With increased computational capabilities, the importance of artificial intelligence (AI) and machine learning (ML) in semiconductor manufacturing is increasing.
While there are already many reported use cases of how AI can be applied in areas such as wafer scheduling and dispatching, unit process optimization and defect classification, the application of AI looks especially promising in the field of yield data analytics.
The first part of this presentation will be a review of currently available AI use cases in yield analysis, ranging from automatic wafer test pattern detection to tool-combination best-path analytics and supervised and unsupervised ML for pattern classification. Pending permission from our customers, we will show these use cases with actual recent fab data.
We will finish the presentation with our technology roadmap for the application of AI for comprehensive semiconductor analytics to the end of the decade.
