OEE Calculation

Performance is crucial in a production or assembly line, but how can you measure whether operations are being carried out correctly and optimally? Each company has its own measurement system, but the vast majority fall under the underlying concept of OEE calculation. The following will present what is required to calculate this KPI and the benefits it brings in terms of control and decision-making.

Shop-floor data capture

To calculate these KPIs, certain data are required, which can be collected manually or automatically through a variety of means (e.g., time collection, PLCs, etc.). In most cases, these data fit into a table of tagged events where effective working hours and any incidents that may occur are reported.

All these events must be associated with the level of granularity we want to achieve, which may be plant, line, operator, product, or even operation. This granularity is largely determined by the existence of standards for a given level of detail; if you have a standard to meet at plant level, this will result in an OEE per plant, and analogously for the chosen granularity. The optimal approach is to meet the standard at the highest level of granularity. If you know the OEE of an operation performed by a specific operator, you will be able to infer the plant’s OEE.

Objectives when calculating OEE

With all these variables in the mix, you can proceed to calculate the KPI. This calculation varies depending on what exactly you want to measure, but in essence it must answer the key questions:

1. How well is the work being done?
2. How many incidents have occurred
3. What types of incidents have occurred?

By answering these questions, we will be able to monitor the status of our production, operator performance, and use this valuable information to make data-driven decisions accordingly. This is where the standards mentioned in the previous point become important; these standards will be the key element, as they are the benchmark to beat. If a standard is poorly estimated, it will lead to an underestimation or overestimation of OEE and therefore it will lack real value.

OEE visualization and navigation

When analyzing this important KPI, we will need a dynamic tool capable of adapting to the multitude of operations to be performed with this coefficient.

There are several dimensions that interact with OEE calculation and its granularity, such as:

• Date: an OEE calculation date; it may be the event date, product completion date, operation completion date, etc.
• Hierarchical analysis: descending from the highest company or plant level down to the user-operation level.
• Incidents: answering the questions of in which operation or product a specific type of incident occurred and what impact it had.
• Company-specific categorizations.

These variables allow us to analyze this KPI across several dimensions, helping to identify potential bottlenecks. Let us assume a practical case:

We observe in the last image that our OEE is 58.65%, somewhat low, so let us identify a potential bottleneck. We start by reviewing the activities

We identify a project with an excessively low OEE. We can focus the case on this specific project and study everything from the most problematic parts to the incidents that have occurred, as illustrated in the following images:

We can see that the most problematic products in the project are 57 and 54, and that their main source of error—despite being performance in terms of incidents—shows that most are displayed in blue; therefore, they are quality-related. To continue exploring in depth, let us see how much these incidents actually weigh on quality.

Simply by selecting the type of incident to be analyzed, we can narrow the focus and verify that 20 of the 255 hours spent are quality defects. This analysis can continue along these lines, identifying that rework occurred in one of the products and an NCR incident in another. By also identifying the operators who performed these operations, we can establish a contingency plan—for example, providing training to these operators to ensure a reduction in quality failures or considering supervision.

As we can see, with a quick look at the data we can identify real situations and improve our production line with targeted measures against bottlenecks.

Conclusions

The calculation of OEE, or performance indicator, in a manufacturing company or one with an operations-based nature is essential, as it allows us to carry out a study that ranges from a high production level down to the identification of critical processes. This KPI, presented in a dynamic visualization that enables intuitive, user-friendly analysis, provides a fundamental tool for improving production, well-being, and decision-making.