What Is OEE in Manufacturing? Definition and Formula

Overall equipment effectiveness (OEE) is a way of calculating operating efficiency for a manufacturing line.

OEE can help you identify productivity losses in your production line. For example, a line can run every scheduled minute of a shift and still turn out far less than it’s built to produce, either from a slow machine or low-quality output that ends up in the reject bin. OEE quantifies these shortfalls so you can evaluate your operation and compare lines and shifts.

This guide defines OEE, breaks down its three components, and shows how to calculate it. It also covers how to evaluate your score compared to common benchmarks, the Six Big Losses that can lower OEE, and how to improve your OEE score.

What is OEE?

OEE, or overall equipment effectiveness, is a measure, or “score,” in manufacturing that combines data on availability, performance, and quality to show how much of your planned production time is truly productive.

Planned production time is the time you schedule a line or machine to run. The most productive time is when a run is making quality parts or products at full speed with no stops. OEE measures the share of planned production time that meets all three of the conditions for truly productive time: availability, performance, and quality.

A production line’s OEE can flag problems in a production line and compare one shift, week, or line to another. This can help you identify issues early, so you can correct them quickly.

3 components of OEE: availability, performance, and quality

OEE combines three elements of productivity: availability, performance, and quality. Each rate compares what happened and what should have happened during a production run:

• Availability. This component measures actual production time available against planned production time. Availability falls when the line is scheduled to run but does not, caused by equipment breakdowns and setup or changeover time.

• Performance. This component measures actual cycle speed against a machine’s ideal cycle speed—the fastest time to make one quality unit, without causing breakdown or waste. It falls when the line runs slower than it should, including short minor stops and reduced speed.

• Quality. This component measures quality units produced against total units produced. Quality falls when the line turns out units you can’t sell, such as scrap and products that need to be reworked.

Every OEE loss falls into one of these three components: a stalled line is an availability problem, a slow line is a performance problem, and a reject pile is a quality problem.

How to calculate OEE

You calculate OEE by multiplying your rates for availability, performance, and quality. 

Each rate is a ratio you build from data you can collect over a single shift or production run. You need five inputs:

• Planned production time. The scheduled number of minutes you plan to run the line.

• Run time. The actual number of minutes the line ran.

• Ideal cycle time. The fastest time that machine can make one quality unit without causing breakdown or waste.

• Total count. Count of every unit produced, whether of good or bad quality.

• Good count. Count of quality units from the run without requiring any rework.

With those numbers, here’s the formulas to calculate each rate:

• Availability = Run time / planned production time

• Performance = (Ideal cycle time x total count) / Run time

• Quality = Good count / Total count

Multiply the three rates together to calculate your OEE:

OEE = Availability x Performance x Quality

For example, if you schedule a line for 500 minutes and it runs for 400, your availability rate is 400 / 500, or 0.8.

The ideal cycle time is one minute per unit, and the line made 360 units, so your performance rate is (1 x 360) / 400, or 0.90. 

Of those 360 units, 350 passed the first time, so your quality rate is 350 / 360, or about 0.97. Multiply the three rates, 0.8 x 0.9 x 0.97, and you get about 0.7, or 70% OEE.

There’s also a shorter formula for OEE:

OEE = Good count x Ideal cycle time / Planned production time

Using the example above, that’s 350 x 1 / 500, which is also 0.7 or 70%. You can use the shorter method when you already have the numbers for good count, ideal cycle time, and planned production time on hand but haven’t worked out the rates for availability, performance, and quality.

The Six Big Losses that lower OEE

Most OEE losses fall into six recurring categories, known as the Six Big Losses: 

• Equipment breakdowns 

• Setup and changeover 

• Idling and minor stops 

• Reduced speed 

• Process defects

• Reduced yield

They come from the total productive maintenance tradition developed by Seiichi Nakajima and formalized by the Japan Institute of Plant Maintenance, where OEE is the primary performance metric.

Each loss maps to one of the three OEE rates, which is what makes the list useful. Once you know which rate is low, you know which losses to look for. Here are the six losses, with an example of how each shows up on the line:

How to improve your OEE score

To improve your OEE, start by breaking your score down by availability, performance, and quality. Identify the area losing the most productive time, and search for the reasons behind those losses first. Here are common losses and suggested fixes:

• Lower changeover time with SMED. If changeovers are hurting your availability rate, use single-minute exchange of die (SMED) to reduce downtime. It splits setup tasks into two categories: those that require the machine to stop and those that can be done while it’s running. Then shift as many stop-required tasks as possible to running time, so the changeover can happen while the line keeps producing. 

• Prevent breakdowns with planned maintenance. For breakdown losses that hurt availability, schedule preventive or autonomous maintenance. Peer-reviewed work on total productive maintenance links this kind of proactive maintenance to better reliability, less downtime, and higher OEE.

• Determine the causes of your minor stops and speed losses. Small losses are easy to overlook one at a time, but they add up across a shift. For performance losses, trace recurring micro-stops and slow cycles back to their causes. 

Whichever fix you apply, re-measure your OEE following implementation. Improvement works as a loop: Measure, find the biggest loss, fix it, then measure again to confirm the gain and reveal the next issue to improve.