Skip to main content
Skip table of contents

OEE

OEE (Overall Equipment Efficiency) is a performance indicator used in production reporting. It measures the level of efficiency and utilization of production equipment and provides information on how well these equipment are utilized and how effectively the production process is running.

OEE is typically calculated based on three key factors: availability, performance, and quality. Each of these factors represents a different aspect of the production process:

Availability: It measures how often the production equipment is available and ready for operation. It takes into account planned and unplanned downtime, such as maintenance, breakdowns, or changeover time.

Performance: It reflects how efficiently the equipment operates at its designed speed. It includes factors such as speed losses, setup time, and performance-related losses during operation.

Quality: It evaluates how well the produced products meet quality standards. It measures factors such as the quantity of defective products, deviations from specifications, etc.

The overall OEE is usually calculated as the product of these three factors, providing an overview of the overall equipment utilization efficiency. OEE is a valuable tool for identifying areas of inefficiency and for improving production processes. By analyzing OEE, one can identify issues, create improvement plans, and monitor progress in achieving production goals.

What you see on the screen

Area

What it shows

How to read it quickly

Gauge widgets (top row)

Four speedometer dials for OEE, Availability, Performance, Quality for the selected period.

Red → needs attention, Amber → below target, Green → meets/exceeds target. Keep an eye on which pillar drags OEE down.

Column chart (middle)

Four bars per time-bucket (grey = OEE, yellow = Availability, blue = Performance, green = Quality). Dashed yellow line = period-average OEE.

Spot spikes or dips in individual pillars. A dip only in blue bars points to speed losses; a dip in yellow bars means stops.

Line-stoppage table (bottom)

Every unplanned stop that occurred inside the selected window, with reason code and start/finish times.

Sort by Reason Code to identify the biggest Availability killers.

image-20250722-074234.png

Why this report matters

Overall Equipment Effectiveness condenses three fundamental questions into one percentage:

Question

OEE Pillar

What it tells you

Was the line available when you needed it?

Availability

Impact of planned vs. unplanned stops

Did the line run at the planned speed?

Performance

Losses from reduced pace, micro-stops, idling

Did the line produce good parts?

Quality

Yield, scrap and rework

Multiplying the three pillars gives OEE. A single number pinpoints how close the line is to its theoretical maximum, while the pillars highlight which loss to attack first.

Where the numbers come from

Step

Source & Logic

End result visible to you

  1. Collect operating time

The system checks each shift calendar and machine signal. It marks every minute as running or stopped.

Availability time and Stoppage time per bucket.

  1. Capture pieces & scrap

Production declarations (good, scrap, discard) are time-stamped. If only total counts are declared, P4 apportions them proportionally across the interval.

Good quantity, Scrap quantity.

  1. Pull cycle-time target

For each material P4 stores the best-known Minimum Production Time (ideal seconds per part).

Benchmark speed for Performance.

  1. Crunch the pillars

Availability = 1 – (stoppage ÷ available time)

Performance = (ideal cycle × good qty) ÷ runtime

Quality = good qty ÷ (good + scrap)

  1. Calculate OEE

OEE = Availability × Performance × Quality

Refresh cadence
All calculations roll up to the last midnight and are recalculated every hour. Granularity (hour / day / 7 d / 30 d) follows the OEE Production Line Interval setting in Admin.

Interpreting & acting on the results

Situation you see

Likely root cause

Typical next step

Low Availability, good Perf & Qual.

Unplanned downtimes, change-overs, missing operators.

Drill into stoppage list, launch RCA, schedule maintenance or SMED activity.

Low Performance, others ok

Micro–stops, speed restrictions, poor scheduling of mix.

Check shift logs, verify cycle-time standards, balance product mix.

Low Quality only

Process instability, start-up rejects.

Review scrap codes, tighten quality checks, adjust parameters.

All pillars healthy but OEE still < target

Very aggressive target, or multiple small losses.

Re-validate theoretical capacity, engage CI team for step-change.

Benchmarks (rule-of-thumb, discrete manufacturing):

  • < 60 % – fire-fighting zone, focus on Availability.

  • 60–75 % – improving, address speed losses.

  • 75–85 % – competitive, chase yield.

  • 85 % – world-class, sustain & optimise.

FAQ

Q: Can I see multiple lines side-by-side?
A: Group OEE or multiple line OEE reports is in our pipeline for Q1/2026, now there is only this detail report which focuses on one line for clarity.

Q: How often should I look at hourly OEE?
A: Hourly view is best for short interval control (GO/NO-GO meetings). Daily view is ideal for shift hand-over; weekly for CI workshops.

Q: How is scrap counted in this report?
A: Admins can define if the Quality KPI should calculate only with Scrap, Bad Quantity or both combined so its really up to your preference. Performance measures speed, so it uses all produced parts (good + scrap + discard) against the ideal cycle time. Scrap affects Quality, not Performance.

JavaScript errors detected

Please note, these errors can depend on your browser setup.

If this problem persists, please contact our support.