How to Improve OEE on Your Production Line

Improving OEE means choosing the right lever

The most common mistake is treating OEE as a demand for speed. That reduces quality and breaks the equipment. The correct path is to identify the dominant loss and eliminate its cause.

Raising OEE does not mean asking everyone to work faster. It means finding out which of the three factors loses the most and acting on it. The formula and the calculator are in OEE: How to Calculate Overall Equipment Effectiveness. Here the focus is what to do after calculating.

A quick reminder: OEE is the product of availability, performance and quality. That is all you need for this article. The numbers in the example below are illustrative, and the math has been checked.

Starting example (illustrative)

A line runs a 480-minute shift, with 30 minutes of planned stops (meal and meeting). Planned time is 450 minutes. There were 45 minutes of unplanned stops. The ideal cycle is 0.5 minute per part. 729 parts were produced, and 36 were scrapped.

Check: the good parts (693) multiplied by the ideal cycle (0.5) give 346.5 minutes of valuable time. Divided by the 450 planned minutes, that results in 0.77, or 77.0%.

See where the loss is. Of the 450 planned minutes, 45 were stoppages and 40.5 minutes were lost to speed (405 − 364.5). Another 18 minutes were spent making the 36 parts that became scrap (36 × 0.5). Adding up: 45 + 40.5 + 18 = 103.5 minutes of loss, and 450 − 103.5 = 346.5. The math closes.

Lever 1: availability

This is where breakdowns and setups sit. The most effective actions:

• Record every stop with its reason and duration, and attack the largest item on the Pareto chart.
• Reduce setup with the SMED method, separating what can be done while the machine is running.
• Implement autonomous and preventive maintenance, as in autonomous maintenance.
• Track repair time with MTBF and MTTR.
• Make sure material, tooling and operator are at the workstation before the shift starts, avoiding stops caused by shortages.

In the example, reducing stoppages from 45 to 27 minutes raises availability to 423 ÷ 450 = 94.0%.

Lever 2: performance

This is where minor stops and reduced speed sit. They are invisible losses, because the machine "is running."

• Count micro stops (jammed feeder, dirty sensor) on a simple tally sheet.
• Confirm the ideal cycle with the manufacturer and with real trials.
• Investigate why the operator slows down: variable material, unstable adjustment, fear of scrap.
• Standardize process parameters in standardized work.
• Level the line supply so that parts neither run short nor pile up, which also stabilizes the pace.

In the example, raising performance from 90% to 95% lifts OEE to 0.900 × 0.950 × 0.951 = 81.3%.

Lever 3: quality

Scrap and rework consume machine time that has already been spent. The cause usually lies in process variation or human error.

• Use SPC to detect drift before it generates defects.
• Apply poka-yoke to assembly and adjustment errors.
• Address start-up losses with a standard initial adjustment.
• Measure first pass yield.
• Separate scrap by defect type and originating workstation, so the action is specific.

In the example, bringing quality to 99% (from 95.1%) gives 0.900 × 0.900 × 0.990 = 80.2%.

Which lever first?

Before deciding, validate the quality of the data. Short stops with no record, a poorly defined ideal cycle and scrap counted only at the end of the line distort OEE. A wrong indicator leads to the wrong action.

All three paths give a similar gain in this example, between 3.2 and 4.3 points. In practice, the decision depends on the cost and ease of each action. Choose the factor with the lowest number and the clearest cause.

Note also that improvements add up. If all three actions in the example are done together, OEE goes to 0.940 × 0.950 × 0.990 = 88.4%. Check: 0.940 × 0.950 = 0.893 and 0.893 × 0.990 = 0.88407. It is a long-term target, not a promise. For each machine, the realistic ceiling depends on the process and the product.

Note: a high OEE on equipment that is not the bottleneck does not increase plant output. Prioritize the bottleneck (see Theory of Constraints).

Next step

To sustain the gain, turn the actions into routine. Display the shift's OEE on a visible board, with the three factors shown separately. Hold a ten-minute meeting at the start of the day to review yesterday's dominant loss. Assign an owner to each action and check the result the following week. Without this ritual, the indicator becomes a number in a report.

Calculate your line's OEE for two weeks, keeping the three factors separate. Take the lowest and list the three biggest causes of loss. Define one action for each, with an owner and a deadline, and track the result on a simple board. For a broader view of the program, read TPM best practices.

Frequently asked questions

How can I improve OEE quickly?

Identify the factor with the lowest value and attack its biggest cause of loss. Recording stops by reason usually reveals quick gains.

Which factor is usually the hardest to see?

Performance, since minor stops and reduced speed look like part of the routine.

Is 100% OEE possible?

In practice, it is not a realistic target. The goal is to reduce losses continuously, with targets that make sense for each process.

Does a high OEE mean an efficient plant?

Not necessarily. If the equipment is not the bottleneck, the gain does not increase plant output.

Where is the complete formula?

In the article on how to calculate OEE, with the calculator. Here we use only the calculation from the example.

About the author

Vagner Soares

Lean Manufacturing & Behavioral Management Specialist

Over 20 years in the automotive and metalworking industries (GM and Dana), Lean Manufacturing practitioner since 2006. SENAI instructor and mentor in Brazil’s Brasil Mais Produtivo program, delivering consulting, training and audits for 50+ companies, combining quality, productivity and people development.

LinkedInInstagramFull profile →