The calculation of the OEE and what this means

The three factors of availability, performance, and quality, along with the OEE, are expressed as percentages between 0-100%. Example: A system has an availability of 90%, a performance of 95%, and a quality of 98%. In this case, the OEE of the system would be 90% x 95% x 98% = 84.1%.

Availability, Performance, Quality

Due to multiplication, the OEE depends on all three factors: availability, performance, and quality. This means that if one of the factors is poor, it pulls the entire OEE down, even though the other two factors remain very good. Therefore, the individual factors of availability, performance, and quality should always be examined. This helps come closer to the root causes of problems.

Let’s take a closer look at the three factors:

The availability factor provides insight into how long the system was actually producing while it was scheduled for production. Any downtime that occurs reduces the availability of the system. There is a distinction between planned downtimes (e.g., setup, planned maintenance) and unplanned downtimes (sudden machine failures requiring unplanned repairs).

The performance factor indicates how far from the maximum production speed production is taking place. From our perspective, it is necessary to always consider the product being manufactured at that moment. Not all products can be produced at the same speed on the system. A detailed explanation of the performance factor can be found here.

The quality factor indicates how much of the produced product can be processed further or sold to customers. This is calculated by relating the quantity of good products to the total quantity. The quantity of good products can be calculated through the scrap numbers.

System Availability and OEE

What is system availability?

System availability is a crucial factor for determining overall equipment effectiveness (OEE). It indicates how much time the system has been productive. If the system is idle when it should be producing, productive time is lost, leading to a decrease in OEE.

Availability is defined as the ratio of actual productive time to planned productive time (according to the production schedule). Sometimes, planned downtimes are subtracted from this planned time to prevent them from negatively affecting availability. The reasons why this can lead to problems will be explained further below.

How is system availability calculated?

System availability is calculated using the quotient availability = actual productive time / planned productive time (e.g., in hours), extended with the factor of 100% to present it as a percentage (0-100%).

The planned productive time here corresponds to the time in which the system is scheduled for production according to the production plan.

Downtimes: Planned, Unplanned, No Production Planned

Ultimately, availability depends on the number and length of downtimes. There are different opinions on which downtimes should be considered in the calculation of availability.

If the system is idle, there can be various reasons for this. In many cases, the system will not be planned for production. The system is ready and could produce, but there is no production order because, for example, there are not enough orders, no worker is planned for the system, etc. These reasons are not the responsibility of production planning and should therefore not be included in the OEE calculation, as it would obscure the actual problems.

In addition, there are planned and unplanned downtimes. Planned downtimes are those that are part of the production process, such as setup processes or material changes. Some OEE calculations do not observe these planned downtimes. We advise against this: while these planned downtimes cannot be prevented, a detailed look and comparison of these downtimes often pays off. For example, different products often experience varying setup times, which can indicate disturbances in the production process.

Unplanned downtimes, on the other hand, unequivocally contribute to reduced availability. These include unexpected machine failures due to equipment damage, proactive, unplanned maintenance of machine parts to prevent damage, material congestion, and sudden worker absence due to illness, as well as short interruptions such as restroom breaks and pauses.

Here you can read how ENLYZE automatically detects and distinguishes downtimes.

How can I increase system availability?

First and foremost, when considering planned downtimes, it is possible to reduce unnecessary planned downtimes. Even if they are planned, they remain downtimes when no production can occur, which are often not optimized due to faulty planning. Other factors that influence system availability include:

• Frequency of failures and breakdowns

• Duration of maintenance and repair work

• Quality of maintenance and repair work

• Availability of spare parts

• Qualification and experience of the staff

• Environmental conditions

OEE: How Performance and Quality are Calculated

What does performance mean in the context of OEE?

Performance is the key factor for the technical condition and proper setting of the machine. It indicates how much product is produced by the machine in relation to how much could be produced.

Determining the "possible" amount of end product is quite significant. Ultimately, it affects how high the performance factor will be. In industry, manufacturer specifications are often referenced, which unfortunately is not realistic. Machines can sometimes be decades old, and these maximum values have not been achieved for years. How to do this better is explained in this article.

How is performance calculated?

Performance is calculated based on the current throughput (amount of product per time, e.g., hour) in relation to the maximum possible throughput.

During traditional, manual OEE recording, the amount of material produced per hour is often weighed, documented, and compared at regular intervals. The reference value, the maximum throughput, is often based on the manufacturer's specifications, which frequently cite an unrealistically high throughput per hour for their machines. Moreover, the performance of a system changes with age, and different products may only be processed at certain throughputs due to process conditions. Therefore, it is worthwhile to set the reference value for different products individually and update it more frequently.

At ENLYZE, we use throughput from machine data for performance determination. The PLC continuously collects speed data from which the throughput (integral of speed over time) can be calculated. We use this to continuously display the current performance. Additionally, the maximum throughput is automatically adjusted when we record new top performances.

Conclusion: How should the maximum throughput be determined?

We recommend not determining the maximum throughput based on manufacturer's specifications but rather based on the machine's experiential values. Over a longer period, the throughput should be routinely determined and the best achieved value set as the maximum.

Furthermore, a separate performance calculation for each product manufactured on the system is advisable. Two completely different products A and B, with different diameters, quality characteristics, etc., cannot be compared using a single performance value.

More on this topic can be found in this in-depth article on performance losses

What is quality and how is it calculated?

The quality of a product indicates how much end product (= good quantity) is usable, i.e., not scrap. It is calculated using the quotient quality = good quantity / total quantity. The total quantity corresponds to the amount of raw materials processed by the system.

Total quantities can be easily calculated from machine data. Alternately, they can be manually weighed, either through the amount of raw materials or through the sum of good quantity (amount of end products) + scrap.

We have gathered more information on quality and scrap in this article

OEE in the Context of Manufacturing

Advantages of OEE Calculation in Manufacturing

OEE is a very powerful tool capable of uncovering and solving a multitude of problems. It can help improve the effectiveness of a system by pointing out weaknesses that can then be eliminated.

OEE is also very useful for evaluating the effects of changes. Frequent use cases include the introduction of new materials. Through OEE-supported experiments, the productivity of the new material can be optimized quickly.

What is a good OEE value?

A statement about a "good OEE" cannot be made so easily. Manufacturing processes and systems in industry are simply too varied. A good OEE value is often considered to be around 80 percent, which often holds true for our customers. However, it should be noted that OEE is often misunderstood as an absolute number and is often distorted to meet unrealistic targets.

Thus, the reference values for performance and availability can be set too low, resulting in an OEE of 100 percent or more. However, when the OEE value is measured based on well-chosen indicators, it is a reliable index for productivity in manufacturing. It indicates how much time a machine is truly utilized productively.

Feel free to sign up here for our 4-week bootcamp. Over 4 weeks, you will receive emails twice a week, covering everything around OEE, such as typical implementation problems and how you can truly profit from it. At ENLYZE, for instance, we have solved the issue of well-chosen target specifications by calculating your OEE based on data-driven and verifiable machine data.

Is an OEE of 100% possible?

Short answer: Yes, but unlikely. Many companies “trick” in calculating such an OEE.

Various approaches to calculation make it difficult to compare the OEE value with other companies and decide whether it is “good” or “bad.” In practice, it is often said that an OEE value of 60 to 70 percent is considered average. A “good” OEE value, as mentioned earlier, is already considered to be at 80 percent.

We place more value on calculating the correct OEE and based on that uncover realistic potential for improvement. Such an integration of OEE into your production allows you to identify real issues in manufacturing and continuously steer productivity to a maximum specific to the operation.

OEE and “Lean Manufacturing”

The OEE is an essential metric for Lean Manufacturing. The OEE measures the productivity of a factory concerning quality, output (performance), and load (availability). By calculating the OEE, a company can determine how much product it produces compared to its potential maximum and where improvement potentials lie.

Lean Manufacturing aims to avoid waste and maximize production. OEE contributes to achieving this goal by making the productivity of a factory measurable. This enables companies to identify weaknesses and take measures to improve OEE. Increasing OEE also enhances efficiency according to the principles of Lean Manufacturing.

Moreover, OEE supports the implementation of Lean Manufacturing by creating transparency, thereby prioritizing the biggest problems in manufacturing and the greatest potentials for improvements. In this way, companies can determine where most machine failures occur and what causes are responsible for them using OEE. Consequently, measures can be taken to reduce failure rates, thereby enhancing manufacturing efficiency.

The Six Big Losses in Manufacturing

In production, there are 6 major losses that negatively impact overall equipment effectiveness (OEE). To improve OEE, these losses must be systematically minimized. They can be effectively sorted by the three OEE factors:

Availability:

1. Scheduled maintenance: Process-related downtimes, such as setups or reconfigurations.

2. Unplanned downtimes: Too frequent unplanned breakdowns arising, for example, from technical errors.

Performance:

3. Short stops: Brief stops (a few seconds) are often not considered in availability but notably reduce performance per hour.

4. Slow cycle times: The system operates slower than it should – a typical performance drop.

Quality:

5. Ongoing scrap: Damage to the product occurs due to unforeseen process errors.

6. Startup scrap: Process-related, a certain amount of scrap is often produced when starting systems until they really get warmed up.

Do you want to dive deeper into the topic of the 6 Big Losses? Learn here how the 6 Big Losses help identify the largest optimization levers on the shop floor.

Implementing OEE in Manufacturing

The Challenges of OEE Calculation

Calculating overall equipment effectiveness is a complex task that brings many challenges. First, the data needed for the calculation must be gathered and processed. This is often a time-consuming and tedious task, especially when done manually, as the data frequently comes in different formats and systems.

Additionally, the relevance of OEE depends on how detailed the data is collected. How is it ensured that the correct reasons for downtimes (stoppages) are always provided? Why is the system currently running slower than usual, and can the worker identify the correct data for this?

To correctly determine OEE, several points must be considered.

Steps to Implement an OEE System:

1. First, it should be clear how OEE is to be calculated. By default, it is calculated based on availability, performance, and quality, but it can be expanded as needed.

2. Next, measures for the regular collection of these factors should be defined. This can be done traditionally and manually or fully automatically via specialized systems. For instance, a system's performance can be determined by weighing the end products produced within a specific time frame (e.g., 1 hour) or calculated automatically based on speeds.

3. OEE should be regularly calculated from these factors and, more importantly, made accessible to relevant parties. It makes no sense for the production manager to calculate OEE every Friday and enter it into an Excel sheet to which only he has access.

4. From OEE or its factors, the major weaknesses of production can then be identified. This helps derive targeted possible improvement measures. If availability is unexpectedly low, for example, production plans may need to be revised. If this is due to unplanned downtimes, a deeper analysis based on machine parameters must take place. What errors are accumulating, and what causes are there for them?

→ Again, the more data available, the better problems can be analyzed. Workers should systematically document anomalies at the system to find connections. If OEE is calculated fully automatically, these interactions can be analyzed based on the collected machine data.

Calculating OEE (automated)

OEE is an important KPI for production; however, traditional calculation is often cumbersome and error-prone as it is based on many different data points. Solutions are systems that automatically calculate OEE.

During manual collection, production and machine data are documented at regular intervals either in manual logs or in MES/ERP systems. However, manual data collection is time-consuming and leads to inaccurate data.

By automatically reading machine data, OEE can be continuously captured automatically in the background. Data quality improves, and floor staff are relieved and can focus on value-added work rather than on tedious manual data collection.

Today's manufacturing management systems (MES) often lack this access to machine data and require a large number of manual entries to calculate OEE. OEE is often calculated using simple sensors, such as light barriers and pulse signals from the PLC. In the latter case, only the PLC's signal indicating whether the system is running or not is measured. Unfortunately, this value is mistakenly provided by many vendors as OEE but only represents the system's availability.

Systems that capture all machine data, such as those from ENLYZE, provide in-depth insights into systems. All relevant data from the PLC and possibly other required sensor data is read directly at the machine and continuously collected. OEE can then be continuously and realistically calculated from this data. Additionally, extensive analyses beyond OEE are enabled. This allows for direct tracing of causes for quality and production issues. In this way, you have your production always in view and can quickly respond to problems.