All you want to know about


17 Questions about the role of OEE within Continuous Improvement

1. 6 Steps to make waste visible by using OEE

Don’t compare OEE… make waste visible!

So to compare OEE is not a good idea?

Heriberto Ramirez •  Ok I got you, to compare OEE of our machines or even plants in different countries is not a good idea…

In the end what we really want is to make the waste visible and drive improvement where is needed.

Arno Koch • Ok Heriberto, this is a great starting point!
So let’s agree that comparing OEE’s is not exactly fulfilling this need.
How could you fulfill it, what could be done?

How to make waste visible by using OEE?

  1. Make sure that on the equipment you want to focus your improvement activities all (ALL) losses are visualized. Equipment that cannot be supported at this point is not yet bothered. The OEE industry Standard will guide you, in order to get as many as possible losses on the table.
  2. Take away any ‘treat’ that may be experienced by the crews who are exposing themselves and becomes vulnerable. Instead support them and encourage them in being honest.
  3. So if an OEE is low and the crew can exactly show where all the losses are: reward them! They just gave you a huge present namely an identified potential for improvement! When a team is showing of with window-dressing high numbers and barely any losses; challenge them! “Look better, find us the losses!”
  4. If you now see losses clearly, study how the improvement proceeds. Ie setup times have been improved by 50% on machine A and 5% on machine B. What was different on machine A? What can team B learn from team A. Now instead of threatening teams (this is what comparing feels like) you support an open communication. Remember; the enemy is not called plant B, it is called Competitor X.
  5. Challenge teams to halve whatever loss they found. And when they did so: offer them a stage where they can present how and what they did to get this result. The psychological effect is immense. Reward structured loss reduction activities, discourage improvements by ‘shots from the hip’ or adding space, people, money or complexity.
  6. You want to compare? Compare the amount of successfully closed improvement activities (halving a loss without investment etc.) Compare the amount of certain losses reduced, in a way they themselves thought of as ‘honest comparison’. Compare the depth of analysis between teams, the stability of OEE, and maybe the height and stability of quality rates.

Don’t compare OEE – Compare losses reduced

But whenever you compare, ask yourselves: “What will be the result on the behavior of the teams”. If it is not supporting real improvement: refrain from it.

There is much more to say about this subject but this should give you enough hints to find your way. Let us hear how your company is going to proceed on this?

Also read:

Comparing OEE’s

World Class OEE

Continuous Improvement

List with expected world class OEE’s

Evaluating performance of machines

Shopfloor Management with OEE

2. How can OEE be abused? How should it be used?

Q: How can OEE be abused, how should it be used?

Malcolm Jones •    Overall Equipment Effectiveness (OEE) is fast becoming a widely used measure for manufacturing industry, but it is also one of the more misunderstood and misused measures and causing much confusion.

What is OEE for?

The simple answer is “Improvement”. OEE is an improvement measure and is used as part of the improvement cycle. Unfortunately, much is made of the 85% ‘World Class Standard’ an arbitrary target found in the original TPM literature. Not only is this target out of date (Nissan in Sunderland are running welding lines at 92-93% OEE) it gives the wrong message. A customer has no interest in your OEE – that is an internal measure, which relates to your efficiency and costs. The customer is far more interested in a measure such as On Time In Full (OTIF) ie did I get my order? Running a manufacturing business on an arbitrary efficiency measure rather than a customer satisfaction measure is a recipe for disaster. The best use of an OEE target such as 85% is to recognise that if you are reaching that level and the customer is still not getting his orders on time, then you may have a capacity constraint.

How can OEE help?

OEE does not tell us if we have a problem, the customer does. What OEE does do is help us analyse the problem and make improvements. This is why Toyota use it as a spot measure on a particular machine where there is a capacity or quality problem.  Calculating the OEE of anything other than a discrete machine or automated line is pointless; we have far better measures of the efficiency of a factory or department as a whole.

OEE developed out of the need for improvement groups to have a way of measuring and analysing equipment problems as part of their Define, Measure, Analyse, Improve, Control cycle. OEE defines the expected performance of a machine, measures it and provides a loss structure for analysis, which leads to improvement. It can then be used as a tracking measure to see if improvement is being sustained ie if control is sufficient.

What does OEE measure?

At its simplest, OEE measures the Availability, Performance and Output Quality of a machine.


A machine is available if it is ready to produce, as opposed to being broken down or having some changes or adjustments made. The definition of availability allows for planned maintenance, when the machine is not meant to be available to production, but makes no allowance for changeovers etc. No machine with changeovers can ever be 100% available. The reason for taking such a hard line is that changeovers are a major loss to both efficiency and flexibility, so the OEE analysis focuses attention on it by making no changeover allowances.


Performance efficiency measures the output during available time compared to a standard. Here there can be debate about what the standard output should be. A good rule of thumb is to make the performance calculation based on best known performance. This may be greater or less than design speed. My argument is that if a machine has never reached its design performance it is not helpful to measure against that. On the other hand, if it has consistently outperformed the design spec you can have (and I have seen) performance figures of 140%, which can hide poor availability. This is always remembering that one purpose of OEE is to help tell you if you have the capacity to meet customer demand.


Output Quality is a First Time Through measure – what percentage of the output was right first time, without any rework. FTT measures are always the best quality measures. The issue in OEE is that sometimes the quality feedback is not immediate. In FMCG businesses, a customer complaint can be received three months or more after production. In these cases it is best not to include quality in the OEE calculation and use a more customer focused measure for quality – number of complaints etc. If there is no way we can use the Quality component of OEE in a real time improvement cycle, then it is pointless to measure it.

Loss Analysis

The next level of analysis is the seven (or six or eight or sixteen) losses. Within OEE we usually talk about seven losses, although TPM loss structures have been known to define 23 losses in all.

Availability losses

Availability losses are primarily Breakdowns and Changeovers. Changeovers can be separated into Tool changes, Material changes and Reduced Yield at start up, but fundamentally these are the same issue. Further analysis reveals breakdowns to have two fundamental types, those due to deterioration because of inadequate maintenance and those due to inherent machine characteristics.

This gives us three basic responses to availability issues – improve changeovers through SMED, improve basic maintenance and improve machine characteristics. Depending on the Pareto analysis of losses we may need to act on one, two or all three of these.

Performance losses

Performance losses are usually separated into speed loss and minor stops – is the machine running slow, or is it stop-starting? The definition of minor stop is also open to debate – originally it was less than ten minutes, then five minutes, then three minutes. The pragmatic approach is to say that if you can measure the amount of time lost for a stop it is a breakdown, not a minor stop. If you can only record the quantity of stops, then they are minor stops.

There is some practical use for the speed/minor stop distinction – if a machine is running slow we can always speed it up, whereas if it is jamming we need to look at the physical mechanism and try to remove the cause of the jams (my favourite example is where we found the root cause was when metal washers were being loaded into a hopper with a metal shovel, which damaged some, which then jammed the feed – the solution was a plastic shovel!).

We can however also make a useful distinction between performance losses due to deterioration or contamination and those caused by inherent machine characteristics. As with breakdowns this gives us two improvement approaches – better maintenance or equipment re-design.


The only reason to measure and analyse anything is to improve it. If we are not going to use the whole improvement cycle there is no point in measuring OEE. It tells us nothing we do not already know. At a gross level all OEE tells you is how much you made compared to what you wanted to make, and any schedule adherence measure would tell you that already. Averaging OEE’s over whole plants or time periods just hides issues – OEE is a specific measure for use in specific improvement projects.

The biggest misuse of OEE is to use it to compare different processes, plants or machines. OEE is not a useful executive KPI. It is not even a very useful operational measure. It is an improvement measure, for people who want to improve their equipment performance.

How to massage your OEE

  1. When the machine breaks down, log it to planned maintenance
  2. Do changeovers during planned maintenance or at weekends if not 24/7
  3. Use an easy performance standard
  4. Measure the best machine and quote that figure
  5. Set arbitrary targets and achieve them through the above

Using the above strategy you should be able to report decent OEE’s and even make some money if pay is OEE performance related. What this will not do however is improve your ability to meet customer demand.

How to improve performance

  1. Measure against customer demand (OTIF or similar)
  2. Measure OEE on constraints or problem equipment
  3. Set realistic performance standards
  4. Analyse losses to identify issues for improvement
  5. Use the whole improvement cycle
Biography Malcolm Jones founded Productivity Europe in 1989 to develop support and facilitation services in World Class Manufacturing techniques. He learnt from Japanese masters such as Shigeo Shingo and the Total Productivity group at the Japan Management Association, and has edited three books on World Class Manufacturing techniques and practices. Productivity Europe are leaders in lean manufacturing training and consulting in the UK. Our experienced consultants help you establish a World Class Manufacturing vision through Lean Manufacturing, Total Quality (Six Sigma) and Total Productive Maintenance training.

More info:

Evaluating performance of machines

Can 85% OEE be NOT world Class?

Compare OEE when defined according the OEE Industry Standard

How high should OEE be?

World Class OEE

Comparing OEE’s to benchmark machines

3. (How) can Data Collection become part of the improvement process?

Q: We measure Productivity. Now we start with OEE as easy as possible. How can I arrange that workers are not just filling out files with data. Currently we measure productivity/present time.

Arno Koch •    It is not the data-acquisition that is the most difficult part in OEE. Most OEE implementations fail or show a lack of power because what is (not) done with the acquired data. Or, even worse, the data acquisition process itselves already destroyed the process of involving the people that would have made the difference.

Data Collection involving the shopfloor

If we listen to most software suppliers, one might think it is just a matter of installing that beautiful IT solution, and the world of OEE is bright shining and improvement comes automatically.
Unfortunately this is not how real life with real people works. OEE implementations are hard work, a serious management task, that needs a good deal of understanding in human psychology and skills in changing team dynamics and human behaviour.
So, the first quest you’ll have to face is this one: “HOW is the data-aquisition of my OEE data going to influence WHO in my improvement process, and what will be the CONSEQUENCES.”

Who reports what of OEE, when, why to who?

The next question is: “Who is going to report what information to whom at what moment, resulting in the behaviour needed to realize the desired change” (aha, what was the desired change exactly? Are you sure that’s it?)
If you take those quests really serious, OEE might, just might be your (yours? or whoms?) compass towards the desired (which?) improvement.
Puzzled? Right! That is what OEE should do to achieve its real power!

4. Our people don't like to collect data. How to address this?

I am an industrial engineering supervisor. I work for an automotive sensors, solenoids, and switches supplier for the automotive market. We are 53rd out of the 150 top suppliers.

I have some questions for you regarding OEE implementation and software:

Data Collection

Q1: Data collection can be a problem (Garbage in Garbage out) and the best of software solutions can fail for lack of complete and accurate data. Production people see any form of data collection as a hassle. How does OEE address this?

Arno Koch •    Instead of adding an extra hurdle, lets find a way to reduce the operators burden,  by gathering ONLY the data that is at the bottom line necessary. Normally the operator collects a lot of data that is not leading in any way to improvement. By eliminating such registrations it should be a relieve to the operator.

OEE is a real Operator tool that should help him or hear to do the job with fewer problems. It goes a little to far here, but the book “OEE for the production team” tells all about it. Maybe you want to join one of the workshops in Vegas September 99, I will be telling you all about OEE.

5. How to make Continuous Improvement Activities a part of daily work?

After the implementation: Continuous Improvement

Q: How to make Continuous Improvement Activities a part of daily work?

This question regularly is being asked when companies join a Lean– or TPM journey.

Arno Koch •  Continuous Improvement indeed should not be a project, an incidental activity or a staff-member job. So how to engage everybody in the factory to get involved in improvement activities on a daily basis?

The role of management

First of all it is the task -and main purpose of being- of the complete management crew to Govern Change.

Concrete this means the management has to design, create and establish a work-environment where people CAN and shall perform their task in a solid, accurate and smooth way.

Work to the rule

Unfortunately, in many companies it is not because, yet nevertheless the working environment, the procedures, the processes, that the job is being done… Otherwise how can it be that the workforce can threaten the company with  “work to the rule”…

In other words: Although it is organised as it is, people manage to perform their tasks, having to ‘be creative’ and even break the rules all the time in order to succeed.

What is your Value Stream?

As a manager, question yourselves how deep your basic knowledge about your value-stream really is. Do you have detailed insights what when where happens? Do you really know what is happening at the shopfloor; at the places where the actual value is being created? Would you believe when you where told that probably over 95% of your main resources do NOT lead to value addition?

Having Continuous Improvement Activities become integrated as part of daily work is primarily a task of the management. That’s what governing change is all about. Having ‘to govern the change’ in order to achieve serious improvements, management needs to be seriously concerned about 4 questions:

  1. What are we aiming for, where are we heading?
  2. Where are we now AND what is in between ‘now’ and ‘then’?
  3. How to eliminate that gap?
  4. How do we sustain that improvement?

1. What are we aiming for, where are we heading?

Having a True North

Can you be specific when being asked:

  • Where is your company/department/line in 3, 5 and 10 years?
  • When I come here in 3 years, what is the first thing I will notice? What will it look like, how will I feel here, what will our customers say about us, … etc.
  • How will our people know what to do, whether they are on track?
  • How do we align our peoples needs with our companies needs?

Governing Change begins with having a clear picture about what you are changing into. Continuous Improvement starts by being able to clearly communicate this, in a way that those who should actually realize that change can fully identify with this picture.

2. Where are we now AND what is in between ‘now’ and ‘then’?

Visualizing all losses

When it is clear what my machine, my line, department or company should look like, what challenges it should be able to face, the next question is: “Where are we now and what is in between”.

This is a crucial part in managing change: If we forget to analyse- and visualize the gap, it will be very difficult to proactively work through this ‘hidden landscape’.

OEE and Makigami are good startingpoints

At a conversion-level here we use OEE; this instrument, when well applied, will inform us about the dynamics of the actual value creation process. It will give us insight what can be done to stabilize and improve it. It will also lead use (via 5 Why’s) to the rest of the process and organisation.

Valuestream maps and especially makigami-process mapping will reveal the losses and hidden potentials within our organisation. These methodologies will first of all work for us as learning tools, where the workforce ánd its management will gain a unified and deep understanding what is actually happening in this (according to the ISO book) so smoothly looking processes that are nevertheless consuming so many resources and causing so many problems. (Check: How is the ratio between direct- and indirect personnel in your company? Why is that?).

Ask for loss visualizations!

As a manager it is your task to ask for such visualizations and to allow the people involved to analyse and report the losses on a regular basis. (Instead of having others -that do not really know the works-, tell them how they performed). Please feel the difference:

  • When YOU can tell your management in detail about your value creation process and what you need to perform better, versus
  • When a staff-member or manager tells you how HE thinks you performed

In the first case you might discuss what- how- and when you need support;
in the second case you will most probably argue the numbers, start defending or explaining and get frustrated in the end, feeling not understood…

3. Continuous Improvement:

Continuously Eliminating Losses

As soon as we can see what is hindering, what is between where we are now and where we are heading for, we can start the improvement. Improvement in that sense, mainly is not a matter of doing new things. It is about eliminating those things that do not really matter, that do not contribute to what we want to achieve at first. A well designed OEE and a well performed Value Stream Analysis (including a deep loss analysis!) tells us quite explicitly WHAT should be eliminated.

Involve the people in- and from the process!

This elimination can ONLY be done by the people directly involved, in a multidisciplinary team, having a 360 degree view on the situation. I promise you: There is not ONE person in your organisation that fully understands whatever process you are going to investigate! (Even if you just received whatever ISO/QS certificate…)

This team needs to learn to work in a structured PDCA way and must be capable of doing decent phenomenon descriptions, true root-cause analyses, draw Operating Principles and make process designs. Every member of your company can learn this. Visualizing, analyzing and eliminating losses ought to be an essential part of each employees work description. It is a perfect learning instrument and will result in a continuously improving the organisation.

Be though on WHAT, keep away from HOW

It is the task of the management to be clear about WHAT should happen (“eliminate losses in order to achieve our goals”). It is that task of the people that execute the process to improve it, while the management will facilitate such activities as an integrated part of their jobs.

So here is one of the difficult cultural changes: Instead of assigning improvement tasks to staff members or even external parties, these resources (mainly FTE time) is given to the actual performers of the task. 

They will now learn how to do this, and by doing it they will gain a deep understanding of the functioning of equipment, processes, and whatever subject is being studied to improve. And by designing the new way of working by the executioners themselves, this will create ownership, leading to a more sustainable improvement. (You now might believe they can to this: Let me tell you I have not yet met a problem those teams could not solve…)

4. Standardizing optimal conditions

Making the best way the easiest way

As you will be aware, the last step of a PDCA team (you may call them Kaizen Groups, Small Activity Groups,  DMAIC cirkels) is always to create sustainability by standardizing the new way.

Far too often this is ‘forgotten’ or skipped because ‘there was no time left’. Yet this step again is a key to continuous improvement being part of the daily job.

The best way is the only way

First of all there should always be just one best way to achieve the best result given a certain situation. And that best way is designed and determined by the experts in the field: the people who perform that task. Finding the best standard is -again- a learning event. Here a multitude of solutions may be possible, yet just one will pass the critics of the experts. And they will find a way how to sustain it. Although that might look easy, this proves to be a harsh task…

Standardizing is not a matter of writing procedures: in the optimal form there is nothing written at all: it is embedded in the process, usually in a visual way. Compare it to the parking lot. You do not have to read a guideline to know what is the best way to park your car. No driver ever thinks about this, yet imagine how difficult it must have been to design the optimal layout of the parking lot (and feel what an effect is has if this job has not bee done properly)!


Embedding continuous improvement activities in daily work is not a matter of applying a few tricks, learning a few tools or assigning some tasks. It needs a clear vision of what you want to achieve with your continuous Improvement.  How are you going to design and establish an environment that engages everybody in the company to contribute to this achievements? This is your task: to Govern Change!

See also:

6. We want to see our progress. Should our software report the trend?

Trend Data

Q: I do agree that what gets measured gets improved. Software does a good job collecting data and giving it back in reports. What is lacking is trend data. What should good OEE software report?

Arno Koch •    The first OEE software I developed (the ‘OEE Toolkit’) was designed to report the right information that you need to get real focussed improvement. It reports always based on the three issues as told above (plus more if you do it clever, and good software should do so…). The quality of OEE software is determined by the quality of its analyses and its ability to influence the behavior of its users. Unfortunately most software is designed from an IT point of view, rather than from a ‘improvers’ point of view.

Labor Effectiveness

Q: Hmm, but our main measure of productivity is pieces per person hour and labor dollars per piece.

Arno Koch •    So you measure the result of your productivity. Let me ask you a question: Do you know how much potential productivity you lost? And can you pinpoint where precise you lost it? And what to solve first?

Q: To this we attempt to not speed up production but rather to reduce labor. Unless the capacity is needed – then we recalculate takt time.

Arno Koch •    If you look through Muda Glasses, there is only one ideal takt time! Not running so means having a loss….

Q: To this end we kaizen cells, attack scrap and rework, and stabilize machinery. We currently track and trend all of these areas and have been since 1989. And yes we do believe that all this leads to driving down costs.
Profit = Selling Price – Costs

Arno Koch •    I would modify your algorithm a little:
Profit = Volume x ( Selling Price – Costs)
How to raise volume? Deliver your customer:
• Highest quality
• Lowest cost
• Highest delivery…

Again: Can you be sure you attack the right loss, where is the major LOSS (=cost quality delivery) to fix?

Hey, I am glad to have linked up with you (it’s a small world these days).

Nice chatting to you Bob!
I hope I could tickle your mind to ‘learn to see’ (Jones…et al.)

7. Should we optimize for maximal capacity or to follow demand?

Maximum Output or Demand?

Q: Lean does not strive for maximum output of the machine but rather producing to customer demand.

Arno Koch •    Sure! What does the customer want? The right quality for the best price at the moment he needs it. Right? Question is: How to do so?

  • Would it help if your machines would run when you need them?
  • Would it help if it had the output performance what it was designed for, so you can produce cheap and quickly?
  • Would it help if you would produce zero defects?

Right? OEE makes visible where it is NOT like that….

What means ‘Lean’? It means: Using Muda glasses: detecting all your losses and solve them. OEE detects the losses on the actual spot were it is all about.

Generating money by adding value… where is this done? At the machine… How to look for losses when actually adding value? OEE! Viola…

What to track?

Q: So, output of a cell (that can have a group of machines and people) is tracked hour to hour against takt time and downtime recorded and addressed as a part of the cells environment. Wouldn’t OEE in this case be counterproductive?

Arno Koch •    When you need to use your machine, your machine would be ideal if:
It runs without stopping, at maximum speed with no quality loss.
That is the definition of 100 OEE. Everything else is a loss. Big question is: Where am I losing?? OEE makes this visible and clear. And you are right: Some equipment has over-capacity. So if you need to reduce it’s speed because there is no demand… that is a loss.

But think about it: Are you able to run your machine on maximum speed? Mostly not, and mostly this is not desired. But than, are you able to run the machine in a controlled way on a lower speed, precisely on the needed speed to fulfill demand? Most machine cannot do so either… That is the REAL problem that needs to be solved!

8. Although a high OEE, we have a low Planning/Schedule adherence

Q:  It is theoretically possible to have high OEE and very low schedule adherence, as OEE has no component to measure ‘are we making the correct product today?’.

Arno Koch •    This is not just possible, it is most often the normal situation in OEE implementations. Why is that?

To understand why OEE implementations sometimes tend to be counterproductive to the strive for creating flow in the factory in order to follow customer demand, lets try to find the root cause for the phenomenon;

First of all: What does “Quality” in the Quality rate of OEE mean? It means the product has been produced according to its specifications.

Now what are the specifications that need to be met in a World Class environment? The products needs to be produced according to its:

  • technical specifications
  • specification of delivery
  • specification of costs…

… and all those specifications need to be leveled against customers needs and expectations.

If those specs are not known, wrongly defined, or not applied within the OEE, this is not a problem of the OEE as a measure, but a problem of the use and implementation of the measure.

How to prevent low schedule adherence?

Step 1

Step one to prevent a misalignment between what is produced (due to sub-optimization at the machine) and what is required, is to make sure On Time Delivery becomes a quality parameter on the machine.

So if the product is not produced On Time, it is not within spec, the quality rate in OEE goes down due to not meeting the specification of  delivery time.

Now the assumption often is this: “If the scheduler has the right scheduling tools and access to the status of the current schedule execution by operations, and access to ERP to ensure that raw material exists, then scenarios can be modelled to conveniently place orders into the existing schedule to deliver it, and all the other production orders, on time.”

In other words: “If we produce according to the right schedule, it is OK.”

Unfortunately, this is rarely true. Why? Because the ‘schedule’ is a construct, an opinion, about what is the best way to produce the required orders, based on a whole series of assumptions.

The schedule rarely fulfills customers real demand (it is even worst: since most B2B customers also ‘schedule’, they do not order according to their real demand…) so even if OEE would take ‘schedule fulfillment’ as a Q parameter, this does not guarantee real customer satisfaction.

Conclusion: Even if meeting the schedule would be a parameter of OEE, it is far from sure that customer demand would be met.

Now the question remains: Why is it so hard to meet a schedule, even if the schedule is planned with the best tools and planners?

This is because reality is rarely the same as the assumptions within the planning. In other words: Whatever you plan in advance; it will be probably differently in real life… So than why plan in advance?

Instead of running the optimal planning, you might consider to run your optimal real life!

One of the reasons why the planning is usually far from ‘practical’ and ‘real’ is the fact that the planner and its software are not -or not tight enough- connected to the real life, not being able to respond adequate to changes and deviations to the plannings assumptions.

Knowing this, it is not surprising to see how typically when an order portfolio is planned by the production-team, it will run much smoother than one planned by someone outside of the team.

Step 2

In order to have as much knowledge about the current actual situation around the equipment available, let the production team it selves plan the details of a certain order portfolio. “Those are the orders you have to produce, these are the deadlines, feel free to decide how you run them”

The problem that may arise next, is the supply of raw materials. If the production team plans its own routine of production, how do we know what raw materials are needed when?

Of course we might ask the team to also plan its raw materials, in other words, the team takes over the complete task of the current planner, yet only for the part it needs to run the machine.

We then would still be confronted with variances in real life situations. A typical ‘push planning’ (MRP/ERP systems try to push the planning through the value-chain)  can hardly cope with this, unless we grow buffers, security stocks etc. Which then again create new problems.

Step 3

Create ‘supermarkets’ using 2-bin and fifo buffers to guarantee the delivery of raw materials to the line. Planners and Buyers should be able to define bin-sizes and arrange bin swapping mechanisms with the suppliers. If not they have to learn this.

Step 4.

Give the line as direct as possible access to the real customer demand, show them directly what has been sold to the user, so they can respond adequate to fluctuations in demand.

Planners now take care of CAPACITY planning and not about DETAIL planning. Orders are assigned to machines (based on rules agreed with the teams) and keep track of the flow of bins, the lines of fifo’s etc.

At this point a high OEE can only be achieved if a high customer satisfaction has been achieved, this without complex IT systems, multi layered planning schedules, etc.




9. is 85% OEE Really World Class? Or Not?

Q: What is a ‘World Class OEE’?

Dennis McCarthy •     When I say that Effectiveness is a measure of how well you did what you planned to do I want to be sure that the measure excludes time when you didn’t expect to run the machine.   For sure the goal is to maximise effectiveness.

85% World Class

The “85% OEE is world class”  comes from a TPM book written by Mr Nakajima who set out what he rightly considered to be world class in a machine shop in 1970’s.  This was based on zero breakdowns, 10% of time on set ups, 5% performance loss due speed losses and 1% quality defects.  That is still pretty impressive for a machine shop but even a poor process plant would run much higher OEE.
A more robust measure of world class is the OEE improvement trend. From a standing start a 50% increase in OEE over 3 years is a common achievement which often exceeds what they thought of as 100%,
The key for me is to keep it simple, aim for 80% of the answers with 20% of the data and use the measure to drive the improvement trend.  As you approach 100% you will also improve process capability and can increase the precision of your measure to raise the bar further.  OBE (Overall Batch Effectiveness) has a higher data capture overhead so only use it if you need to.

See also:

Can 85% OEE be NOT world Class?

List with expected world class OEE’s

How high should the OEE be?

Compare OEE when defined according the OEE Industry Standard

10. 85% OEE

Q: I have been reading in different sources about the OEE measurement and I have not been able to find any concrete agreement about what the value for OEE should be. Could you please give me any references? How high should OEE be?

Is 85% OEE world class?

I several places I read that there is no common agreement but that it is generally suggested that:

  • bellow 65% OEE should be considered unacceptable, since it represents a very low competitiveness and a great number of economic losses.
  • A value between 65 and 75% is considered as regular, only acceptable if it is being improved.
  • An OEE between 75 and 85 % means an acceptable level with slight economic losses and which can be easily improved to World Class levels.
  • A high OEE value between 85 and 95 percent is considered as good and equals World Class values; it represents a high level of competitiveness.
  • And finally, a value above 95% is considered as excellence; an excellent competitiveness and World Class values.

Arno Koch  •  In general speaking, I do not agree. There are many angles to approach this question. Let me try to elaborate a bit on it.

What number are we looking at?

First of all there is no such thing as ‘an OEE number’ that could be referenced to, simply because OEE can be defined in a thousand different ways.

How was the OEE defined?

Would you like to see an OEE of 85%? Well I’ll get it for you tomorrow on your machine! It is just a matter of changing some of the definitions;

Do not laugh! I have heard about consultants who earn money in this way…

What is the criterion for “good”?

So let’s assume we are going to use the OEE industry Standard for all machines. Now,

  • What is a ‘good OEE’ for a machine that continuously runs just one bulk product?
  • And what OEE is ‘good’ for a machine that runs 60 different products a day?

Stable OEE or not?

Let’s take the first machine, running just one product all day, feeding a line. Is 95% now OK?

  • We’ll not if the downstream equipment would only process the volume of 65%.

Now would 65% be a good OEE? No, not necessarily! Why not?

  • If the machine jumps fort and back between 45% and 95%, with an average of 65% OEE, it still would cause a lot of problems!

Producing Quality, or not?

Ok, now we have a stable OEE of 65%. Is it now OK?

  • No, the OEE is not OK if the machine runs at a quality-rate of 90% (or any other number that is not (near) 100%!

A high OEE at high cost?

We’ll lets go on: Now it is stable at 65% with a quality-rate of 100%. Now OK?

  • What if we need to make huge and expensive efforts to get this done? Maybe now your costs are so high you are losing money with your beautiful machine!

Understand what’s going on!

So you see, it just does not makes sense to focus on the OEE number as such. This is a typical western style of thinking that many managers suffer from. Deming repeatedly referred to this as the lack of profound systemic knowledge. In other words:

Instead of looking at a number,

one should understand really what is going on.

What to look for in your OEE?

Some important questions with regard to the OEE number could be:

  1. Is there an OEE level we can run the machine stable and reliable? I.e. can we run continuously between 40 and 44%? I would call that already a hunch of ‘World Class’ since I have rarely seen such machine! It would be an indicator for a process that is -or can be- in control.
  2. Is there a range of OEE in which we can run a stable OEE; I.e. could we run any desired OEE between 20 and 50% on demand for days. (Wow, ever seen that?)
  3. What OEE is desired to fulfill demand, and are we able to run such an OEE on this equipment?
  4. Are we able to run the machine without rejects?
  5. Can we run the machine without any unexpected interruptions?
  6. Are we able to setup from one product and the other, following demand flawlessly?

Understand the process and its losses!

To go short; stop focusing on the height of the number. Start to focus on those components that may be indicators for an understood- and controlled process. As a result, the number will go up, costs will go down, etc. 

How high are OEE numbers normally?

In my experience, measuring over 2000 machines in all kind of branches and continents, most machines will not exceed 35 to 45% of OEE. This assumes the OEE Industry Standard definitions where applied.

In some branches (like pharma) the typical machine runs far less, while in others (like –some- automotive) it can be higher. Some particular types of machines (like extrusion-machines) tend to have higher OEE’s by the nature and the stage of development of the equipment.

After some years of TPM implementations and applying lean principles to prevent the loss of flow, equipment (again very generalized) may grow in the 60’s or even 70’s which (as a number…) is quite good.

What about the economic losses?

Let’s take my statement about the average machine in the average factory running 35-45% OEE. Are they running economic losses? Most of them earn quite some money. Even worse: The conversion cost (if we may believe the controllers, which you should be very careful with) is mostly just a small part of the total cost. This explains why such low OEE’s are commonly accepted (otherwise they would not be there…)

Speed, reliability, flexibility

The real economic loss is not in the height of the OEE, it is in the lack of flexibility and reliability of it. Think about it; what would it mean to the cost of the supply-chain if every step would immediately respond and follow the demand. Just imagine the economic consequences….

More info:

Evaluating performance of machines

Can 85% OEE be NOT world Class?

Compare OEE when defined according the OEE Industry Standard

Use and abuse of OEE

World Class OEE

Comparing OEE’s to benchmark machines

11. 7 Reasons not to desire 85% OEE

Q: Usually, 85% OEE is considered to be ‘World Class’. You say it isn’t. How come?

Situations where 85% OEE is not World Class

Arno Koch •    I’ll give you several reasons or situations when 85% OEE is FAR from World Class:

Situation 1. Not all losses are included in the definition of the OEE

It is possible to run any machine at 85% OEE from today, simply by adapting the OEE definition… Just take out breaks, maintenance, setups, all kind of waiting, or calculate OEE with a maximum speed that is lower than theoretically possible (the so called ‘realistic speed’) and hoopla: there is your 85%. Don’t laugh! This is not exceptional! Most definitions I saw where ‘manipulating’ OEE by hiding all kind of losses, resulting in a too high OEE.

Situation 2: Quality rate is low

  • As long as quality rate is not extreme high, the process obviously somehow can become unstable. The quality problems are probably only the tip of the iceberg. After all, if the ‘system’ could handle this, there would not be any quality problems.
  • The costs of the scrap products are high, even for a ‘cheap’ product such as egg cartons, for example (no disrespect intended for this fine product!) Just take a moment to calculate how quickly your profit margin erodes when you produce more scrap! Once you have passed that tipping point, yes, you are making more products, but in fact you are quickly incurring loss! You are throwing money away! This can absolutely occur at a ‘beautiful OEE of 85%’, in which case you are throwing away your money almost continuously and at high speed…

Situation 3: OEE fluctuates, process is not in control

An average OEE of 85% does not mean the machine runs at 85% all the time. If the OEE fluctuates between 30% and 100% OEE, in the end still 85% might remain on average…. This will be a very expensive process (meaning low efficiency).

Situation 4. Stock increases

Running large batches usually enhances the OEE; less setups, ‘keep running once it runs’….

But (if) this is not what the customer demands; we will increase stock, thus causing tons of negative side effects. In this situation 85% OEE can never be called world class, in the contrary!

Situation 5: High effectiveness is achieved at low efficiency

If the 85% OEE is achieved by investing heavily or by increasing the used resources, the costs will go up. Whether this is a good idea -in economic terms- needs to be calculated. From a point of view of improving by loss reduction, this is the wrong way to go.

Situation 6: High effectiveness is achieved without binding the workforce

Although I would not directly see how to do so, it might be possible to get the equipment to run at 85% OEE without involving the workforce around it. If well standardized, this might even be sustainable. However since the workforce was not involved the question will be:

  • Will they have enough understanding of the process to correct i.e. abnormal situations (resulting in a crash when this occurs)?
  • Will they be able to give input to an improvement process? In other words: probably the process will not be able to grow, based on workforces’ knowledge.

Since we all know how difficult it is to create standardized- and sustainable processes, it will be most plausible that this situation leads to another effect:

Situation 7: 85% achieved, but it is not sustainable

We know that when ‘the shit hits the fan’, all is possible. If 85% OEE is achieved by pushing and tricking, it will be highly questionable whether this will be sustainable and ‘in control’. This type of situations is potentially dangerous. It seems everything is OK, but actually it is a ticking time bomb. Machine crashes, accidents, or sudden loss of capacity or quality will be waiting to occur because there is a situation that is under stress and just waiting to become out of control.

Resuming: This all means, there are situations where running 38% OEE might be far more world Class than running 85%…

Again: The OEE number it selves, seen without its context, is a non-number, it has no value as such.

More info:

Evaluating performance of machines

How high should OEE be?

Compare OEE when defined according the OEE Industry Standard

Use and abuse of OEE

World Class OEE

Comparing OEE’s to benchmark machines

12. What does OEE NOT show?

Q: What does OEE not show?

Daniel Högfeldt •    OEE is a good way for looking at a single machine’s performance. However it does not really say how the entire plant is performing. It says how the plant is performing based on how much each machine was scheduled to run but not based on how machines are utilized in the plant. It does not say anything about if the plant makes a profit or not. An example on how the situation could look like in a plant is shown in example 2.

Example 2: How a plant could look like.

Let us say that the plant has 15 machines on the shop floor. They used to have many orders and ran all the machines every day. Their overall plant OEE used to be 63%. These days the orders have dropped and they only run 5 out of the 15 machines. On those 5 machines they have a plant OEE of 90%. Is the plant doing better now than it used to?

OEE would not say anything about the situation in example 2. If that number is the only thing that is looked on then management could think that the plant is doing a lot better now than it used to, 90% OEE now and 63% OEE before.

Of course the plants are looking at what kind of profit they are making, but looking at the OEE number will not provide any other information than how effective the machines/plants are in relationship to the planned time it had to operate.

There are however easy ways to include some utilization calculations on the same sheet as the OEE calculation is made (Koch 2003).

OEE is not the perfect tool for comparing plants. The corporate does not really have control of what is put into the calculation. The number is very easy to manipulate. One example could be if a plant has a low OEE number on some machines but want to invest in new machines because of new orders coming in. Then the plant could change the ideal cycle times to get a higher OEE number and then claim that they need new machines, while the truth could be that all they need to do is to work on reducing waste on the machines to get more capacity.

Master’s Thesis: PLANT EFFICIENCY A value stream mapping and overall equipment effectiveness study

Download as PDF

13. Is there a list with OEE Benchmark numbers?

Arno Koch •    OEE benchmark Lists: Don’t even think about it! No, NO one that understands OEE will set up such a list. If you find one it is useless since OEE’s in general can not be compared. And it will tell you someone was trying to compare apples with cabbage.

Comparing Apples with Cabbage

This topic has been discussed more often; please start reading at:

How high should the OEE be?

—————–Q: Our management is looking for one universal tool for OEE calculation so they can compare the performance of equipment. How do you feel about this?

Arno Koch •   It is a persistent wish of many managers to request to compare the OEE’s of different machines. Unfortunately this seems to make sense, but actually not only leads to wrong conclusions, but also to counter productive behavior.

What does is to to the team?

OEE is primarily an operator tool to support operators on getting focus on the losses of their equipment. They have tons of knowledge of the equipment which can be used for structured improvement through Small Group Activities. Although OEE information can come handy for managers, they should never use it as a stick to hit shop floor.

Punishing people

Why? Because in such case managers are punishing people for making an effort on improvement (even if this is ‘just measuring’ OEE, which is the basic need for TPM structured improvement). Instead managers should make any effort they can, to support shop floor teams to improve the OEE;

Basic rule: No Blame, No Excuses, Total Commitment.

So the answer to your question depends on the intention of your management.

The power of TPM is in the fact that it visualizes where the losses are, it focuses everyone on the right things to do and it offers tools to get everyone’s involvement in a structured improvement strategy. If the OEE is only used ‘on the day of reckoning’, so as a punishing system, I don’t think you will ever manage to build teams that are showing pride to their equipment and are having fun to make it run to the best performance; for that is what good functioning TPM teams can be recognized at!

A2 •    Bluntly comparing OEE’s of different machines in different processes in order to ‘benchmark’ is at least very tricky, maybe I would dare to say it is just silly! Even from a economical point of view.

The silliness of comparing OEE’s

First of all most OEE calculations are being made based upon incomparable definitions. I.e. in one case a break is affecting the OEE in the other it isn’t. Of course the way the maximum speed is being defined alone is already enough reason to have at least doubt about any comparison. But even if you would have fully standardized ways to set up the definitions; just look at one of a thousand examples I could imagine: A machine has an OEE of 55%, another has 40%. Looking into detail you see the 40% machine is running at a very stable level; day in day out the same OEE. Sometimes it stands due to planned maintenance and than it runs smoothly 40% OEE at 99% quality. The other machine is one day running fast and the other day slow, the third day it breaks down at a very inconvenient moment and the next day it runs at 95%. It drives everybody nuts but on average it has an OEE of 55%. So which of the two would you prefer??

Another reason why it just comparing OEE’s can lead to wrong conclusions; take the same two machines from the example above and lets look at the people and the improvement process. The 55% machine has a team that is ‘just trying to keep it up’ (hard enough, even very difficult; running around all day, tired like dog at the end of the day but no structural improvements and tomorrow it starts all over again…) the other team started with an OEE of 15%. They formed a production team with an engineer and a maintenance guy, they measured their losses and they started to improve their production process through small group activities in which they even invited people from the purchasing department. At the moment they work on a plan to improve OEE from current 40% to 48% which will be presented to the management next week.

What do you really want?

What kind of operators and teams would you prefer? Would you compliment the 55% team for showing a higher number?
Meaning: The number itself is not that important. If you look at the number you should look at the number for a period of time; is it stable or unstable? Is there any improvement? How is the number composed? Even two identical machines with an identical OEE may face a complete different situation if one is running faster and has a lower quality-rate, and the other runs slower with no quality losses. Both might have the same OEE…

But maybe the most important issue about the number is: is it a number, or better a tool, to the production-team? Does the operator understand it and lives up to it? Is it HIS or HERS? Because in the end you can not have a successful TPM implementation without operator focus on the losses!

OK, I’ll give you óne more thing to consider….

Day one the same machine, same team, same product runs 53%. The next day, same situation, again 53%. Now was it equal?

Day one had 80% availability, 70% Performance, 95% Quality

Day two had 70% availability, 95 Performance, 80% Quality.

Tell me, what did your management would have missed when comparing OEE’s …

14. Can we compare OEE when defined according to the OEE Standard?

Q: I defined OEE according to the OEE Industry Standard. Can we now benchmark our machines?

Arno Koch •    This question is popping up over and over again. Obviously the thinking is: When we standardize the definition of the metric we can use it to compare equipment and even factories.

Compare two shifts with 56% OEE…

Imagine the early shift of machine X has an OEE of  56%.
The late shift on this machine also has an OEE of 56%.

How do those two shifts on the same machine compare? Are they equal?
The early shift was running an availability of 75%, Performance of 85% and quality of 95%.
The late shift was running an availability of  85%, Performance of 95% and quality of 75%.

The same OEE but on top of the equal amount of good product, the late shift produced a huge volume of scrap. Same OEE, different cost-price.

Higher OEE having a negative effect…

There are many situations where a higher OEE has even a negative effect:

  • The OEE was higher, but the process became unstable, less reliable
  • The OEE was higher, yet there was no demand and stock increases
  • The OEE was higher because the crew was pushed over the limits causing stress, illness, risk of accidents etc.
  • The OEE was higher, yet the cost also where higher due to additional resources applied.


Even on ONE machine with an identical definition of OEE it is not possible to compare two OEE numbers without knowing the parameters behind the number.

And even when knowing those parameters it is still difficult. Was the shift running many small batches or one continuous batch? What products are allocated to what machine and shift?  Just to name some reasons for fluctuations.

The question now is: Why does everybody want to compare machines and factories? How will that lead to improvement? Remember what Dr Demings answer was?

This standard aims to offer a methodology that breaks through this ‘comparison’ thinking by comparing each machine against its own theoretical maximum, allowing the crew to identify where it is losing capacity and the management to support any initiative and activity to improve this.

15. What should I do to be able to compare the OEE's of my machines

Q: I want to start comparing OEE’s of  our machines. How should I proceed?

Arno Koch •    To see what the effect of comparing OEE’s is, let’s compare two countries on how they do on Traffic. Country A has a mobility rate of 18% and Country B 13.5%. So hurray for country A, right?

So what needs to be done to get those losers from country B to get their mobility go Up? Let’s look into the situation; Country A is Great! A remarkable 40% of the time their cars are running! And the average top-speed of their cars is a wonderful 230 Km/h, and yes then you’ve got to go fast and that is what they do: on average the A’ans are cruising a wonderful 130Km/h. Of course at such speeds you may miss a curve or hit a tree, that’s in the game, but 80% of the trips go without any problem. So their mobility rate is 40% x 56% x 80% =   18%, Well done Country A!

Now let’s compare that to Country B. Only 15% of the time the cars are running; here they use a well-functioning public transportation system. And all cars are limited at 100 km/h. They are cruising on an average 90km/h and tend not to hit any trees; 99,99% of the trips have no problems. So their mobility rate is 15% x 90% x 99,99% = 13,5%.


So who is doing better here? You think this is a ridiculous comparison? YES! I also think so. And yet this is what you are asking for when you start to compare factories based on OEE… I can tell you real life stories from real companies that are even more astonishing! And yet people want to abuse OEE in this way.

Let me take a shortcut and tell you why you can’t even compare the OEE’s from ONE machine running the same product in two shifts.

The early shift runs 46% OEE and the noon shift also runs 46%. So who is best? Or are they the same?

  • Shift A runs 60% of the time on 80% of the max speed, with 95% quality: 46% OEE
  • Shift B runs 80% of the time (wow) on 95% speed (Hurray) producing 60% Quality (oops): 46% OEE

Now what when the machines are different, the complexity of the product, the product mix, the lot-sizes…. And still you want to put everything in ONE number to compare this to another number?? Tell me, what are you going to achieve now??

Conclusion: OEE as a number cannot be used to compare factories, machines or even shifts.

Q: Wait wait; I would build in some Weighting factors, that would solve the problem!

Arno Koch •    Aha… Ok, in Country A there are 4,8 times more cars than in country B, so what would the weighting factor now? And what would the criteria be to define such factor?

Let’s assume we have a factor (I would love to join the meetings where they are established…) and as a result country A becomes 17,2%  and B 19,3%.

Now what? What will the management do to improve the results?? And what will be the real effects?

It is this kind of ‘management by excel‘ that often leads to horrifying and catastrophic fail-decisions.

So before running into a huge pitfall, think twice:

  •  What do you really want to achieve?
  •  How can you really visualize where you are and where you are heading?
  •  How can you help the factories to move in this direction (it is the task of the management to give the direction and to govern the change in this direction, remember the speeches of Deming, if not Google for them)

My advice when a company starts comparing OEE’s: Sell shares!

This might not have been the answer you wanted but it is probably the one you need…

Good luck!

More info:

Evaluating performance of machines

Can 85% OEE be NOT world Class?

Compare OEE when defined according the OEE Industry Standard

Use and abuse of OEE

World Class OEE

How high should OEE be?

16. What is the correct definition of 'SMART'?

Q: SMART: What is the correct definition? How does it relate to OEE?

How should we use SMART when improving the OEE?

Arno Koch •  Indeed different explanations can be found, particularly about the A and R part. Let me explain SMART in relation to improvement of OEE.

What is SMART?

SMART is an acronym giving criteria for goals or targets to be met. You will usually find something like:

  • S – Specific
  • M – Measurable
  • A – Accepted/Acceptable/Assignable
  • R – Realistic/Realizable/Relevant
  • T – Time-bound

SMART: Specific

If you ask your team to improve OEE from 45 to 55%, be specific: Are you sure this is your target?

Imagine the quality rate drops by 5% yet the performance rate goes up enough to compensate this AND achieve the 55%… Now, is the target being met?

If the team decides to do less change-overs or to skip maintenance, the OEE will go up and they may meet the target. Is that what you want? Be Specific!

SMART: Measurable

Keep in mind OEE does only measure EFFECTIVENESS; it does not measure efficiency! So you may need secondary KPI’s to specify your real achievable’s!

An assumption is that when something can be measured, it also can be monitored. Unfortunately, as many OEE users know, having measured and collected the data, does not mean the information you need is presented in a way it triggers the right response… Here the quality of your OEE software will be crucial.

Can you MONITOR the achievements?

SMART: Acceptable

To meet the target, the efforts and the conditions to meet the target need to be acceptable. Meaning: You will here need to define the conditions, the boundaries, the NoGo’s.

When raising OEE, ‘acceptable’ could mean: Stock will NOT raise, No security and safety risks, no quality risks, no extra personnel, etc etc.

Acceptable also means: Conditions have to be met to get there. You can not ask to improve the OEE when the conditions to do so are not given.

A team running on 110% workload, can not be asked to invest time in a Kaizen activity. You can not ask for more safety, quality etc. when the team is forced to skip procedures by putting them under immense pressure.

SMART: Relevant

Warning: Do not attempt to discuss here whether something is ‘realistic’. In 1962 going to the moon was absolutely not ‘realistic’ to most of the people. In 1969 it was done. If that could be done more than 40 years ago …

Important to your team is to understand the RELEVANCE to achieve the goal. WHY do we need to go from 45 to 55% OEE? And why 55%? Why not 53% or 56,4%.

Explain your goal-cascade, inform the team about your policy deployment, your hoshin kanri efforts. People who know WHY they have to bring an effort will more likely do it when they know WHY!

SMART: Time-Bound

Time-bound is not just about setting a ‘deadline’. It is about having a time-line. A time-line is closely tight to a PLAN

  • What needs to be achieved when?
  • How can we follow up?
  • Is the team still on track?
  • Do they need help?
  • Can we celebrate some successes?
  • ….
Following this SMART definition might help 
to have a simple yet powerful framework to improve your OEE

See also:

17. Where can we download the registration form you used in your book?

Q: Where can I find an excel spreadsheet to make my registration form for OEE?

Arno Koch •    
Below find some excel examples for the OEE form as described in the book “OEE for the production team“.

Download it, adapt it to suite your needs and start registering!

Front side

The front side contains all the information to calculate the availability.

Do not define more activities than will fit on this sheet of paper; it will add no value. Although it should be almost self-explaining, you may want to use the book “OEE for the productionteam” if you need mare information.

Basic OEE Registration form (front)

Basic OEE form in Excel (front)

Back Side

On the back side, all data about ‘output’ is being collected. Keep it Speedy and Simple (KISS)! and make sure the team calculates its OEE at each shift ending!

Basic OEE Registration form (back)

Basic OEE form in excel (back)

Download Excel OEE Registration Form

Answer still not found?

If you checked all the sections in the Academy, yet still not found the answer to your question, ask it here.
To help you best, please be as specific as possible and please understand:
We do not want to reply to anonymous questions.