Reliable and value-adding AI implementations in an industrial context currently require a holistic approach: It includes hardware, IT infrastructure, a machine learning (ML) model and industrial-grade operations of data- and software pipelines. 

ML models typically rely on static historic extracts of generally dynamic data for training. However, data and concepts may change rapidly and even in a hidden manner, especially in industrial environments. 

In this webinar we will outline the realization of robust industrial AI relevant to the Semiconductor industry: 

• Leveraging distributed compute, e.g., using Industrial Edge installations we will outline how continuous learning and optimization with ML and mathematical optimization methods allows to reduce manual effort and increases reliability and robustness in continuous productive operation. 
• As a result, we are able to demonstrate efficiency impact by reducing power and resource consumption or increase yields in Integrated Circuits (IC) Manufacturing as well as Wafer Manufacturing plants.