SEMI is supporting a proof-of-concept project at Cornell University to explore innovative applications of Data and AI to semiconductor manufacturing and the supply-chain ecosystem. The project also provides members with early access to leading talent with cross-disciplinary training in both Data-AI and microelectronics.
For details on how to participate, please contact Pushkar Apte at [email protected]
POC Project Overview:
Goal set to build an accurate AI model that can optimize 2 process modules (lithography and etch).
The project has already completed 2 phases where the team demonstrated an accurate AI model, and methodologies to overcome data-sharing barrier - a major roadblock for AI implementation in semiconductors.
POC Project Results:
From Phases 1 &2 – Accurate AI model was developed that predicts the device dimensions accurately based on process parameters.
POC Project Plans:
Phase 3 (starting soon) –
Goal is to create a virtual innovation environment with digital twins and secure data sharing.
SEMI102 ARTIFICIAL INTELLIGENCE FOR THIN FILM MANUFACTURING
This course describes how machine learning and AI-based approaches to research, development, and production bring advantages to cleanroom processes. The team from Cornell has applied AI approaches to optimize lithography and etching processes involved in the development of an RF wake-up NEMS (Nano ElectroMechnical system) switch that needs a well-controlled gap between a moving shuttle and a contact.
