You may already be familiar with some of the metrics used to measure the performance of your assets. If you have had some exposure to the common thinking on this topic, then you are likely familiar with the terms OEE (Overall Equipment Effectiveness) and MTBF (Mean Time Between Failure). Both of these metrics can be very informative and drive specific behaviors when applied correctly. When applied incorrectly they could lead to confusion and frustration. In this post, I hope to shed some light on which of these measures may be more meaningful for your organization.
Let’s start with Overall Equipment Effectiveness.
Overall Equipment Effectiveness (OEE)
The layman’s version of OEE is simply defined as what percentage of the time that you were capable of running good product at full capacity and with acceptable quality did you do so? Perfection at OEE is 100%. Anything more than that and you are breaking a few of the basic laws of physics, but we will save that discussion for another time.
The common formula for OEE is as follows:
OEE = Overall Equipment Effectiveness
A = Availability = The percentage of the time that you actually ran and produced something.
P = Performance Efficiency = During this time that you actually produced, at what percentage of your rated capacity (speed, throughput, etc.) did you actually run.
Q = Quality Rate = For all of the units that you produced, what percentage of them were acceptable for sale to the customer, meeting the stated quality requirements.
Availability = 80%
Performance Efficiency = 90%
Quality Rate = 99%
OEE = 80% X 90% X 99%
OEE = 71.28%
OEE is a fantastic measure for a few important reasons:
- Nowhere in the definition do the words “maintenance” or “operations” appear. It brings us together under one common measure. It facilitates two teams who can easily work at cross-purposes to work together. For example, availability losses can be experienced as a result of both maintenance and operating practices.
- It forces us to focus on the loss of potential rather than what we have accomplished with no idea of what might be possible.
- It brings together three aspects that are critical to meeting our customer’s needs. For example, if we only measured availability, we might overlook the fact that we ran at 50% speed.
OEE is a bit of a universal metric. It includes many of the critical aspects of your performance into a single metric. If you include some safety and cost measures you are approaching a well formed maintenance scorecard.
Mean Time Between Failure (MTBF)
Mean Time Between Failure tells us how often, on average (mean), we should expect to see an asset fail to fulfill its function. Higher MTBF performance is always considered to be better, as it indicates that your asset performs it intended function longer without any interruption attributed to failures. Downtime that may exist associated with planned maintenance activities are excluded from this calculation.
A simple definition for MTBF is:
MTBF = Operating Hours ÷ Number of Failures
Operating Hours = The time frame (frequency of measurement).
Number of Failures = The number of occurrences where the asset failed to fulfill its function.
Note: MTBF is the reciprocal of the Failure Rate (1/MTBF = Failure Rate).
Assume we measure MTBF on a monthly basis (30 days or 720 hours), and we have experienced 10 failures in this time frame.
MTBF = 720 hours ÷ 10 failures = 72 hours
MTBF = 30 days ÷ 10 failures = 3 days
This means that under current conditions, we should expect to experience a failure (loss of function) every 72 hour.
MTBF is a relative measure. Measuring MTBF on a highly critical asset or group of assets can be very meaningful. Measuring MTBF across an entire department or facility provides less value.
It is also important to note that MTBF does not consider situations where assets are running at a reduced rate, nor does it consider the size and duration of the event (not all failures are equal – some are longer and have a greater impact than others).
The key to leveraging MTBF is to understand where these failures came from, and to develop focused strategies that will eliminate them. Your Root Cause Failure Analysis (RCFA) skills becomes very important when you are measuring MTBF.
Which is right for you? OEE or MTBF?
My short answer to this question is that if you can measure OEE in a meaningful way, then this is the metric for you. It covers many of the bases. Sadly, OEE does not fit in all situations.
Discrete manufacturing is the production of distinct items. Automobiles, furniture, toys, smartphones, and airplanes are examples of discrete manufacturing products. You can easily identify a single item produced by discreet manufacturing.
If this definition fits, then OEE is for you. You should be concerned about factors such as availability, performance efficiency, and quality rate. These factors are central to your success.
Process industries are those who run a continuous batch where distinction between one unit and another is not easily identified. Power generation, oil production, transport, and refining, as well as many petrochemical processes fall into this category.
Unfortunately, in these types of environments, OEE might not serve our purposes for the following reasons:
- Availability is often so critical in process industries that we see a high level of redundant equipment (with a corresponding capital investment). If we experience a failure of one piece of equipment, we switch over to the redundant unit and continue on. For this reason, availability losses are usually greatly minimized.
- Performance efficiency losses can exist, but rather than run slow, we often immediately bring the backup unit online and begin a fix to the underperforming unit.
- It is often difficult, if not impossible, to identify quality losses in process industries. What does a bad kilowatt look like?
In these situations, establishing MTBF as a meaningful measure on specific highly critical assets and trending your performance upward will likely provide much more value than OEE.
Metrics are a funny thing. When used correctly and with the right application, they can provide a lot of value and focus to an organization. Used incorrectly, they tend to cause more harm than good.
When selecting the best metrics for your team, consider the following advice:
- Keep it simple. Too many measures cloud the focus.
- Make sure we know what actions to take in order to improve the performance of each metric. What must we do differently in order to improve? It sounds obvious, but surprisingly it is often overlooked.
- Talk about it often. A metric is only as good as the leader who draws the team together to own it and act on it.