Project Description

AUTOMATING INDUSTRIAL MANUFACTURING PROCESSES WITH MACHINE LEARNING
OPTIMATE’S STORY

“Together with Motius, we successfully proved that our design optimization concept works and that it has the potential to revolutionize the metal cutting industry.”

Jonas Steiling
CEO
Optimate

THE CHALLENGE

With a limited budget and three months time, Optimate aimed to prove that Machine Learning can help to predict if sheet metal parts can be optimized while keeping optimal product quality and functionality.

THE SOLUTION

A web demonstrator backed by a Machine Learning algorithm that easily visualizes the optimization potential of a metal building part and thereby serves as a proof of concept.

THE IMPACT

A business that sets out to shape the industrial manufacturing industry through AI-based optimization technology.

TECHNOLOGIES

  • Machine Learning
  • Convolutional Neural Networks

TEAM SIZE

YEAR

  • 2019 / 2020

CAN MACHINE LEARNING HELP TO PREDICT
IF SHEET METAL PARTS CAN BE OPTIMIZED?

The Optimate team noticed that sheet metal parts are often designed in a non-optimal way. On the one hand, there was the cost factor. In the metal manufacturing industry, every piece of metal wasted is money lost. On the other hand, there was the time factor. Although optimization efforts existed before, they were not time-efficient enough to justify the complex process.

They were looking for a way to automate the manual optimization efforts that existed before. The right solution would not only save time and money but would also keep the optimal product quality.

In the scope of TRUMPF’s intrapreneurship program, the solution had to secure further investments that enable Optimate to become a spin-off company.

Optimate set out to build a design optimization web demonstrator backed by a machine learning algorithm as a proof of concept.

CHOOSING THE RIGHT MACHINE LEARNING ALGORITHM

The first challenge was generating labeled data for sheet metal parts because nobody had done that before. We had to collect big amounts of data, validate relevant and sort out irrelevant data. After that, we had to choose between supervised and semi- supervised learning. In order to realistically assess, which kind of ML algorithm would work best, we tested both options through hyperparamter tuning. The tests revealed that the semi-supervised machine learning delivered slightly better results.

BUILDING THE WEB DEMONSTRATOR

The web demonstrator had to be backed by the developed ML algorithm and easily visualize the optimization potential of a metal building part. Again, Optimate and Motius collaborated closely and used their combined expertise to take this final step towards the POC.
The result was a web demonstrator that represents the whole desired optimization process. It analyzes uploaded files through machine learning and visualizes the metal part’s optimization potentials through a user-friendly traffic light system.

Interested in further insights?

Jannik Finkbohner
Head of Think Tank | Stuttgart
jannik.finkbohner@motius.de

Read more about the story on our blog

READ MORE
Optimate-team

The Optimate Team: Max Hesselbarth, Martina Trinczek, Jonas Steiling

REVOLUTIONIZING THE INDUSTRIAL MANUFACTURING INDUSTRY

The web demonstrator successfully served as a PoC for the Optimate team and secured further investments. Their design optimization concept that successfully enabled Optimate to become a spin-off company. This means that Optimate sets out to shape the industrial manufacturing through design optimization powered by cutting-edge technologies.

The successful project is also the basis for a long-term collaboration between Optimate and Motius to further optimize the tool and establish Optimate in the market.

“We are very satisfied with the cooperation with Motius and look forward to taking the next steps together.”
Jonas Steiling, CEO, Optimate

“Besides the success of our initial project, I am excited for our long- term collaboration with Optimate to shape the industrial manufacturing industry.”
Jannik Finkbohner, Head of Think Tank Stuttgart, Motius

WANT TO BUILD A PRODUCT OF THE FUTURE?

TALK TO US
JOIN MOTIUS

HOW ALLIANZ OPTIMIZES
CUSTOMER SUPPORT

USING A HOLISTIC CHATBOT SOLUTION

READ THE STORY