OEE Calculation Explained in 500 Words

Overall equipment effectiveness (OEE) is a group of measurements designed to highlight opportunities to make assets more productive through process improvement. OEE calculations are an important metric and analytic tool in lean manufacturing. As with most lean-related concepts, even though the math behind OEE is basic and straight-forward, thinking is required. Always keep in mind that the goal is to drive behavior that results in improved machine productivity.

The Math

In simple terms, the top-level equation for an OEE calculation is % Available x % Productive x % Good Quality. Let’s plug in hypothetical numbers for each: “Available” means “how much of the time a machine was expected to run was it running?” If it is expected to run for ten hours a day, but mechanical breakdowns only allowed it to run nine hours, then the per cent available would be 90% (9/10). The key loss in availability is machine down time. “Productive” means while the machine was running, how close to its maximum speed was it running? If the maximum speed is 100 parts per hour, and the machine ran at 80 parts per hour, then productivity would be 80% (80/100). “Good Quality” means the percent of parts that meet standard and can be used to fill customer orders. If 5 parts out of 100 have some sort of problem preventing them from being shipped, then good quality would equal 95% ((100-5)/100). The OEE calculation in this example would be 90% Availability x 80% productivity x 95% Good quality = or 68.4%.

Is 68.4% good or bad? While lean practitioners often refer to 85% as being world class, and 60% being pretty typical, the really key point is that with the right amount of problem solving, you have the opportunity to increase your output by 31.6% (100% - 68.4%) at no additional cost beyond direct materials. So the point of OEE is to get you to look at machine down time, at productivity losses, and at quality losses, and try to discover ways to reduce them, subsequently increasing OEE.

Think Beyond the Equation

So why is thinking required? Because at each level there are circumstances that make the measurement less clear cut. In the above example, the “expected to run” time was ten hours. Suppose management’s reaction to the one hour downtime was to extend the machine run schedule to eleven hours. Now does the equation become (10/11 = 90.9%)? It shouldn’t, because OEE would rise without any actual process improvement. In the productivity example, the calculation is easy because the maximum speed is a constant. But what is the arbiter of what the maximum speed should be? Can the maximum speed vary by product type? If so, what does that do to the equation?

The point of the questions above is not to introduce confusion, but to introduce caution. Constant tinkering with the measurement definitions will change the OEE just as certainly as implementing process improvement. Spend the time to set up the right measurement definitions from the start.