OEE is the standard metric manufacturers use to measure how well a machine or production line is actually performing against its potential. OEE tells you where your planned time and output are actually leaking away. Most plants find their real number is lower than they expected.
OEE in manufacturing is a percentage that measures how much of your planned production time is fully productive. A machine running at 100% OEE is producing only good parts, at its ideal cycle rate, with no unplanned stoppages during planned production time.
In practice, no machine hits 100%, but tracking OEE gives you a number you can act on.
The metric was developed as part of Total Productive Maintenance (TPM) and is now standard across discrete manufacturing, process manufacturing, and assembly operations. It works for a CNC cell, a press brake line, an injection molding machine, or any equipment where you can measure runtime and output.
OEE excludes scheduled breaks and other planned non-production periods. Whether planned changeovers are included varies by organization, but in TPM-based OEE systems setup and changeover time is typically treated as an Availability loss. The formula only measures what happens during planned production time.
The OEE formula multiplies three factors together: OEE = Availability × Performance × Quality.
Each factor is calculated as a decimal (or percentage) and the three are multiplied:
Because you multiply three fractions together, OEE drops quickly when each factor dips even a few percent. An Availability of 90%, Performance of 90%, and Quality of 90% gives you an OEE of 72.9%, not 90%.
To calculate OEE, gather four numbers from one shift: Planned Production Time, actual Run Time, Total Count produced, and Good Count. Everything else is derived from those.
Here is a worked example using a CNC machining cell running an 8-hour shift:
Step 1: Availability — Run Time / Planned Production Time = 390 / 450 = 86.7%
Step 2: Performance — (Total Count × Ideal Cycle Time) / Run Time = (330 × 1) / 390 = 84.6%
Step 3: Quality — Good Count / Total Count = 315 / 330 = 95.5%
Step 4: OEE — 0.867 × 0.846 × 0.955 = 70.0%
That 70% is a reasonable result for a plant that has started tracking. The breakdown tells you the biggest lever is Performance (84.6%), not Quality. The first investigation should be slow cycles and minor stops, not the inspection process.
Note on the 60-minute downtime: in this example, 20 minutes of that unplanned stop was traced to an empty consumable bin with no grinding discs available for the next operation. That material shortage cut Availability from a potential 91.1% down to 86.7%.
A world-class OEE score is 85%, a benchmark widely cited in lean manufacturing and TPM literature. Most manufacturers, when they first start tracking honestly, land between 40% and 60%. Newly instrumented machines often measure around 40%, not because the plant is poorly run, but because losses that were invisible are now being counted.
An OEE of 85% means 85% of planned production time is producing good parts at full speed. For context: a plant running at 60% OEE has 40% of planned time either stopped, running slow, or producing scrap. That is a significant amount of recoverable capacity without adding a single shift or machine.
The 85% world-class figure is a reference point, not a target every plant should chase immediately. A plant at 55% improving to 70% has created more value than a plant at 80% grinding toward 85%.
The Six Big Losses are the six specific ways a machine can fail to produce good parts at full speed, and every one maps directly to an OEE component. The framework was formalized in TPM and gives maintenance and operations teams a common language for diagnosing where OEE is being lost.
Planned stops (scheduled maintenance, planned breaks, planned changeovers logged in advance) are not one of the Six Big Losses. They are excluded from Planned Production Time before OEE is calculated.
In most plants, unplanned downtime and minor stops account for the largest losses. Minor stops are especially underreported because operators clear them in seconds and never log them, yet they compound into significant lost time over a shift.
For a press brake line or injection molding machine, startup rejects are worth watching separately. The first 10–20 shots after a mold change are often scrap, and if run counts are small, that can move Quality by several percent.
Understanding which material and supply failures create the most Availability losses is part of the diagnostic work. The top causes of stockouts in manufacturing map directly to the unplanned downtime loss category.
Improving OEE starts with identifying which of the three factors is lowest, then targeting the specific loss driving it. Generic advice about "reducing downtime" is not a plan. Finding that your CNC cell loses 45 minutes per shift to minor stops on the chip conveyor: that is a plan.
For Availability losses:
For Performance losses:
For Quality losses:
A practical improvement cycle: measure OEE honestly for four weeks, identify the largest loss category, run a focused improvement event on one machine, measure again. Small manufacturers do not need elaborate software to do this. A whiteboard and a consistent logging habit get results.
For Availability losses linked to consumable shortages, the fix is a correctly sized replenishment loop: reorder points set to consumption rate times lead time, with a small safety cushion for supplier variation. Sized right, the consumable is at the machine when the operator reaches for it, not delayed in a replenishment cycle.
Missing materials cause unplanned downtime, and unplanned downtime directly reduces Availability. A welding cell that stops because welding wire ran out, a CNC cell that idles because no grinding discs are available, a press brake line waiting on the correct material specification: these are OEE losses that have nothing to do with the machine itself.
In the worked example above, 20 of the 60 downtime minutes came from an empty consumable bin. That single material shortage dropped Availability from a potential 91.1% to 86.7%, and drove the final OEE from a potential 73.6% down to 70.0%.
Kanban replenishment systems address this directly. By setting a reorder trigger based on consumption rate and lead time, a well-run two-bin kanban system keeps consumables available at the point of use without relying on operators to raise purchase orders. When the first bin empties, the shop switches to the second bin, and the empty bin becomes the signal to refill the first, so material is already on its way before the second bin runs down.
The connection between inventory management and OEE is direct and measurable. Stockouts are one of the most common sources of unplanned downtime that plants fail to categorize correctly. They appear in maintenance logs as "waiting" rather than in procurement data as a supply failure.
Most of the Availability you lose to material shortages comes from the same place: nobody is counting the grinding discs or the welding wire, because counting consumables is a job no one wants and no one keeps up. So the bin runs dry, the cell idles, and it shows up in your log as "waiting."
Arda is the modern version of the two-bin system that closes that gap. Physical cards sit in the bins on your floor. When a bin hits its replenishment point, someone scans the card and the reorder is triggered — no counts, no spreadsheets, no operator chasing down a purchase order. You never count your consumables again; the material just shows up before the cell runs out. That is one whole category of unplanned downtime taken off your Availability number.
If material shortages are quietly eating your OEE, see how Arda keeps point-of-use consumables flowing without adding a system your team has to babysit.
No, you do not need software to track OEE reliably. A paper log or a simple spreadsheet is sufficient. The data collection discipline matters more than the tool. The minimum you need is a shift log that captures: shift start time, planned stops (with times), unplanned stop events (start, end, and reason), total parts counted, and good parts counted.
From those five data points, you calculate Availability, Performance, and Quality for every shift. Plot the daily OEE number on a whiteboard chart visible to the production team. Visibility alone drives improvement. Operators who can see OEE trending down before end of shift will act on it.
If you want to move beyond manual logging, basic OPC-UA data from modern CNCs, injection molding machines, and press brakes feeds into simple spreadsheet dashboards or low-cost manufacturing execution tools. The goal is not a sophisticated analytics platform. It is a consistent number that your team trusts and updates every shift.
The kanban reorder point for consumables is one data point worth feeding into the same tracking system. When a material-related downtime event occurs, you can immediately check whether the reorder point was set correctly or whether replenishment failed upstream.
What does OEE stand for? OEE stands for Overall Equipment Effectiveness. It is a percentage that measures how much of planned production time is fully productive, accounting for availability, performance, and quality losses.
What is the OEE formula? OEE = Availability × Performance × Quality. Availability is Run Time divided by Planned Production Time. Performance is Total Count times Ideal Cycle Time divided by Run Time. Quality is Good Count divided by Total Count.
What is included in Planned Production Time? Planned Production Time is the shift or scheduled run time minus any planned stops: scheduled breaks, planned maintenance, and planned changeovers logged in advance. Unplanned stops remain inside Planned Production Time and are captured as downtime losses.
What is the difference between OEE and TEEP? TEEP (Total Effective Equipment Performance) adds utilization to OEE. It measures productive time against all available calendar time, including planned stops and unscheduled non-production time. OEE measures performance during planned production only. TEEP is a broader capacity metric; OEE is an operational efficiency metric.
Can OEE be applied to a whole production line? Yes, OEE can be calculated for a single machine, a production cell, a line, or a plant. For a line, the slowest or most constrained machine typically drives the overall number. Tracking OEE at machine level is more useful for improvement work because it pinpoints which equipment is causing losses.
Why is my OEE lower than I expected? Most OEE surprises come from minor stops that were never logged and from Performance losses that operators normalize over time. A machine that always runs slightly slow does not trigger alarms, but the Performance factor captures it. Start with four weeks of honest data collection before drawing conclusions.
What is the relationship between OEE and lean manufacturing? OEE was developed as part of Total Productive Maintenance (TPM), which is one pillar of lean manufacturing. In lean terms, OEE losses are manifestations of waste. Downtime creates waiting, slow cycles reduce flow efficiency, and defects create direct waste through scrap and rework. OEE gives lean teams a quantified starting point for improvement.
Does planned maintenance count against OEE? No. Planned maintenance is a planned stop and is excluded from Planned Production Time before OEE is calculated. Only unplanned stoppages (unexpected breakdowns, waiting on materials, unplanned adjustments) reduce the Availability factor.
How often should OEE be calculated? Calculate OEE every shift. Daily or weekly roll-ups are useful for trend monitoring, but shift-level data is what allows operators and supervisors to connect specific events to OEE results. The closer the measurement is to the event, the more actionable the data.
.svg)
.svg)
.svg)
%201.svg){kind=logo}
.png)
.svg)
.svg)
