What is OEE and how is it calculated?

OEE, Overall Equipment Effectiveness, is one of the most widely used key performance indicators in the manufacturing industry. A single figure between 0 and 100% is meant to summarize how well a machine, a line, or an entire factory is utilizing its potential. However, many use this metric without really knowing what it measures, how it is calculated, or what a good value is in practice.

This guide goes through OEE from the ground up. We explain what the metric consists of, how to calculate it step-by-step, what is hidden behind the numbers, and why the "85% world class" standard is a myth that leads production managers astray. Once you have read this article, we recommend reading the second part on why OEE is not enough as a standalone KPI, where we explain how the metric should be used to actually drive improvement.

What is OEE?

OEE stands for Overall Equipment Effectiveness. In Swedish, it is roughly translated to "övergripande utrustningseffektivitet". The metric originates from Total Productive Maintenance (TPM) during the 1980s and has since become the standard for measuring how effectively manufacturing equipment is utilized.

The metric answers a simple question. Out of the time the equipment could potentially produce, how much does it actually produce in practice at the right speed and with the right quality?

OEE consists of three components that we multiply together:

OEE = Availability × Performance × Quality

Availability measures the portion of planned production time that the machine was actually running. Stops, both planned and unplanned, reduce availability.

Performance measures how fast the machine ran. Speed losses, micro-stops, and idling drag down performance.

Quality measures the share of produced units that met the quality requirements. Scrap and rework lower the quality.

Since we multiply the three parts, rather than adding them, a small deficiency in each part can result in a large total loss. Three sub-values of 90% yield an OEE of 73%, not 90%. This is one of the most important insights of the entire OEE model. The losses compound each other.

Is OEE called something else in Swedish?

In Swedish, OEE is called TAK, and it is exactly the same metric. TAK stands for Tillgänglighet (Availability), Anläggningsutbyte (Performance), and Kvalitetsutbyte (Quality). In English, the terms are Availability, Performance, Quality.

• Availability is called Tillgänglighet.

• Performance is called Anläggningsutbyte.

• Quality is called Kvalitetsutbyte.

This means you calculate TAK in the exact same way as OEE. TAK can be compared directly to OEE.

How do you calculate OEE?

The basic formula looks like this:

Availability = operating time ÷ planned production time
Performance = (actual production × ideal cycle time) ÷ operating time
Quality = good units ÷ total units produced

A calculation example. A line is scheduled to run for 8 hours (480 minutes). Due to a stop, it ran for 408 minutes. This gives an availability of 85%. During the operating time, it produced 800 units. The ideal cycle time is 0.5 minutes per unit, which corresponds to 816 possible units. Performance is 98%. Out of the 800 units, 780 were approved. Quality is 97.5%.

85% availability × 98% performance × 97.5% quality gives an OEE of 81%.

That is the calculation. The difficulty does not lie in the math. The hard part is measuring the right things, defining "planned production time" consistently, and handling changeovers, micro-stops, planned maintenance, and speed losses in a way that provides comparable figures over time. That is where most OEE projects are won or lost, not in the formula.

What lies behind the number: 6 big losses

An OEE figure is a summary. For it to be useful, we need to know what is dragging it down. TPM (Total Productive Maintenance) identifies six big losses that correspond to the three components of OEE.

Availability losses:

1. Breakdowns and unplanned stops

2. Setup and adjustments

Performance losses:
3. Idling and minor stops
4. Reduced speed

Quality losses:
5. Startup losses
6. Defects and rework during normal production

Micro-stops are included in the category of idling and minor stops within the six big losses and normally affect the performance part of OEE. They can be tracked separately in the analysis to highlight recurring problems, but they do not constitute a seventh category in the model. Micro-stops are short interruptions lasting seconds or a single minute. A bottle that gets stuck, a sensor that triggers an alarm, a sheet of paper that goes askew. Each is insignificant. Collectively, they can account for a large portion of performance losses. They are difficult to capture without automated data collection, as they are too short and too frequent to be recorded manually.

Understanding these losses is the difference between an OEE number that is merely reported and one that can be acted upon. Without a connection to the underlying losses, OEE remains a reporting figure. With a connection, it becomes a tool for prioritizing where improvement efforts will have the greatest impact.

Is 85% OEE really world-class?

Many articles claim that 85% is "world class" and that factories should aim for it. The truth is that 85% serves as an abstract reference point that rarely aligns with reality. In practice, the average OEE value in the industry is significantly lower, around 50% to 60% in Northern Europe. This level is confirmed by research from Chalmers on productivity in the manufacturing industry (Gopalakrishnan, Subramaniyan, and Skoogh, 2022) and is also found in previously published OEE research (Ljungberg 1998; Ingemansson 2004). It also aligns with Good Solutions' own customer data from the 300+ factories measuring on the platform, looking at factories that start measuring before improvement work has begun.

This means two things. First, most factories have huge potential for improvement without needing new equipment or new investments. The difference between 55% and 65% OEE corresponds to 18% more production from the same machine park. Second, the chase for 85% often becomes counterproductive. Production managers inflate the figure by discounting time that "doesn't count," such as changeovers, planned maintenance, and planned stops. The result is a number that looks good but does not reflect actual effectiveness.

The question "What is a good OEE value?" has no universal answer. It depends on the industry, production type, how the metric is defined internally, and what you are comparing it to. A more useful approach is to compare the factory to itself over time, focusing on the direction of travel, not on an absolute figure taken from a textbook.

Different ways to measure OEE

There are four common levels for how OEE data is collected, ranging from simple to more advanced.

Manual collection. Operators record stops and production by hand, often on paper or in spreadsheets. It is easy to start with, but data is often incomplete and hard to trust over time. Micro-stops are rarely captured.

Semi-automatic collection. The machine reports production and stops automatically, while the operator supplements this with reasons. This is often the most valuable level, as it combines reliable machine data with the operator's knowledge of why the stops occur.

Automatic collection from the machine's control system. Data is retrieved directly from a PLC or similar. This provides high precision in times and quantities, but cannot determine the cause of a stop on its own.

Integrated collection with ERP, MES, and maintenance systems. OEE data is linked to orders, articles, planning, and maintenance. This gives a more complete picture, but requires integrations to work and data quality to be high at every stage.

Most factories that succeed with OEE often use a combination. Automatic collection of times and quantities, supplemented by operator coding of causes. This ensures both precision and understanding.

What a good OEE system does, beyond calculating

An OEE system that only displays a number rarely helps a factory move forward. A system worth the investment highlights losses in real time, while the stop is ongoing, so the right person can take action immediately. It supports the operators' work by making it easy to classify stops and register scrap. It enables deep loss analysis, so OEE can be broken down by line, article, shift, cause, and time. And it integrates with the rest of the operations, so OEE data, quality data, maintenance data, and energy data can be viewed together.

These characteristics, not the length of the feature list, determine whether a system creates value in daily operations. Simply so that you can make better decisions faster and go from finding losses in production to implementing the right improvements and securing results over time.

How Good Solutions works with OEE

The platform from Good Solutions is built on the principle that OEE should go from reporting to action. Measurement is the means, improvement is the goal. The platform combines machine connectivity, operator tools, dashboards, timelines, reports, quality management, maintenance, andon, energy, and operational implementation in one coherent tool. Today, the platform supports 300+ factories, from individual production lines to global corporations.

Machine connection is done via common standards like OPC UA or via a proprietary IoT solution, such as RS IoT 4G from Good Solutions, which collects operating data from various sources. RS IoT 4G can, for example, analyze machine vibrations, power consumption, other digital signals, or a standard 24 V connection. It works on both new and legacy machines and requires neither a local network nor IT support to get started. The data is sent directly to Good Solutions' cloud service via the industrial 4G network. This allows even older equipment to provide reliable OEE data, which is a prerequisite for improvement work to encompass the entire factory.

Sibbhultsverken was able to improve OEE by 19.4% in 12 months. At Barilla Wasa, net production increased by 15% while CO₂ consumption decreased by 28%.

That type of result is not built by OEE measurement itself, but by the systematic improvement work that the platform makes easy to carry out.

Read more about how others have increased their factory's productivity

FAQ

What is a good OEE value?
It depends on the industry, production type, and how the metric is defined internally. The industrial average in Northern Europe is between 50 and 60%. The most important thing is not the number in absolute terms, but the direction over time and what the loss analysis shows. Comparing OEE against the factory's own history is usually more valuable than comparing against external benchmarks.

Is 85% OEE really world-class?
It is a widespread belief but in practice an abstract reference point. Few factories reach 85%, and those that report it have often defined away a large portion of the time. A more useful goal is to focus on the losses the factory actually has and improve OEE incrementally based on actual figures.

How do you calculate OEE for a shift?
Availability = operating time in the shift ÷ planned production time for the shift. Performance = (actual production × ideal cycle time) ÷ operating time. Quality = good units ÷ total units produced. The three are multiplied. The key is that definitions are consistent over time. Otherwise, it is impossible to compare figures between shifts.

What are the 6 big losses?
They are breakdowns, setups, idling and minor stops, reduced speed, defects in normal production, and defects during startup. The first two reduce availability, the middle two performance, and the last two quality. Micro-stops are included in the category of idling and minor stops within the six big losses and normally affect the performance part of OEE. They can be tracked separately in the analysis to highlight recurring problems, but do not constitute a seventh category in the model.

Do we need a separate OEE system or is our ERP enough?
ERP systems are built for business processes, i.e., orders, inventory, finance, and planning. They rarely have real-time data, machine connectivity, or the operator tools required for meaningful OEE work. A dedicated OEE system is used alongside the ERP and integrated with it. The common division is that the ERP keeps track of what is produced and the OEE system keeps track of how it is produced.