We recently received an e-mail regarding OEE calculations for batch processes and more specifically the effect on down stream equipment that is directly dependent (perhaps integrated) on the batch process. While the inquiry was specifically related to the printing industry, batch processing is found throughout manufacturing. Our more recent experiences pertain to heat treating operations where parts are loaded into a stationary fixed-load oven as opposed to a continuous belt process.
Batch processing will inherently cause directly integrated downstream equipment (such as cooling, quenching, or coating processes) to be idle. In many cases it doesn’t make sense to measure the OEE of each co-dependent piece of equipment that are part of the same line or process. Unless there is a strong case otherwise, it may be better to de-integrate or de-couple subsequent downstream processes.
Batch processing presents a myriad of challenges for line balancing, batch sizes, and capacity management in general. We presented two articles in April 2009 that addressed the topic of where OEE should be measured. Click here for Part I or Click here for Part II.
Scheduling Concerns – Theory of Constraints
Ideally, we want to measure OEE at the bottleneck operation. When we apply the Theory of Constraints to our production process, we can assure that the flow of material is optimized through the whole system. The key of course is to make sure that we have correctly identified the bottleneck operation. In many cases this is the batch process.
While we are often challenged to balance our production operations, the real goal is to create a schedule that can be driven by demand. Rather than build excess inventories of parts that aren’t required, we want to be able to synchronize our operations to produce on demand and as required to keep the bottleneck operation running. Build only what is necessary: the right part, the right quantity, at the right time.
Through my own experience, I have realized the greatest successes using the Theory of Constraints to establish our material flows and production scheduling strategy for batch processes. Although an in-depth discussion is beyond the scope of this article, I highly recommend reading the following books that convey the concepts and application through a well written and uniquely entertaining style:
- In his book “The Goal“, Dr. Eliyahu A. Goldratt presents a unique story of a troubled plant and the steps they took to turn the operation around.
- Another book titled “Velocity“, from the AGI-Goldratt Institute and Jeff Cox also demonstrates how the Theory of Constraints and Lean Six Sigma can work together to bring operations to all new level of performance, efficiency, and effectiveness.
I am fond of the “fable” based story line presented by these books as it is allows you to create an image of the operation in your own mind while maintaining an objective view. The analogies and references used in these books also serve as excellent instruction aids that can be used when teaching your own teams how the Theory of Constraints work. We can quickly realize that the companies presented in either of the above books are not much different from our own. As such, we are quickly pulled into the story to see what happens and how the journey unfolds as the story unfolds.
Please leave your comments regarding this or other topics. We appreciate your feedback. Also, remember to get your free OEE spreadsheets. See our free downloads page or click on the file you want from the “Orange” box file on the sidebar.
Until Next Time – STAY lean!
Great site…keep up the good work. I read a lot of blogs on a daily basis and for the most part, people lack substance but, I just wanted to make a quick comment to say I’m glad I found your blog. Thanks,
A definite great read..
In batch processes, apply the OEE to the bottleneck equipment of a series of equipments may not help to increase the capacity of the whole production. If you have a bottleneck between equipments, like at the in process control analysis, you will not improve the capacity of the whole process improving the OEE of the bottleneck. What procedure would you apply in this case? Which machine would be selected to apply OEE?
Hi Beatriz, thank you for your question. Consistent with the Theory of Constraints, the very short answer is that I would improve and apply OEE to the process that is constraining the TRUE bottleneck. If it is not a capacity constraint, I would be inclined to use the Lean Six Sigma tools to improve the process.
Note that there is no rule that says you can’t or shouldn’t measure OEE at multiple steps in the process. In other words, this does not have to be a choice of “here or there” – it can be both. There are two objectives to managing the bottleneck: 1) Never Starve the Bottleneck for Product, and 2) Make Sure the Bottleneck is ALWAYS running as Scheduled. If you have identified the TRUE bottleneck, you are effectively managing throughput of the whole system.
I would also attempt to understand whether the constraint ahead of the bottleneck has the capacity to handle the demands placed on it and how that process is being scheduled. Is it really a system constraint or is it poor planning? Is the scheduling based on a balanced line or variable demand pulled from the system? Random stock outs can appear when a balanced capacity model is being used to schedule production on upstream processes. It may even be best to decouple these processes from the bottleneck itself.
Also note that too many companies invest a tremendous amount of capital, time, and effort to improve processes that will have no effect on the throughput of the system as a whole.
If you need more information to understand the concept, I highly recommend reading “Velocity: Combining Lean, Six Sigma and the Theory of Constraints to Achieve Breakthrough Performance – A Business Novel“. This is a terrific read and can be finished in a few short days.
Sorry, I think the question was not well formulated.
In a batch process, suppose I examined the equipments and determined as the bottleneck the one with the bigger process time.
But, then looking to the lead time of the batchs it was noticed that the batchs stay waiting for the approval of quality control for a longer time, so this is the restriction of the system.
If it was a continuous process, improving the OEE of the bottleneck would increase the capacity of the whole process.
But in this case, is it still worthy measuring the OEE of the equipment as it is not the restriction of the system?
This is an excellent question. Yes, it is still worth measuring the OEE of the equipment if excessive downtime would jeopardize your inventories and / or deliveries to your customers. Why does it take so long for Quality Control to give approval?
Based on what you describe, I would recommend finding ways to reduce the time it takes to get the approval from Quality Control. If the approval process a time dependent function (life cycle testing such as salt spray testing), you may want to consider another form of controlled product release that could be based on Reliability data (Weibull Analysis).
If the approval process is constrained by human effort, you can add resources to expedite the approval process. Your situation reminds me of the one described in the book I recommended in my earlier reply.
The analysts in an off-site R&D facility were responsible for approving a process before it could be run through a specialized key piece of equipment. This approval process could range from minutes to days and weeks. Delays caused major disruptions to the flow of work through the plant.
Eventually, the analysts found ways to reduce their approval time and the key piece of equipment remained the bottleneck.
Thank you again for your question.
I like very much this site because the comments are based in pratical applications and help a lot of users.
I have a doubt.
The performance efficiency is a factor in the calculation of OEE. It is a function of cycle time (theoretical time to produce a unit of product):
Performance Efficiency = (Ideal Cycle Time x Total Prod Qt) / Net Operating Time) x 100
Total Prod Qt: total quantity of products
Cycle time is easily measured in continuous production lines.
However, I am calculating the OEE for a batch process equipment.
It is a tablet coating equipment. We fill the machine with the entire batch of tablets, the process is carried out and then we discharge the batch of coated tablets. I was thinking of dividing the batch production time by the total number of tablets processed to determine the cycle time of each batch, but I don`t know if it is correct to call it cycle time? Is there another formula to calculate the performance for batch processes?
Dears,
I’m very honored to your response and Helping Data , and I hope that you can help me with an issue :
I’m working in an ink manufacture , mainly our process consists of three main stages (mixing, milling, filtration & packaging).
we have a variety of products depending on customer needs (color – special needs …)
For example, during production of 1 ton of white color ink , there are 4 steps :
1- Preparation of raw materials, vessel & weighing (20-30 Min)
2- Mixing of the recipe (Time Varied according to the recipe mixture)
3- Milling (Time Varied according to the recipe mixture and batch size)
4- Filtration & packaging (Time Varied according to the recipe mixture and batch size)
The issue is that the system is a batch process with a huge No. of varieties and Variables , calculation of ideal cycle time for a single process ??
how should we start ?? taking into consideration the start point (( Machines Or Products )) .
Can you help us?
Looking forward for hearing from you.
BR
Thank you for your inquiry Mahmoud! We can appreciate some of the complexity that batch processing introduces and more so where custom orders are concerned.
We would first attempt to determine which machine is the bottleneck process and focus our efforts there. Although each of these processes may be decoupled (separated), the bottleneck will determine the net throughput for the entire process.
We suggest focusing on the Filtration and Packaging process first as this may be easier to establish a production rate. Flow rates can be equated to fill rates for containers making it possible to formulate an ideal cycle time for that part of the operation.
We would also suggest looking at historical production data for each of the operations. In essence, time study data can be gathered from this history to determine what factors influence cycle times accordingly. This historical data serves as surrogate data for future formulations and could be used to determine cycle times in kind.
We highly recommend reading “The Goal” by Eliyahu Goldratt where the theory of constraints is thoroughly explained. The ultimate objective of OEE is to reduce waste in the form lost time. Identifying and eliminating the forms of waste that exist in your processes will increase yields, throughput, and ultimately … profits.
With the limited information we have, it is difficult to determine where the real process constraints are. Balancing resources in the form of machines or batch sizes is dependent on available capacity.
Thank you for visiting.
Hello,
We have a powder coating unit wherein we need to measure the Performance of the Unit. To do so, we do not know how to capture the data as it is a batch process.
We have a booth wherein the powder is sprayed onto the components and it is sent to the ovens in batches for baking.
Knowing the above scenario, kindly suggest how to go about.