measurable components, Availability, Performance and Quality. A X P X Q id est:

70% Availability (0.7) X 60% Performance rate (0.6) X 10% Quality rate (0.9) = 38% OEE

It is impossible to get 100% OEE. The ideal benchmark is 80-85%.

TEEP (Total Effective Equipment Productivity) represents production of good parts vs total available time which is also based on 24 X 7 through the year.

TEEP = OEE X Loading time.

In a five days a week working facility

Loading = (5 days x 24 hours) / (7 days x 24 hours) = 71.4%

TART is not to be confused with OEE & TEEP.

Umang

 

[palmer]

OEE may be set up for 365/24/7 but I don't think that is a fair assessment. Several factors play into the perception of this...

1. Equipment has to be maintained or it will breakdown. Time HAS to be set aside to perform this operation.

2. We are geared toward goals. From the time we are children, we have the perception 100% is the goal and IS acheivable. Under OEE it is NOT achievable. It is deflating for employees to see the numbers from OEE because of this.

We have a new performance evaluation based on 1 thru 5. We were told that 3 is 100%. Very few get 4's and virtuually no one gets 5's. WTH?

It is very deflating and an example to compare to OEE and human perception....

 

[Citizen Kane]

Hi !

Same discussion with 3 for 100% here. I also think is unrealistic and reported like this only to be reported.

The indexes used in KPIs should be understandable and should reflect the situation as it is. If a KPI isn't used, then why should we keep it ?

Also, it's true, how the operators should understand this ? In this mathematic-semi-empirical-theoretical-(...)al calculations we are loosing the common sense understanding of the indexes. But, on the other hand, we ask them to reach what they cannot understand.

 

[jasonb067]

Availability x performance x quality

Where availability is (Operating time / planned production time)

performance is (Total pieces / Operating Time) / Ideal Run time)

quality is (good pieces / total pieces)


Every bit of TPM training I have had indicates that we will define planned production time as time actually scheduled to run. So, if by design we make something perishable we may run 24/7/365. That would be our planned production time. But, for example, in a traditional automotive setting with a week down in July we would not include those 7 days. Again, if we do not plan to run production on weekends because our business model says 5 days a week then we do not include saturday or sunday.

That is OEE.

If we are trying to determine what our utilization of "expensive equipment" is there are other terms and calculations we can use. But, that is not the origional inetention of OEE as I was trained. The intention of OEE was to assign an oveall KPI to scrap, machine set up time beyond what was planned, unscheduled down time for material stock out, machine fail etc.

Now, any organization can determine what and how they want to measure anything based on their business model. They can also call that measurment whatever they want.

I think, however, using OEE to describe Equipment Utilization or facility capacity is mis-leading to someone outside of the organization which may need the OEE information to make decisions (customers, stakeholders, machine designers etc.). I can have my organization refer to the number 7.067067067 as π but I will not be line with the rest of the world.

I may be way off base here but this opinion is based on my training on OEE, what it means to an organization and the most effective ways to use it.

 

[wmarhel]

Every bit of TPM training I have had indicates that we will define planned production time as time actually scheduled to run.

True, and similarly, the maintenance activities should be "planned" and scheduled as well. Unfortunately, what too many organizations have is a maintenance department that doesn't really do any maintenance. In fact, they are essentially repair people, and only really touch the machine once it has failed. Unfortunate, but true.

Wayne

 

[palmer]

True, and similarly, the maintenance activities should be "planned" and scheduled as well. Unfortunately, what too many organizations have is a maintenance department that doesn't really do any maintenance. In fact, they are essentially repair people, and only really touch the machine once it has failed. Unfortunate, but true.

Wayne

This is so true. I believe that maintenance was originally created to fix rather than prevent. This does include having to change oil, filters, brushes, etc... which were requirements for machines and equipment but other than that they were reactive not proactive.

I used to get into "conversations" with a maintenance person over performing things on equipment while it was down for changeovers, QA, issues, etc...

He was posted in my area to fix any breakdowns on my production lines. If the machine was stopped for any of the reasons I mentioned he would just stand there and wait. We would discuss his needing to change parts, filters, inspect problem areas, etc.... while we were down. He would say it wasn't his job to do those things until it was scheduled.

I got around that way of thinking (this was a union shop BTW) by telling him I was scheduling him to do this or that on the production lines while they were down. I had grievances filed and all kinds of headaches.

But guess what. My production lines broke down less than anyone elses and they ran better than any of the others.....

Go figure....

Might be why I got promoted to Industrial Engineering after that....

 

[Fillemon]

Just take the 'theoretic' calculation method and adapt it so that
- 0% is when everything goes wrong
- 100% is when everyhting goes as you would wish.

Hence:
- If your factory wil never never never run 365/24/7, than it's foolish to take this as benchmark.
- If every hour there is one piece taken out of production (e.g. for testing, sampling), than do not consider this as a NOK piece. For a worker, seeing "100% OEE" on a board is much more rewarding than seeing "98% OEE" and always have to think "but it would be 100% if not every hour this test piece would be taken".
- ...

 

[palmer]

Just take the 'theoretic' calculation method and adapt it so that
- 0% is when everything goes wrong
- 100% is when everyhting goes as you would wish.

Hence:
- If your factory wil never never never run 365/24/7, than it's foolish to take this as benchmark.
- If every hour there is one piece taken out of production (e.g. for testing, sampling), than do not consider this as a NOK piece. For a worker, seeing "100% OEE" on a board is much more rewarding than seeing "98% OEE" and always have to think "but it would be 100% if not every hour this test piece would be taken".
- ...

I completely agree with using actual and not theoretical.

The information you are providing needs to be put in layman's terms for the hourly employee. Almost all of them understand 100% but not 98% IS 100%.

Heck, most managers are in this boat as well.