By Hilary Constable, Guest Contributor
A Powerful Tool
Overall Equipment Effectiveness, or OEE, is a powerful performance measurement derived from three elements: availability, performance, and quality. Measuring OEE is a best practice and very common in organizations leveraging common improvement programs such as six sigma or lean transformation.
Combining availability, performance, and quality provides a clear representation of how a process is operating. Yes, you read that correctly. While “equipment” is in the name, it’s really about the entire process for producing a product or providing a service. The equipment is just on component. Consider a few examples:
- A shortage, of test chemicals, means a blood analysis machine will sit idle – low availability
- Poor training, of a new employee, leads to the assembly line running at a slow pace – poor performance
- Lack of calibration, on a measurement tool, causes defects to “escape” from an inspection process – a quality disaster
While availability, performance, and quality are all important, bringing all three together in one overall measurement is half of the magic of OEE. Chasing one element, just to hit a number, will often lead to problems with the other elements. The other half of the magic is tracking OEE to show progress, identify issues which need to be investigated, finding the root cause of those issues, and eliminating those root causes.
Using OEE, as a performance measurement, requires an understanding of each element and how they work together. Let’s look closer at availability, performance, and quality.
Availability is the amount of time a piece of equipment is running divided by the amount of time it should be running. Is the equipment available when it is supposed to be?
Let’s say a hospital plans to run a centrifuge for 15 hours per day. Due to various reasons, the centrifuge actually runs 13.5 hours per day. The availability of the centrifuge is 13.5 divided by 15 or 90%.
A number of reasons could be causing availability to be less than planned:
- Shortage of people
- Unplanned maintenance
- Shortage of supplies
The calculation for availability is clear, but that doesn’t mean it’s easy. One of the main challenges with availability is agreeing on what counts as “available.” Some organizations calculate the available time based on every minute the machine was available to run. Others calculate it based on how much time the staff was also available to run it. A machine is rarely running off to a meeting or taking a break for lunch. The same cannot be said about the people to operate the machine.
Many process improvement enthusiasts presume any machine should be available 24 hours a day and 7 days a week. This grandiose approach might feel good. But, it does reduce the effectiveness of OEE as a performance measurement. The staff at a restaurant cannot run the pizza oven 24 hours a day. The restaurant is only open 12 hours a day and there is no business reason to run 24 hours a day.
The last think you want is a performance measurement which drives the wrong behavior. When it comes to availability, you want people to focus on why equipment was not running when it was supposed to be. You don’t want people finding creative ways to run machines 24 hours a day – when that is not the goal.
Performance is actual run rate divided by the ideal run rate. Is the equipment performing as it is supposed to – when it is running?
Let’s say our centrifuge should take 15 minutes per batch – the ideal run rate. During the 13.5 hours when the centrifuge is staffed and running, the total batches completed is 45, or 18 minutes per batch. The performance of the centrifuge is 15 divided by 18 or 83%.
A number of reasons could be causing performance to be less than ideal:
- Lack of training
- Delayed maintenance
- Poor instructions
As with availability, the calculation for performance is straight forward math. One of the main challenges is setting the “ideal run rate.” A few people advocate a machine should be expected to run at the maximum speed provided in the original specifications from the manufacturer. These original specifications might have been developed for an entirely different application.
The foundation of performance should reflect how the a machine behaves in the specific process under ideal conditions. Do not muddy the waters by basing an ideal run rate on limited data, such as the fastest operator, or on a process which hasn’t been standardized. Identify the ideal conditions and what the run rate should be when operators are following the standard operating procedures and work instructions.
The last thing anyone needs is a performance measurement which promotes dangerous behavior just to hit some fictitious engineering standard. You don’t want people running equipment too fast and causing a danger to themselves and others.
“You cannot mandate productivity; you must provide the tools to let people become their best.”
– Steve Jobs
Quality is the products, or services, which meet standards without rework or any other alterations. This is also known as “first pass yield.”
One possible way to see quality in a centrifuge is in how many of the samples separate as expected – the sole function of the process. Let’s say 41 of the 45 batches we ran separated correctly and could be passed to the next step in the process. The quality of the centrifuge is 41 divided by 45 or 91%.
A number of reasons could be causing quality, or first pass yield, to be less than perfect:
- Failure to follow directions
- Worn parts on equipment
- Poorly calibrated instruments
Beyond someone not counting accurately, defining quality is a challenge for many. An analysis commissioned by the National Quality Forum identified 1,367 quality measurements – used by 48 state and regional healthcare systems. Knowing exactly what is required for a first pass yield is easier said than done.
Many organizations are rely upon rework and rejection to sort out poor quality. They will often ignore these first pass failures in the official measurement. Claims of 99% or 100% quality are often a signal that it’s time to look much closer at what is being reported.
A quality measurement is only valuable if it is reporting the true situation with an emphasis on first pass yield. Finding products, or services, which don’t meet the standard on the first try serves up those golden nuggets for root cause analysis and continuous improvement.
Bring It Home
In our centrifuge example, we’re talking about a critical step in the process for a blood test. We have found:
- Availability = 13.5 divided by 15 or 90%
- Performance = 15 divided by 18 or 83%
- Quality = 41 divided by 45 or 91%
I suspect you can see where this is going!
The overall equipment effectiveness is 90% times 83% times 93% which equals 68%.
This centrifuge can be relied on to perform as expected 68% of the time. The rest of the time, there will be an issue with availability, performance, or quality. Beyond the potential for a misdiagnosis, or other direct impact on patient health, the hospital and our entire healthcare system is slowed unnecessary waste and aggrevation.
OEE paints a vastly different picture from other performance measurements which don’t include all three elements.
Healthcare workers have enough to worry about without worrying about whether the equipment they rely on will work as expected. When a machine fails, someone’s work gets more complicated. They have to stop to repair it, wait for someone else to repair it, or find another one they can use instead.
Imagine a cart of blood samples rolls up to the centrifuge and they all need to be processed within an hour of being drawn from the patients. If the Centrifugal Operator suddenly can’t use the centrifuge, what happens to their stress level?
If there’s another centrifuge available nearby, that could solve the problem temporarily. If not, the operator will have to track one down. It becomes a race against the clock.
What if this is a common occurrence at this facility? How would the operator feel going into work every day knowing this struggle is likely to come up?
Imagine the difference for the Centrifugal Operator who goes into work knowing all the tools they need will be available and able to keep up with their workload. Which operator is more engaged? Which one is more likely to continue working where they are?
One of the first steps toward an engaged workforce is meeting the bare minimum of ensuring the tools they need are available and able to perform as expected. OEE gives a way to understand more fully what it is like to work with those tools.
An availability rate of 89% doesn’t sound so bad until the performance rate and quality rate are added. Then it becomes clear the operator is experiencing some sort of issue almost two out of every three times they use the equipment. That sounds awful.
For Your Customers
Employees aren’t the only ones who may feel frustrated or stressed by faulty equipment. Apply the principles of OEE to an MRI scanner or a defibrillator. Patients need to be able to rely on accurate test results and treatment. They want to receive those results and treatment when they expect to, no matter how many other people are also waiting.
In those cases where time is short and critical, no one wants to hear they can’t be helped because there’s an issue with the equipment.
Think back to the last time you were on an airplane and there was any sort of delay, even a short one. How did that feel? How long did it take for you to calculate the length of your layover and whether there would still be enough time to make your connection? It’s frustrating and promotes doubt into the airline’s ability to transport their passengers safely and on time.
As you work through these calculations, absorb the concepts behind them, and add them to your knowledge of Lean, remember they are only tools. What really matters is knowing when to use which ones and how to ensure your accuracy in order to produce a reliable vehicle, deliver a beautiful birthday cake that will make someone’s day, arrive on time, or save a life.