We have stated that policies and procedures will have an impact on your OEE implementation strategy.  One reader commented on Part I of this post stating that “OEE should be measured at the ‘design’ bottleneck process / piece of equipment that sets the pace of the line.”  While this is certainly an effective approach, the question is whether or not company policy or procedure supports the measurement of OEE in this manner.  Nothing is as simple as it looks.  Take this to the boardroom and see what kind of response you get.  We’re flexible.

As such, this becomes yet another consideration for what is being measured, how the data going to be used, and what is the significance of the results.  While we didn’t elude to a multi-series post, the comment was indeed timely.  The risk of not understanding the data could result in other inefficiencies that are built into the process that could mask either upstream or downstream disruptions.

Inventory – Hiding Opportunities

Whenever we think of the “bottleneck”, we instantly turn to the Theory of Constraints.  The objective is to ensure that the bottleneck operation is performing as required – no disruptions.  In many cases, process engineers will anticipate the bottleneck and incorporate buffers or safety stock into the process to minimize the effect of any potential process disruptions.

On one hand, the inventory, whether in the form of off-line storage or internalized, by using a buffer (or part queue), will in essence minimize or eliminate the effects of external disruptions.  On the other hand, there is a premium to be paid to carry the excess inventory as well.

While buffers or part queue’s can serve as a visual indicator of how well the process is performing, assuming the method used to calculate the queue quantities is correct, our previous post was eluding to the fact that many manufacturers incorporate contingency strategies into the process after the fact such as inventory that was not part of the original process design or reworking product on line.

Incorporating a rework station as part of the manufacturing process because the tooling or equipment is not capable of producing a quality part at rate may eventually be absorbed as part of the “normal” or standard operating procedure.  As such, it is important to manage standardized operating procedures in conjunction with Value Stream Maps to avoid degradation from the base line process.

OEE can serve as an isolated diagnostic tool and as a metric to monitor and manage your overall operation.  Company policy should consider how OEE is to be applied.  While most companies manage OEE for all processes, they are typically managed individually.  Many companies also calculate weighted department, plant, and customer driven OEE indices.

Regardless of the OEE index reported, it is important to understand the complexities introduced by product mix and volumes when considering the use of a weighted OEE index.  The variability of the individual OEE factors compounds the understanding of the net OEE index even more.

We have provided FREE Files for you to download and use at your convenience.  A detailed discussion is also provided in our OEE tutorial.  See the “FREE Files” BOX on the sidebar.

We look forward to your comments.  If you prefer, please send an e-mail to leanexecution@gmail.com

We look forward to hearing from you.

Until next time, STAY – lean!

4 thoughts on “OEE Where do we Measure – Part II

  1. This is good information. I’m glad I found it. Thank you for providing this. I provide professional web design services and solutions and can relate.

  2. Roland Pagatpat :
    I have difficulty of calculating OEE for most of our process. For example our press lamination, the machine has multiple cavity (6 cavity). Each cavity we are loading multiple layer (4 layers) with multiple panel also (4 panels each) on each panel multiple units.
    The Variables are: (depends on the product, each product has different recipe)
    1. Number of layers per cavity – 6 cavities
    2. Number of panel per layers – range (2 to 6 panels per layer)
    3. Number of units per panel – range (4 to 50 units per panel)
    4. Pressing time (cycle time) – 30 to 60 minutes
    How can i compute our OEE? For now we are just computing the performance using the number of loads for the whole day (24 hrs)
    for example:
    I have chosen one product: 1 load
    Number of cavity – 6
    Number of layers per cavity – 6
    Number of panels per layers – 4
    Number of unit per panel – 20
    Cycle is 60 minutes
    How can we compute the availability and performance if the cycle time and loading is always varying?
    Hope you can answer me..

    1. Roland, this is clearly a high mix / low volume application. I recommend creating a matrix that identifies each unique product and cycle time. If possible I would assign a unique part number or other identifier to represent each product. I will discuss a proposed solution and follow with an example to allow you to calculate your OEE, however, I would highly recommend visiting our Downloads page and also refer you to review the following original posts:

    2. OEE For Multiple Parts – Single Machine (Multipart Processes)
    3. Calculating OEE – The Real Formula with Examples
    4. Weighted OEE – How to Calculate Total Plant OEE
    5. Practical OEE – How to Calculate and Use Weighted OEE
    6. OEE and Standardized Work
    7. Using the product matrix, we can calculate the Earned Hours or Ideal Time for each Product Type made during the shift. Ideal Time for Part A = (Quantity Good Part A) x Cycle Time Part A. Repeat this same calculation for each product combination. To calculate OEE for the shift, we then use the following formula: OEE (Shift) = (Sum Total Ideal Time for All GOOD Parts) / (Total Net Available Time).

      Regarding cycle time calculations, in this case I recommend calculating the cycle time based on “Last Part Out to Next Part Out”. This allows you to isolate the load and cycle time for each unique product.

      You will see that we can calculate OEE without having to calculate the individual Availability, Performance, and Quality factors. If required, we can easily calculate the total lost time but we will not be able to isolate the performance cycle time loss. Total Lost Time = Net Available Time – Sum Ideal Time for ALL Parts. Note that ALL parts is based on the total quantity of products made. Time Loss Factor = Lost Time / Net Available Time.

      Assume 4 products were produced as summarized and shown in the table below. If the shift has a Net Available Time of 440 Minutes we can easily calculate OEE.

      ID, Cycle Time (Minutes), Total Quantity, Good Quantity, Ideal Time (Total), Ideal Time (Good)
      A 30 2 2 60 60
      B 45 2 1 90 45
      C 60 1 1 60 60
      D 30 3 3 90 90
      Total 300 255

      From this chart we can calculate OEE = 255 / 440 = 57.9%
      Quality = 255 / 300 = 85%
      Total Lost Time = 185 minutes
      Cycle Loss + Downtime = 140 minutes

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.