OEE formula – bad idea to cook up

Why a ‘convenient’ OEE formula is a bad idea

In devising an OEE formula, it is important that you are truthful about losses like downtimes, rejections and rework. OEE is not a number that you flaunt to others. It is a mirror that helps you evaluate and improve the performance of your shop. It is a number that you constantly strive to improve, not a number that you achieve and then sit back and relax.

Cooked up OEE formula is a bad idea
Cooked up OEE calculation – a bad idea

A lot of firms consider most downtimes as part of the process, and NOT as downtimes. E.g., part unloading / loading, inspection, setup change. This flies against the logic of the use of OEE calculation. It artificially inflates the OEE number, but is counterproductive.

Availability is defined as = Run time / Planned production time.
Run time = Planned production time – Downtime.
Downtime is the time when the process was supposed to be running but did not run. This includes part unloading and loading, inspection, setup changes, breakdowns, etc.

Here’s why you must consider ALL downtimes as downtimes
I can get an artificially high Availability of 95 % by excluding some down times, and making the numerator in the Availability equation go up. However, if I do this, I’ll just assume that there is very little further improvement possible, and just sit back and relax. Let’s say the setup time for a part on a CNC lathe is 2 hours. If I do not consider this as downtime, this setup time is going to remain constant for years together. If I do consider it as downtime, I’ll keep on trying to reduce it, by using quick change tooling, automatic tool presetter, etc. Similarly for other downtimes. The fact is that there are a huge variety of OEE formulae for calculating A,P, Q and OEE. Each firm defines its own cooking formula, and does the OEE calculation to get OEE numbers to suit its own taste.

OEE formula – examples of cooking up numbers
The formula of OEE calculation can devised in various ways to cook up good looking OEE numbers. Here are some recipes, and what each dish tastes like.

Cooking up OEE formula - recipes
Recipes for cooking up OEE formula

OEE Recipe 1

In the Availability part of the formula for OEE calculation, do not consider setup change times as downtime, because “How can you consider this a downtime ? After one batch of parts is done, you have to do a setup change and it takes time, right ? So it’s actually part of the cycle”. The taste: Maybe you can reduce a 2 hour setup time to 15 minutes by the use of quick change tooling, or by using a tool presetter, or by work and tool offset probes. You’ll never do this, because you have hidden this downtime waste forever and will never know that it exists.

OEE Recipe 2

In the Performance part of the formula for calculation of OEE, make the part change time a part of the cycle time, because “How can you consider this a downtime ? You have to unload the completed part and load a new part, right ? So it’s actually part of the cycle”. The taste: Maybe you can reduce the part change time by using a pallet changer, or with better and faster clamps on the fixture, a better and faster crane, or a robot. You’ll never do this, because you have hidden this manufacturing downtime waste forever and will never know that it exists.

OEE Recipe 3

In the Availability part of the formula for OEE calculation, do not consider chip cleaning, insert changing and part inspection times as downtime, because “All these are part of the process. They are not downtimes.” The taste: Maybe you can eliminate the chip cleaning time by using better chip breaker geometry and cutting parameters, reduce the frequency of insert changes by using an insert with a better life. Maybe you can speed up the inspection or reduce its frequency. You’ll never do this, again because you have hidden these downtime wastes in manufacturing forever and will never know that they exist.

As you can see, the taste in all these recipes is uniformly bad. These are only some examples of bad practices in the OEE formula. There really is no point in hiding downtimes to inflate OEE numbers. This comes from an obsession with the absolute value of the OEE. Instead, use it as a measure of productivity that needs to be improved. Set a target for periodic improvement, not an absolute value. Just let ALL the downtimes be visible, out there, so that you will keep reducing them and never forget them.

This is the key tenet of LEAN – be honest about your wastes. Do not hide them.

OEE target: Absolute vs. incremental target

You have deployed a machine monitoring system to go down the Industry 4.0 path. The system is working, you are getting data, and now you want to start using the OEE numbers to improve productivity. You have two options:

  1. Set an absolute target OEE number, and work towards achieving this number with a convenient OEE formula.
  2. Set an incremental target for OEE improvement, and work towards achieving this honestly.
OEE formula and calcuation - absolute vs incremental OEE targets

What happens in absolute target OEE
  • You set a target like 70 %.
  • People down the line in your organization cook up number with a convenient OEE formula. Example: They consider most downtimes as part of the process and not as downtimes – like setup time, inspection, part load/unload, insert change, etc.
  • Everyone focuses on achieving the target OEE number by fair or foul means (usually foul, because fair is difficult).
  • Once this number is achieved, everyone sits back and relaxes, thinking they’ve done their job and that there is no more scope for improvement.

What happens in incremental target OEE
  • You set a target for improvement, say 1 % per month.
  • It does not matter what the starting number is with the formula used. The starting number could be 70 % (when in reality it is 45 %), but nobody is looking at this at all. What everyone is focused on is the target 1 % increment.
  • People can make up any convenient formula of OEE that they want. You are however looking at the increment, and the increment cannot be cooked up, because the same formula is used every month.
  • There is no target OEE number that people can achieve and then go to sleep. The incremental number is forever.

Option 2 is obviously the better option, because it is independent of the OEE formula.

Don’t be obsessed with the absolute value of your OEE. This leads to people at various levels cooking up formulae for OEE calculation to show good values. Instead, use it as a measure of productivity that needs to be improved. Set a target for periodic improvement, not an absolute value.

Action point
Instead of getting an artificially high OEE number with a convenient OEE formula, patting ourselves on the back and sitting around complacently, we must get the true number and strive to improve it. See this page for what OEE really means, and what should be your correct method of OEE calculation.

Why do we cook up the OEE formula ?

We hear that there is a ‘World class OEE’ , and everyone tells us that we are no good unless we achieve this. ? So we strive to achieve this number, by any means available.
What is world class OEE ? Should you obsess over it ?

World Class OEE is 85 %. OEE is a concept that is part of the Toyota Production System (TPS). Toyota is an automobile manufacturer, so this standard was devised for mass production autombile manufacturing. To achieve this, you would need these numbers for Availability, Performance and Quality:
A = 90 % (10 % downtime between cycles)
P = 95 % (5 % downtime within cycles)
Q = 99 % (1 % part rejection)

OEE = 100 x (0.90 x 0.95 x 0.99) = 85 %

If your machines are doing mass production, by all means please strive to achieve World class OEE by fair means (not by cooking up the OEE formula).

If your machines are not doing mass production, you cannot achieve this number. In mass production the same part runs for months together. There is no downtime for setup change, the process is set, and various other downtimes can be reduced by daily analysis and fine tuning of the process and equipment. In small batch production your production batch sizes are small, frequent setup changes add hugely to the downtime, and the process, tooling and machines are not fine tuned to any of the parts that you are making.

In small batch production, OEEs between 50 % and 60 % are typical. The more your product mix and frequency of setups, the lower the OEE, Within a single organization itself, multiple plants or cells could be doing different types of parts, some mass production and some small batch production.

Instead of focusing on a single absolute value of OEE, it is best to use it as a measure of productivity that needs to be improved continuously. It could be just 40 % now, but the point is to increase it steadily, by say 2 % every month. Set an incremental target, like X% a month, instead of an absolute one.

So just use the proper OEE formula and do the OEE calculation as it is supposed to be done, do not cook up the formula of OEE and the numbers.

The harm of not being truthful in the OEE fomula

OEE is a really neat invention to measure how well you are using your machines, It has a direct impact on your profits. In a single number, it measures:
1. How much time it is running out of the total available time. Availability.
2. How efficiently it is producing in the time that it is running. Performance.
3. How good the quality of the parts are. Quality.
OEE = Availability x Performance x Quality.

To improve profits, you need to cut costs. To cut costs, you need to cut waste of machine capacity. OEE is a measure of the waste of machine capacity. The higher the OEE, the lower the waste.
If you are truthful about all your wastes, you get a truthful value of OEE.

If you are a manufacturer planning to expand capacity by buying more machines
OEE helps you estimate your true capacity when you want to take on more orders, and to decide if you need to buy more machines. E.g., if your OEE is 50%, this means you are using only 50% of the capacity that you have. When you are determining the number of machines to buy in a capacity expansion, you need to consider the fact that only 50 % of the additional capacity is going to be used to produce parts. This would be impossible if you are looking at numbers derived from wrong ingredients in the OEE formula.

If you are a customer looking to place a long term order for parts on a sub-contractor
You can use OEE to determine if the vendor is actually capable of making the parts consistently over the long term, and delivering on time. You will typically go to the vendor’s plant to check on the capacity, the number of machines they have. The vendor will convince you that they already have the additional capacity to take on your order, and show you the machines.

This is where you will have to use the vendor’s OEE numbers for each machine to determine what percentage of the available capacity they actually use. You cannot merely estimate the capacity to produce parts using the average cycle times and setup times of your parts. E.g., if the average cycle time of a part on a CNC HMC is 30 minutes and the available time per day is 22 hours, the theoretical production is 44 parts per day. If the OEE is 60%, this however drops to 26 parts. The capacity is therefore actually 26 parts per day. Again, you cannot get an accurate estimate here without knowing the correct value of OEE. You need to delve deep into their OEE formula and how they are tracking their OEE numbers to determine if the numbers they are showing you are accurate. Wrong numbers can harm your production plans greatly.


My (almost) daily encounter with lotus flowers

A few streets away from my home is a footpath shop selling lotus flowers, that I pass every day on my way to work. The shop only has lotus flowers and nothing else, they’re always the same colour, and they seem to get sold out by the evening.

Calculation of OEE - 1

The lotus happens to be the national flower of India, and for good reason – it is a stunning looking flower. It also has some peculiar qualities that very few other plant species have. The lotus has the ability to control the temperature of its flowers between 30 and 35 degrees C even though the ambient temperature may be higher or lower, just as humans and other warm blooded animals do. Scientists think this is for attracting insects for pollination. The flowers are hermaphrodite (have both male and female organs) and are pollinated by insects. The flowers and leaves are water-repellent, and exhibit something called the ‘lotus effect’. They never get wet, and water droplets roll off them like balls of mercury, washing away all dirt.

Lotus flowers in pond
Lotus stem

The roots of the plant are inside the soil at the bottom of the pond, and the leaves float on the surface. The stem has holes that run right through, providing a pathway for air from above the water surface to reach the underwater parts. Dried lotus seeds (Makhana in Hindi, Fox nut or Gorgon nut in English) are a great snack, lightly roasted with some salt and a couple of spices. The lotus stem (Kamal Kakdi in Hindi) too is eaten in many parts of the world. In India it is made into sambar, sabzi and pickle.

All in all, a remarkable plant !

About his blog from LEANworx: Plug-and-play Industry 4.0 system for MSMEs.

Share :

Leave a Reply