Home > Forums
Topic Title: OEE Headache
Topic Summary: A Question Regarding OEE
Created On: 08/14/2019 09:05 PM
Linear : Threading
Send to a Friend Send to a Friend
Search Topic Search Topic
Topic Tools Topic Tools
View similar topics View similar topics
View topic in raw text format. Print this topic.
08/15/2019 11:57 AM
Print this message

Author Icon
Mark_Saunders
Mark Saunders



Hi all, new here. I have lurked for a little while, reading a lot of posts and articles before making the leap and signing up. ??

I do have a specific question regarding OEE but first a little background. I work for a global manufacturing company; our head office is based in the US but I work in one of our sites in Australia. We are in the FMCG and Beer, Wine & Spirits industry, specifically printing labels.

We are in the process of looking at different performance metrics to use Operationally with OEE being top of the list. About 12 months ago we introduced a new MIS into the business which incorporates an automated data collection system linking our production equipment. This has given us access to far more accurate real time data than we have ever had access to previously from an operations perspective.

The issue I am having is with the formula's used to calculate a meaningful OEE score. Unlike discrete manufacturers our processes involve a lot of variability from job to job, and machine to machine. There is quite a bit of waste built into our job estimates and planning, but the biggest hurdle I am facing is that it is possible in our case of being able to produce our products using less materials than estimated as well as producing them in less time than estimated/scheduled. Using the generic OEE formula's it is possible to have multiple instances of Quality, Performance or Availability values greater than 100%.

I have been able to come up with formulas for both Quality and Availability that still give us a meaningful result that do not exceed 100%, but the Performance metric continues to give me headaches.

1. Has anyone else had a similar experience and found a solution?
2. Is tailoring OEE formulas to suit your own processes OK or is there a better performance measure to use in these types of cases?

Still fairly new to Lean concepts and methodologies so my apologies if I am missing something very simple. ??

Thanks
08/19/2019 04:28 PM
Print this message

Author Icon
KenAbbs
Ken Abbs



If performance is over 100% then your ideal cycle time is set too high in your formula. Perhaps try to base it of your best performance for a meaningful period vs a design number.
08/19/2019 04:28 PM
Print this message

Author Icon
MarkGalloway
Mark Galloway



Good question. I don't know your current situation, but one question is do you know the maximum design performance that the process can produce? One approach is to use the maximum or theoretical rate for the process and use that as your comparison. This has helped me figure out how much waste was in the system and also drove us to a mark that may not be possible to attain but the closer we got to it, the better we got. Feel free to contact me for more details. I am in a new industry recently and we have been working on this for the past several months trying to develop meaningful OEE metrics.
08/19/2019 04:28 PM
Print this message

Author Icon
Robert_Simonis
Robert Simonis



Do not try to calculate OEE for multiple steps or overall for a system or a factory. OEE should be for the one pacemaker or bottleneck process step in a value stream. In that one step, calculate how much you could produce if it ran non-stop at it's speed when it was brand new (or the best speed it ever ran) (no changeover, no set up, no waiting, no downtime, etc.) Now look at how much you did produce. (Produced x 100) / Capacity = % Utilized. Anything less than 100% is waste. Figure out why it is less than 100% and attack those problems.
08/19/2019 04:28 PM
Print this message

Author Icon
DannyLuo
Danny Luo



Mark,

I am thinking how your estimation is worked out?Is it related to equipment parameters?Actually, performance is equal or below 100%.

Danny
08/26/2019 10:35 AM
Print this message

Author Icon
Mark_Saunders
Mark Saunders



Thank you all for your replies.

Ken and Mark, our ideal cycle time is variable, from job to job and press to press due to the complexity of label embellishments (some can have few embellishments and others have many). Maximum design performance for the process is scaled based on what value-added components are included and how much the press configuration is changed in Set-up. But you have given me something to think about as I could potentially devise an ideal or theoretical cycle time for each machine based on 'best case' scenarios using the data we've collected over the past 12 months.

Robert, thanks, that also has given me something to mull over. I have come across several differing articles and books that have conflicting views on how OEE should be used and in what capacity.
I feel like it's not a fit for all generically, but the concepts can be adapted to suit based on processes and needs?

Identifying the major wastes in our process is the driver here but I also want a meaningful performance metric that engages operators.
08/28/2019 09:23 AM
Print this message

Author Icon
mhunmann
Michael Unmann



Mark

Ideally you should use Maximum Demonstrated Rate (MDR) if you have the historical data. The "scheduled rate" or "designed rate" does not allow for improvement, it will give your goal a ceiling. By using the MDR you open the opportunity to increase performance by allowing the resetting of the baseline.
08/30/2019 10:15 AM
Print this message

Author Icon
MarkGalloway
Mark Galloway



The maximum rate is based on the equipment parameters design criteria. I like to show that the process is capable of achieving that design rate even if it is for a few minutes. That will help show the team that it is possible to reach the rate. You are correct that performance should be at or below 100%.
09/18/2019 08:24 AM
Print this message

Author Icon
jaidynmoore
jaidyn moore



OEE is the percentage of the time that a machine is producing "sel-able" parts at the "standard" rate (or std UPH). There are two ways to detemine OEE, depending on how you want to use the information:

1- simple: calculate how many parts the machine is capable of producing in one week (168 hours) at the standard rate. For example, if the standard rate is 1 part per hour, then the machine can make 168 parts in one week. Then compare that with the actual number os parts made duing one week. OEE will be the ratio of these two. In the above example, if the machine actually produced 80 parts in one week, then OEE=80/186.

This method will quickly tell you how productive your machine is. However, it will not tell you "where" you have opportunities for improvement. For that you need to use a detailed method as below:

2- OEE = Availablity x Utilization x speed efficiency x rate of quality

Among these 4 factors, the lower number can be the place you need to focus.

hope this helps
Note: These forums are moderated by the Lean Enterprise Institute. All posts are reviewed within 24-48 hours prior to appearing on the site. Views expressed in these forums do not necessarily represent the views of the Lean Enterprise Institute.