OEE Optimization by Artificial Intelligence

Industry 4.0 applications generate large amounts of data. Collecting data is good, analyzing this data is better – and translating the findings into results is the best. With oee.ai we have overcome the hurdle of simple and flexible capture. To visually prepare the data so that a person can recognize patterns is also solved. Making push notifications when the OEE is below Y% for more than X minutes also works. But with artificial intelligence there is much more …

Artificial intelligence as the upper grip and machine learning as a specific sub-area are developing rapidly. The method is efficient: Through the permanent processing of large amounts of data (Big Data) based on statistical analysis, the algorithm learns, so to speak, from experience – more, faster and better – than humans can. Then we let this work do the computer.

The application of OEE analysis and optimization is predestined for the pattern analysis of large amounts of data that humans no longer overlook:

  • Interference frequency as a function of the layer
  • Set-up times depending on the products to be prepared
  • Availability losses as a function of time

Using human intelligence and good data processing to perform a graphical analysis is relatively easy for experienced plant operators. Here are three examples from different companies:

Example 1: A very unstable running product. After setting it only struggles to reach the ridge line and before it comes to performance, the lot is already finished. The OEE for this order was 31.6%.

Example 2: A generally well-running product had five major disruptions during the runtime from 9:30 am to 6:00 pm.

Example 3: Between 1:00 am and 4:00 am, the equipped product was significantly worse than the other products on the system.

Artificial intelligence, however, looks for these patterns and exceptions around the clock and can prioritize the importance of individual events in the overall context – and graphically prepare the person for decision-making in graphic form.

An “abnormality cockpit” shows the special features in the data stream. This algorithm learns independently and thus becomes more and more secure in the classification over time. Humans are judged whether an event is an exception, and therefore not pursued further (e.g., components of the replacement supplier did not have sufficient quality), or whether it is marked for optimization.

This is the OEE improvement process of the future: quantified by an algorithm, data-driven, fact-driven and evaluated by humans and implemented with its solution creativity.

The technology for creating an “abnormality cockpit” is available. At the end of 2015, Google released TensorFlow, its machine learning software, under an open source license. Since then, the tool has become the market leader in this field of application. The TensorFlow framework combines with the programming language Python, which is e.g. Using scikit-learn to build a machine learning framework is a powerful programming environment for solving these issues.

Together with our scientific university environment, we research and develop the OEE analysis through artificial intelligence – the algorithm-based OEE optimization.

What’s Next?

We have sparked your interest. Contact us at info@oee.ai or call us: +49 (0) 241/401 842 75. We will gladly provide you with oee.ai sensors and access to the analysis cockpit for a trial period. Only then you decide. Also here we are easy!