The OEE paradox. Why you don't know what you don't know about your OEE
… and what to do about this lack of knowledge
OEE – Overall Equipment Effectiveness – is one of the most important key performance indicators (KPIs) for any manufacturing company. At the same time, OEE is a highly controversial value. This, however, has often (or has for too long) been ignored.
A finicky construct
A KPI is good if relevant knowledge can be extracted from it. OEE delivers this without a doubt. You need a measure that tells you the level of unplanned losses in your equipment. This is clearly the task of OEE. So far, so good. And now comes the BUT: In practice, OEE often proves to be an entirely finicky construct made up of fuzzy factors.
The informative value of any OEE is therefore only ever as high as the awareness of the pitfalls associated with it.
The first fundamental challenge of OEE is clear: OEE in any plant must be tailored specifically to the company. This is where (good) decisions need to be made. Each of the three loss categories included in OEE – availability, performance, and quality – requires a range of definitions.
But the need for definition starts earlier, at the base of the OEE calculation. Even there, controversial decisions need to be made: Will the calculation of available production time be based on a 24-hour calendar or on calculated net working time? How do you calculate net working time? Which PLANNED stops are included in calculating this, how high do you set the values, and what is the reason for doing so? For example, absentees (due to weekends, holidays, production stops), maintenance, servicing, training, breaks, times for shift handovers, and so on.
It's the sum that counts
- Have decisions about planned stops actually been made in a meaningful way?
- More importantly, is really EVERYONE who uses OEE aware of the factors that have gone into its calculation – and thus into the 100% value?
- Furthermore, how often do you check the validity of these factors/values?
Six big losses
|Planned stops||Maintenance, cleaning …|
|Unplanned stops||Malfunction, power failure, sick leave...|
|Micro Stops||Small stops|
|Speed loss||Speed loss is everything that prevents the (theoretical) maximum speed|
|Startup rejects||Startup rejects are rejected parts caused by heating, starting, or other prior manufacturing stages|
|Production rejects||Production rejects are rejected parts produced during normal operation|
Six Big Losses - Die wichtigsten Verlustursachen der OEE*
BUT – and this is where the OEE comes into play: Collectively, these factors can add up to significant losses. In fact, the same is also true for distortions: It is from the sum of small distortions that the big ones emerge. That is why a high level of awareness of the quality of your OEE is so vital.
For only with this critical KPI quality awareness can your company actually use OEE to improve the added value of the equipment. Only with this awareness will even less obvious, but collectively relevant – in that they are distorting – factors be incorporated – meaningfully and continually – into your OEE (changes due to internal decisions/changes/optimizations, etc.). This is the only way to identify all mechanisms for increasing your effectiveness.
Validity is decisive
Of course, the quality of OEE has a second dimension: The critical awareness of the validity of the values/data used in its calculation. Your OEE can tell you nothing about that – and hence about your actual effectiveness. After all, your OEE knows nothing about their validity. It is the interaction between man and machine that counts here. For only if the calculation is based on reasonable assumptions AND true values will the result be a valid and comparable OEE.
One obstacle that stands in the way of many manufacturing SMEs is a lack of or insufficient data. SMEs often use tried-and-tested, yet older machines (albeit ones that are in perfect working order) to manufacture their products. Moreover, their machinery is heterogeneous, comprising different generations, manufacturers, and proprietary developments.
Any considerations to use available software solutions are often thwarted due to the problem of machine connectivity. Because in the heterogeneous machine pool, there is a Babylonian confusion of tongues: Each machine speaks its own "language" and therefore delivers data in this particular language. And that’s not all. In addition, machines communicate in different dialects and with different accents, depending on their type, origin, age, and so on. Software solutions, however, depend on uniform data in a valid and standard language.
This explains why so many companies are now shying away from moving towards digitalization, even though it would make total sense and would help them. They assume that they need new (different) machines from their existing ones for their journey down the "Smart Factory" road. This is a misconception – albeit a very common one.
Connectivity builds the bridge
Regardless of "language, dialect, and accent", you CAN connect your machines/equipment to your chosen software – the confusion of languages can be solved.
How does this work? You or your software solution provider can do this by connecting a suitably capable "translator" between the machine/equipment and the software: a connectivity tool. This tool, capable of "speaking" the language, can access the data required for each software solution at the very point where it is generated: at the corresponding part of the equipment or machine. This data is then processed, standardized, homogenized, and organized in a meaningful way. Only then is the data sent to the software via a common digital layer – in other words, in a common language without any confusion of tongues.
A connectivity solution such as this is the bridge to valid data acquisition, processing, and use. Moreover (and this is particularly important for SMEs), it is independent of the heterogeneity of the machine pool.
You can use the valid data collected in this way (since it is read directly at the "scene of the action") for more than just your OEE. You can use it to feed the entire spectrum of your desired MES as well as all other components of the ERP software that best suit your needs.
Let's go back to the OEE here:
- Deciding how to define and calculate OEE and all your company's KPIs requires the 'brainpower' of the people in your company.
- The previously unused 'brainpower' of your equipment and machines can provide the suitably valid data.
Optimization in real time
When it comes to your OEE, you enjoy a further advantage of direct data acquisition from your equipment/machinery: Not only does the data thus obtained provide the valid values that go into its calculation, your shop floor also receives details on the exact reasons for the losses.
Clearly, this is key to increasing the effectiveness of the equipment. Armed with this information, your top and shop floor will know the following in real time: WHY is production not running at 100% of its operating time, not running at 100% of its planned speed, or not at 100% of its defined quality? This gives the shop floor the opportunity to react immediately and resolve the reasons for the losses directly. This has a massive impact on your equipment effectiveness and as such on your added value – meaning that your OEE achieves its actual purpose.
Source: vgl: https://www.symestic.com/de-de/oee/six-big-losses-die-wichtigsten-verlustursachen-der-oee (05.11.2010)