Products
OverviewIndustries
Overview- Agriculture and Forestry
- Automotive
- Battery production
- Chemical processing
- Construction and Infrastructure
- Consumer goods
- Energy and Utilities
- Healthcare
- Hydrogen
- Logistics and Transport
- Materials and Resources
- Pharmaceutics
- Recycling and Waste handling
- Renewable energies
- Semiconductors and Electronics
- Testing Technology
Service
OverviewBuy
OverviewCompany
OverviewProduct groups
OverviewEngineering
OvervieweConfigurators and Tools
OverviewEDI (Electronic Data Interchange)
Overview
Switzerland
English
- Home
- References
- Tech-Podcast: Domain knowledge in the industry
Company
Tech Podcast Domain knowledge in the industry
Matthias Lück is a research associate at the Fraunhofer Institute for Industrial Engineering IAO. In the podcast conversation, he explains the importance of domain knowledge in industry. We summarise his seven theses.
- Domain knowledge means: Man knows the way. Every person can name the way from his bedroom to the refrigerator with his eyes closed. That is domain knowledge. I know my way around my domain, my house, and I know where to go. Transferred to industry: There, people have to share their knowledge with other colleagues, with a network of experts.
- Domain knowledge in industry is not whistling in the wind. European companies have a "brute" strength when it comes to knowledge about the processes, the products or the machines. The challenge is to bring the knowledge together with models for machine learning.
- Domain knowledge is important because it helps to develop models, but above all to make them explainable. The acceptance of new techniques such as machine learning stands and falls with whether the human can explain why the machine decided this way or that. Domain knowledge in models prevents uncertainty on the part of the operator.
- The bosses have to talk to the people at the machine. They have to ask questions, they have to have processes explained to them, and ask why it clicks at a certain point or what the screw curves look like.
- The industry is already getting very far with statistical, mathematical methods. But the combination of domain knowledge and new technical approaches such as machine learning takes companies one step further. Domain knowledge gives us a head start over generic models in Europe. Companies need to take advantage of this.
- Tomorrow's production also lives from the creativity of the people in the processes. A model will not replace that. In image processing, creativity is not required, but the machine is superior to the human. In other tasks, humans will continue to set the tone with their knowledge.
- We need to give domain experts a way to continue sharing their knowledge after they leave the workforce - perhaps as a consultant, a trainer who stops by once a week. Not every employee wants that, we have to accept that. But companies that identify motivated employees before they retire and then work with them to prepare knowledge have an advantage.
You can find this and other episodes on the factory of the future in our tech podcast channel "Industry rethought" on all known platforms or you can subscribe directly here via Podigee (only available in German).
Contact person for the Bosch Rexroth Tech Podcast: Susanne Noll
Please feel free to contact Bosch Rexroth!