Production faces an invisible cost of such requests, whereas marketing will peacock charts of rising sales. It’s like bringing a knife to a gun fight.
Yves Bourgeois on
OEE is not a new concept. Its origin dates back all the way to the nineteen-eighties, the decade of economic growth and relocation of manufacturing to the East (and of course the decade of Pac-man). OEE was first described –as a central component of the TPM methodology- in Seiichi Nakajima’s book ‘TPM tenkai’. Not accidently in Japan, given the expansion of the economy, manufacturing boom and need for increased output.
A concept of 40 years old and well-established practice should have inspired an abundance of practical case studies, you might think. Well, unfortunately, the internet is not exactly the treasure chest for the case study raider. Before we partially solve that predicament, let’s make sure we are on the same page conceptually.
OEE doesn’t need a lot of defining, because it is so well known. So just as a brief recap: OEE is a measure of how well a manufacturing process is utilized, compared to its full potential when it is scheduled to be used. An OEE of 100% means that only good parts are produced (100% Quality), at the maximum speed (100% Performance), and without interruption (100% Availability).
When it comes to OEE, the devil is in the detail. Not just in the calculation of the OEE, but in the explanatory value as well. If you know that a production line has a 80% Availability, that’s one thing. If your OEE calculations teach you that is because of a specific machine with intermittent drops in Availability, that’s a different story.
Imagine it would even teach you which part of that machine is causing the downtime. OEE shifts from an informational metric to an operational metric, but it is all based on real-time analysis and enormous detail.
Let’s consider for a moment a contemporary situation in a dairy processing company. Operators and plant managers are scrambling to get the current production at an OEE of 85%. Then, along comes the marketing department with the request to label a batch with a promotional message. In a heavily customer-centric company the mandate for production management to refuse such a request is thin. However, their arguments may be very legitimate.
So with the right level of detail about the Availability and Performance of a specific line and its equipment, production management is now able to evaluate the marketing request with hard figures. Imagine that the required labelling machine has an unavailability of 20%, it is quite easy to calculate the cost for a line that has 4.000 units of output every hour. If their analysis would show that the cost of such a request is higher than the ‘expected’ returns, it is not a matter of customer centricity, but stakeholder thinking that the request is refused.
OEE can balance out the power shifts that are taking place in many customer-centric companies.
Another example is a weaving company. Every production change, operators are required to tie 1.800 knots on every machine. This technical process has a high cost: it’s labor intense and vulnerable for human errors. By setting up an OEE automation process, the company would become able to calculate the cost of every operator on a machine. They would not only finally have hard metrics about the Quality, but also the Availability.
Of specific interest here is the inter-operator variable. Are there people in need of additional training or supervision? Is there a difference between daytime and nighttime operations? How come? And what would be an approach to close that gap. This is a clear example of the benefit of a bottom-up OEE calculation. It contains a lot of details that can be leveraged for operational decisions.
A Belgian company is producing beer (duh!) and packaging in six-packs, but also eight-packs. Due to some arbitrary decisions, it produces like this: 6,8,6 in 24 hours, meaning that the packaging machine setup needs to be changed two times in this sequence. There is no need to change this cycle as it is unclear what the cost is of this sequentialism. OEE would clearly measure the cost of the downtime between batches.
Although people have an underpinning gut feeling for this type of cost, it is much better to base these decisions on hard facts since our intuition can be completely off.
Take this for an example: a yoghurt producing company had a machine that went intermittently down for just 20 seconds, a so-called microstop. Since it did not have a high impact on production – according to the gut feeling of the involved – they let it be. OEE calculations however, showed that the machine was down, for no less than 20 minutes a day, with a cost high above what was deemed acceptable. The problem was raised to the manufacturer who then claimed it was a known problem that could easily be fixed.
OEE is a well-known but underutilized metric. If you’re stuck in manual calculations, top-down high-level measuring and simple trends and benchmarks, you are clearly not reaping all the benefits of OEE.
With the current technology and automation capabilities, the concept can thrive forty years after its conception. It will serve not just as an informational metric, but as an operational tool to further optimize a production environment, which is of specific importance to European companies faced with high costs.
In our next article we will outline how to implement an automated OEE and leverage it as an operational decision making tool.
Read how Factry OEE helps you optimize the use of your production capacity, lower inventory, improve reliability and increase the bottom line.