Creating a Data Ecosystem to Enable End-to-End Advanced Predictive Manufacturing—From Material Supplier to Memory Device Maker

The semiconductor industry has seen growing challenges with material interactions to fab processes impacting yield ramps for next-generation technology nodes with narrower process margin coupled with more complex material requirements. Today’s conventional way of supplier management or control is not comprehensive enough to meet the stringent and ever-increasing quality expectations as demanded by our customers.

In order to truly SHIFT-LEFT, a joint project was initiated between material supplier (Electronics business of Merck, KGaA, Darmstadt, Germany) and memory device maker (Micron). Through this collaborative partnership driven by Senior Leaderships at both companies, an AI-Driven Quality Approach was developed by smart data collaboration to drive the process, yield, and quality improvements. In this presentation, we will share technical approaches and real production improvement implementations carried out by the joint teams for Chemical Mechanical Polishing (CMP) Module. We believe that by extending AI-Driven Quality Approach to the broader supply chain and data ecosystem, end-to-end Advanced Predictive Manufacturing could be accelerated for the broader semiconductor industry.