In a time in which companies need more innovative product ideas than ever, intrapreneurship still has not gained the attention it deserves. A 2018 study among more than 600 companies revealed that only about 20% invest resources in their own employees in order to create radical innovations bottom-up and in-house.
Although intrapreneurship still seems to be ignored by most companies, it offers a range of potential benefits. For instance:
- enables consistent and efficient innovation efforts
- exploits unseen potentials within a company’s competitive landscape
- bootstraps innovative ideas
- reduces time-to-market and
- boosts company culture as well as talent development
Lately, “Internehmertum”, the intrapreneurship program of TRUMPF, gave rise to Optimate – another promising spin-off from TRUMPF’s intrapreneurship program that is entering the market.
Optimization opportunities in industrial metal manufacturing
The original idea behind Optimate comes from Martina, Jonas and Max. Through “design optimization” the team wants to change the way sheet metal parts are processed and thereby save costs as well as time while keeping optimal product quality and functionality.
Optimate minimizes the amount of waste that the metal cutting process produces.
Complementing industry expertise with cutting-edge technology expertise
In the scope of the intrapreneurship program, the Optimate team got a limited budget and three months’ time to develop a Proof of Concept for their idea. To do so, they intended to build a web-based demonstrator backed by a Machine Learning algorithm that would prove the technical feasibility and lead to further investments in Optimate.
So, in order to reach their goal, the Optimate team reached out to Jannik, Head of Think Tank at Motius’ Stuttgart office. Through combining Optimate’s industrial manufacturing expertise with cutting-edge technology expertise from Motius, we set out to prove that Machine Learning can help to predict whether a sheet metal part can be optimized.
In the end, the whole process was supposed to look like this: Input file that contains data, analyze optimization potentials through ML, Visualize optimization potentials in a user-friendly name.
Phase 1: Getting the data that the Machine Learning algorithm needs
The first challenge that Optimate and Motius faced was generating labeled data for sheet metal parts because nobody had done that before. But without this data, we could not have developed a working Machine Learning algorithm.
Under enormous time constraints, the collaboration between Optimate and Motius excelled. Thanks to Optimate’s deep industry expertise, Motius was able to focus on the technical part of the project while weekly sprints kept everyone on track.
On a technical level, the team had to collect big amounts of data, validate relevantly and sort out irrelevant data. Whereas, for instance, Optimate’s team knew which features determine the optimization potential of sheet metal parts, Motius’ team used this information to build the ML algorithm accordingly.
Steps taken in phase 1.
At the end of this phase, the data split, imputation and normalization as well as resampling brought us the data we needed to enter the second phase.
Phase 2: Choosing between supervised and semi-supervised learning
In building the ML algorithm, we had to choose between supervised and semi-supervised learning. The difference is that supervised machine learning algorithms learn only from labeled training data while semi-supervised machine learning algorithms learn from a small amount of labeled and a large amount of unlabeled training data.
“We decided to test both options in order to realistically assess which kind of ML algorithm would work best for our project”, explains Jonas from Optimate.
The Optimate Team: Max Hesselbarth, Martina Trinczek, Jonas Steiling
Jannik from Motius illustrates: “From a tech perspective, this process called “hyperparameter tuning” aims to optimize those parameters that control the learning process of the ML algorithm. In other words, it tunes the model so that it can optimally solve the given problem.”
In this project, the semi-supervised machine learning delivered slightly better results. “Based on the results, it was obvious that Machine Learning can help to predict whether a sheet metal part can be optimized. As soon as we achieved this milestone, we went on to develop the web demonstrator as the final step to the Proof of Concept”, says Max from Optimate.
Phase 3: Building the web demonstrator to bring the idea to life
As said, 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 a PoC.
The result was a web demonstrator that represents the whole desired optimization process from the beginning. It analyzes uploaded files through machine learning and visualizes the metal part’s optimization potentials through a user-friendly traffic light system.
With the web demonstrator, you simply upload your file first.
And then it automatically tells you whether your sheet metal parts are optimizable.
Securing further investments and preparing for the future
“Together with Motius, we successfully proved that our idea works and that it has the potential to optimize the manufacturing process. The web demonstrator perfectly visualized our work and the Proof of Concept that we achieved”, describes the Optimate team.
“As soon as it was clear that Optimate’s team would get the chance to keep developing their idea in their own spin-off company, we started thinking about future improvements and what we need to develop in order to create a profitable business model with them”, says Jannik from Motius.
Based on the successful collaboration in the scope of this project, Optimate and Motius decided to keep working together to further optimize the tool and establish Optimate at the market. The fact that Optimate secured further investments despite current constraints due to the pandemic underlines this potential.
With that outlook, this project is more than just a Proof of Concept. Instead, it is the basis for a long-term collaboration and a business that sets out to shape the industrial manufacturing industry through AI-based optimization technology.
If you want to more about the vast potential of Machine Learning, just talk to our experts below.
In a time in which companies need more innovative product ideas than ever, intrapreneurship still has not gained the attention it deserves. A 2018 study among more than 600 companies revealed that only about 20% invest resources in their own employees in order to create radical innovations bottom-up and in-house.
Although intrapreneurship still seems to be ignored by most companies, it offers a range of potential benefits. For instance:
- enables consistent and efficient innovation efforts
- exploits unseen potentials within a company’s competitive landscape
- bootstraps innovative ideas
- reduces time-to-market and
- boosts company culture as well as talent development
Lately, “Internehmertum”, the intrapreneurship program of TRUMPF, gave rise to Optimate – another promising spin-off from TRUMPF’s intrapreneurship program that is entering the market.
Optimization opportunities in industrial metal manufacturing
The original idea behind Optimate comes from Martina, Jonas and Max. Through “design optimization” the team wants to change the way sheet metal parts are processed and thereby save costs as well as time while keeping optimal product quality and functionality.
Optimate minimizes the amount of waste that the metal cutting process produces.
Complementing industry expertise with cutting-edge technology expertise
In the scope of the intrapreneurship program, the Optimate team got a limited budget and three months’ time to develop a Proof of Concept for their idea. To do so, they intended to build a web-based demonstrator backed by a Machine Learning algorithm that would prove the technical feasibility and lead to further investments in Optimate.
So, in order to reach their goal, the Optimate team reached out to Jannik, Head of Think Tank at Motius’ Stuttgart office. Through combining Optimate’s industrial manufacturing expertise with cutting-edge technology expertise from Motius, we set out to prove that Machine Learning can help to predict whether a sheet metal part can be optimized.
In the end, the whole process was supposed to look like this: Input file that contains data, analyze optimization potentials through ML, Visualize optimization potentials in a user-friendly name.
Phase 1: Getting the data that the Machine Learning algorithm needs
The first challenge that Optimate and Motius faced was generating labeled data for sheet metal parts because nobody had done that before. But without this data, we could not have developed a working Machine Learning algorithm.
Under enormous time constraints, the collaboration between Optimate and Motius excelled. Thanks to Optimate’s deep industry expertise, Motius was able to focus on the technical part of the project while weekly sprints kept everyone on track.
On a technical level, the team had to collect big amounts of data, validate relevantly and sort out irrelevant data. Whereas, for instance, Optimate’s team knew which features determine the optimization potential of sheet metal parts, Motius’ team used this information to build the ML algorithm accordingly.
Steps taken in phase 1.
At the end of this phase, the data split, imputation and normalization as well as resampling brought us the data we needed to enter the second phase.
Phase 2: Choosing between supervised and semi-supervised learning
In building the ML algorithm, we had to choose between supervised and semi-supervised learning. The difference is that supervised machine learning algorithms learn only from labeled training data while semi-supervised machine learning algorithms learn from a small amount of labeled and a large amount of unlabeled training data.
“We decided to test both options in order to realistically assess which kind of ML algorithm would work best for our project”, explains Jonas from Optimate.
The Optimate Team: Max Hesselbarth, Martina Trinczek, Jonas Steiling
Jannik from Motius illustrates: “From a tech perspective, this process called “hyperparameter tuning” aims to optimize those parameters that control the learning process of the ML algorithm. In other words, it tunes the model so that it can optimally solve the given problem.”
In this project, the semi-supervised machine learning delivered slightly better results. “Based on the results, it was obvious that Machine Learning can help to predict whether a sheet metal part can be optimized. As soon as we achieved this milestone, we went on to develop the web demonstrator as the final step to the Proof of Concept”, says Max from Optimate.
Phase 3: Building the web demonstrator to bring the idea to life
As said, 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 a PoC.
The result was a web demonstrator that represents the whole desired optimization process from the beginning. It analyzes uploaded files through machine learning and visualizes the metal part’s optimization potentials through a user-friendly traffic light system.
With the web demonstrator, you simply upload your file first.
And then it automatically tells you whether your sheet metal parts are optimizable.
Securing further investments and preparing for the future
“Together with Motius, we successfully proved that our idea works and that it has the potential to optimize the manufacturing process. The web demonstrator perfectly visualized our work and the Proof of Concept that we achieved”, describes the Optimate team.
“As soon as it was clear that Optimate’s team would get the chance to keep developing their idea in their own spin-off company, we started thinking about future improvements and what we need to develop in order to create a profitable business model with them”, says Jannik from Motius.
Based on the successful collaboration in the scope of this project, Optimate and Motius decided to keep working together to further optimize the tool and establish Optimate at the market. The fact that Optimate secured further investments despite current constraints due to the pandemic underlines this potential.
With that outlook, this project is more than just a Proof of Concept. Instead, it is the basis for a long-term collaboration and a business that sets out to shape the industrial manufacturing industry through AI-based optimization technology.
If you want to more about the vast potential of Machine Learning, just talk to our experts below.
In a time in which companies need more innovative product ideas than ever, intrapreneurship still has not gained the attention it deserves. A 2018 study among more than 600 companies revealed that only about 20% invest resources in their own employees in order to create radical innovations bottom-up and in-house.
Although intrapreneurship still seems to be ignored by most companies, it offers a range of potential benefits. For instance:
- enables consistent and efficient innovation efforts
- exploits unseen potentials within a company’s competitive landscape
- bootstraps innovative ideas
- reduces time-to-market and
- boosts company culture as well as talent development
Lately, “Internehmertum”, the intrapreneurship program of TRUMPF, gave rise to Optimate – another promising spin-off from TRUMPF’s intrapreneurship program that is entering the market.
Optimization opportunities in industrial metal manufacturing
The original idea behind Optimate comes from Martina, Jonas and Max. Through “design optimization” the team wants to change the way sheet metal parts are processed and thereby save costs as well as time while keeping optimal product quality and functionality.
Optimate minimizes the amount of waste that the metal cutting process produces.
Complementing industry expertise with cutting-edge technology expertise
In the scope of the intrapreneurship program, the Optimate team got a limited budget and three months’ time to develop a Proof of Concept for their idea. To do so, they intended to build a web-based demonstrator backed by a Machine Learning algorithm that would prove the technical feasibility and lead to further investments in Optimate.
So, in order to reach their goal, the Optimate team reached out to Jannik, Head of Think Tank at Motius’ Stuttgart office. Through combining Optimate’s industrial manufacturing expertise with cutting-edge technology expertise from Motius, we set out to prove that Machine Learning can help to predict whether a sheet metal part can be optimized.
In the end, the whole process was supposed to look like this: Input file that contains data, analyze optimization potentials through ML, Visualize optimization potentials in a user-friendly name.
Phase 1: Getting the data that the Machine Learning algorithm needs
The first challenge that Optimate and Motius faced was generating labeled data for sheet metal parts because nobody had done that before. But without this data, we could not have developed a working Machine Learning algorithm.
Under enormous time constraints, the collaboration between Optimate and Motius excelled. Thanks to Optimate’s deep industry expertise, Motius was able to focus on the technical part of the project while weekly sprints kept everyone on track.
On a technical level, the team had to collect big amounts of data, validate relevantly and sort out irrelevant data. Whereas, for instance, Optimate’s team knew which features determine the optimization potential of sheet metal parts, Motius’ team used this information to build the ML algorithm accordingly.
Steps taken in phase 1.
At the end of this phase, the data split, imputation and normalization as well as resampling brought us the data we needed to enter the second phase.
Phase 2: Choosing between supervised and semi-supervised learning
In building the ML algorithm, we had to choose between supervised and semi-supervised learning. The difference is that supervised machine learning algorithms learn only from labeled training data while semi-supervised machine learning algorithms learn from a small amount of labeled and a large amount of unlabeled training data.
“We decided to test both options in order to realistically assess which kind of ML algorithm would work best for our project”, explains Jonas from Optimate.
The Optimate Team: Max Hesselbarth, Martina Trinczek, Jonas Steiling
Jannik from Motius illustrates: “From a tech perspective, this process called “hyperparameter tuning” aims to optimize those parameters that control the learning process of the ML algorithm. In other words, it tunes the model so that it can optimally solve the given problem.”
In this project, the semi-supervised machine learning delivered slightly better results. “Based on the results, it was obvious that Machine Learning can help to predict whether a sheet metal part can be optimized. As soon as we achieved this milestone, we went on to develop the web demonstrator as the final step to the Proof of Concept”, says Max from Optimate.
Phase 3: Building the web demonstrator to bring the idea to life
As said, 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 a PoC.
The result was a web demonstrator that represents the whole desired optimization process from the beginning. It analyzes uploaded files through machine learning and visualizes the metal part’s optimization potentials through a user-friendly traffic light system.
With the web demonstrator, you simply upload your file first.
And then it automatically tells you whether your sheet metal parts are optimizable.
Securing further investments and preparing for the future
“Together with Motius, we successfully proved that our idea works and that it has the potential to optimize the manufacturing process. The web demonstrator perfectly visualized our work and the Proof of Concept that we achieved”, describes the Optimate team.
“As soon as it was clear that Optimate’s team would get the chance to keep developing their idea in their own spin-off company, we started thinking about future improvements and what we need to develop in order to create a profitable business model with them”, says Jannik from Motius.
Based on the successful collaboration in the scope of this project, Optimate and Motius decided to keep working together to further optimize the tool and establish Optimate at the market. The fact that Optimate secured further investments despite current constraints due to the pandemic underlines this potential.
With that outlook, this project is more than just a Proof of Concept. Instead, it is the basis for a long-term collaboration and a business that sets out to shape the industrial manufacturing industry through AI-based optimization technology.
If you want to more about the vast potential of Machine Learning, just talk to our experts below.
In a time in which companies need more innovative product ideas than ever, intrapreneurship still has not gained the attention it deserves. A 2018 study among more than 600 companies revealed that only about 20% invest resources in their own employees in order to create radical innovations bottom-up and in-house.
Although intrapreneurship still seems to be ignored by most companies, it offers a range of potential benefits. For instance:
- enables consistent and efficient innovation efforts
- exploits unseen potentials within a company’s competitive landscape
- bootstraps innovative ideas
- reduces time-to-market and
- boosts company culture as well as talent development
Lately, “Internehmertum”, the intrapreneurship program of TRUMPF, gave rise to Optimate – another promising spin-off from TRUMPF’s intrapreneurship program that is entering the market.
Optimization opportunities in industrial metal manufacturing
The original idea behind Optimate comes from Martina, Jonas and Max. Through “design optimization” the team wants to change the way sheet metal parts are processed and thereby save costs as well as time while keeping optimal product quality and functionality.
Optimate minimizes the amount of waste that the metal cutting process produces.
Complementing industry expertise with cutting-edge technology expertise
In the scope of the intrapreneurship program, the Optimate team got a limited budget and three months’ time to develop a Proof of Concept for their idea. To do so, they intended to build a web-based demonstrator backed by a Machine Learning algorithm that would prove the technical feasibility and lead to further investments in Optimate.
So, in order to reach their goal, the Optimate team reached out to Jannik, Head of Think Tank at Motius’ Stuttgart office. Through combining Optimate’s industrial manufacturing expertise with cutting-edge technology expertise from Motius, we set out to prove that Machine Learning can help to predict whether a sheet metal part can be optimized.
In the end, the whole process was supposed to look like this: Input file that contains data, analyze optimization potentials through ML, Visualize optimization potentials in a user-friendly name.
Phase 1: Getting the data that the Machine Learning algorithm needs
The first challenge that Optimate and Motius faced was generating labeled data for sheet metal parts because nobody had done that before. But without this data, we could not have developed a working Machine Learning algorithm.
Under enormous time constraints, the collaboration between Optimate and Motius excelled. Thanks to Optimate’s deep industry expertise, Motius was able to focus on the technical part of the project while weekly sprints kept everyone on track.
On a technical level, the team had to collect big amounts of data, validate relevantly and sort out irrelevant data. Whereas, for instance, Optimate’s team knew which features determine the optimization potential of sheet metal parts, Motius’ team used this information to build the ML algorithm accordingly.
Steps taken in phase 1.
At the end of this phase, the data split, imputation and normalization as well as resampling brought us the data we needed to enter the second phase.
Phase 2: Choosing between supervised and semi-supervised learning
In building the ML algorithm, we had to choose between supervised and semi-supervised learning. The difference is that supervised machine learning algorithms learn only from labeled training data while semi-supervised machine learning algorithms learn from a small amount of labeled and a large amount of unlabeled training data.
“We decided to test both options in order to realistically assess which kind of ML algorithm would work best for our project”, explains Jonas from Optimate.
The Optimate Team: Max Hesselbarth, Martina Trinczek, Jonas Steiling
Jannik from Motius illustrates: “From a tech perspective, this process called “hyperparameter tuning” aims to optimize those parameters that control the learning process of the ML algorithm. In other words, it tunes the model so that it can optimally solve the given problem.”
In this project, the semi-supervised machine learning delivered slightly better results. “Based on the results, it was obvious that Machine Learning can help to predict whether a sheet metal part can be optimized. As soon as we achieved this milestone, we went on to develop the web demonstrator as the final step to the Proof of Concept”, says Max from Optimate.
Phase 3: Building the web demonstrator to bring the idea to life
As said, 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 a PoC.
The result was a web demonstrator that represents the whole desired optimization process from the beginning. It analyzes uploaded files through machine learning and visualizes the metal part’s optimization potentials through a user-friendly traffic light system.
With the web demonstrator, you simply upload your file first.
And then it automatically tells you whether your sheet metal parts are optimizable.
Securing further investments and preparing for the future
“Together with Motius, we successfully proved that our idea works and that it has the potential to optimize the manufacturing process. The web demonstrator perfectly visualized our work and the Proof of Concept that we achieved”, describes the Optimate team.
“As soon as it was clear that Optimate’s team would get the chance to keep developing their idea in their own spin-off company, we started thinking about future improvements and what we need to develop in order to create a profitable business model with them”, says Jannik from Motius.
Based on the successful collaboration in the scope of this project, Optimate and Motius decided to keep working together to further optimize the tool and establish Optimate at the market. The fact that Optimate secured further investments despite current constraints due to the pandemic underlines this potential.
With that outlook, this project is more than just a Proof of Concept. Instead, it is the basis for a long-term collaboration and a business that sets out to shape the industrial manufacturing industry through AI-based optimization technology.
If you want to more about the vast potential of Machine Learning, just talk to our experts below.