Lesson Learnt and Future of AI Applied to Manufacturing
- Frascolla, Valerio1
- Hummert, Matthias2
- Monsees, Tobias2
- Wübben, Dirk2
- Dekorsy, Armin2
- Michailow, Nicola3
- Döricht, Volkmar3
- Niedermeier, Christoph3
- Kaiser, Joachim3
- Bröring, Arne3
- Villnow, Michael3
- Wessel, Daniel4
- Geiser, Florian4
- Wissel, Matthias4
- Viseras, Alberto4
- Han, Bin5
- Richerzhagen, Björn3
- Schotten, Hans5
- Calandra, Davide6
- Lamberti, Fabrizio6
- 1. Intel Deutschland GmbH
- 2. University of Bremen
- 3. Siemens AG
- 4. Motius GmbH
- 5. TU Kaiserslautern
- 6. Politecnico di Torino
Description
This chapter touches on several aspects related to the role of Artificial Intelligence (AI) and Machine Learning (ML) in the manufacturing sector, and is split in different sub-chapters, focusing on specific new technology enablers that have the potential of solving or minimizing known issues in the manufacturing and, more in general, in the Industrial Internet of Things (IIoT) domain. After introducing AI/ML as a technology enabler for the IoT in general and for manufacturing in particular, the next four sections detail two key technology enablers (EdgeML and federated learning scenarios, challenges and tools), one most important area of the IoT system that needs to decrease energy consumption and increase reliability (reduce receiver Processing complexity and enhancing reliability through multi-connectivity uplink connections), and finally a glimpse at the future describing a promising new technology (Embodied AI), its link with millimetre waves connectivity and potential business impact.