Industrial AI at Thyssenkrupp Steel: From idea to scalable solution – what really works
summary
Project with EMG at Thyssenkrupp Steel builds a live industrial AI that tunes hot‑dip galvanizing settings (air‑knife pressure, strip auto‑positioning, coating margin) using sensor streams and a data platform, trained on ~2 years of production data. The system ensembles 108 models into a fitness function that makes decisions under 100 milliseconds, replaces gut tuning, scales across machines, reduces scrap and CO2, and delivers roughly €6,700 savings per month while enabling EMG to productize the models as a digital offering.
AI-written summary · about
chapters (5)
- 0:00Introduction
- 3:40AI adoption challenges
- 6:50Industrial AI in steel
- 10:02AI for galvanizing
- 13:39Project lessons
related talks
31mHow to successfully implement AI in manufacturing
16mKey Points and Lessons Learned in AI Projects in Mechanical Engineering
16mEnterprise and industrial AI: The great threat of complacency
10mAI-Powered Digital Twins: Patterns for Smart Manufacturing
20mDeploying AI in Days, Not Months: Siemens Industrial Edge on
11mDemystifying AI in Smart Manufacturing
22mThe AI Frontier: Redefining Industrial Automation
23mAI Here & Now