From 45 % to 0 % mistakes: How a German AI solves the biggest problem in industry

Instead of only a few clicks instead of months: how intelligent software factories set up perfectly straight away

Imagine a very complicated machine in a factory, for example one that paints auto parts or coated microchips. This machine has many “controllers” and “buttons” (parameters), such as temperature, pressure, speed, duration, voltage, etc.

The problem that needs to be solved

The big problem for factories is: How do you set all these controllers perfectly to achieve the best result?

The best result can mean: the highest quality (e.g. painting without errors), the least committee, the lowest energy consumption or the fastest production time.

The traditional method is trying out (trial and error). You test an attitude, look at the result, turn a bit on a controller, test again, and so on. This is extremely expensive, takes a long time and wastes a lot of material.

The solution from Machoptima

Machoptima replaces this lengthy “try” with an intelligent abbreviation with the help of artificial intelligence (AI).

Your procedure can be described in a circulation

1. Input: A factory gives a few test runs with different settings (input parameters) of Machoptima.
2. Analysis: Machoptima's AI software looks at the results of these tests (e.g. “With setting A, the paint thickness was 95% perfect”, “With only 60%”).
3. Intelligent forecast: Instead of simply guiding the next setting, the AI ​​calculates which new settings will most likely lead to an even better result. It learns from the few data how the controllers are related.
4. Optimization: The AI ​​proposes a new, promising attitude to the factory. You take a new test and the circulation begins all over - but it comes close to the perfect result much, much faster.

The most important thing

• Non-invasive: You do not convert the factory. They only deliver the “perfect recipes” for the existing machines.
• Data efficient: You don't need thousands of test runs. Your AI is so smart that she can learn from very few data. This is crucial because every test run in a factory is expensive.

Success Stories: What does that mean in practice?

• For Bosch (microchips): Machoptima helped to optimize the settings for coating microchips so that the error rate was reduced from 45 % to almost zero (0.003 %). Savings: 85 % time and costs.
• For Mercedes-Benz (car paint): Machoptoma the perfect settings for painting the body. This not only saved tests, but also reduced downhill times in production. Savings: 80 % time and costs.
• For the Max Planck Institute (simulations): Machoptima has helped to make complex computer simulations much more precise by finding the right ones from 9.8 million possible settings-with a fraction of the otherwise necessary test.

Specialized AI optimizes complex manufacturing processes in record time

Machoptima uses a special AI to find the “perfect recipe for attitudes” for complex industrial processes at lightning speed, which saves an enormous amount of time, money and material for the companies.

So they are not a typical AI company that develops chatbots such as chatt, but use a specialized form of AI (machine learning for optimization) in order to fundamentally improve real, physical processes in industry. Your strong connection to top research (Max Planck Institute) gives you a technological lead.

Industrial meets about Machoptima - research

Machoptima is a spin-off project of the Max Planck Institute for Intelligent Systems and part of Max! Mize, the official start-up link in the Max Planck Society.

The goal: to transferred groundbreaking research into industrial applications. With AI-based technology, Machoptima companies help to make production processes more efficient, more precisely and error-free.

Machoptima combines scientific excellence with practical use - for a flawless industry of tomorrow.

Dr. Richard Hagl

I would be happy to serve as your personal advisor.

Machoptima Interim Manager

LinkedIn