AI Application in Semiconductor Manufacturing: A Patent-driven Approach

As semiconductor structures continue to shrink, semiconductor manufacturers are facing increasing challenges, such as yield degradation, process complexity, and rising production costs. While artificial intelligence (AI) is expected to improve profitability, limited research has systematically investigated how AI is actually being used in this context. Therefore, this study explores the current application of AI technologies in semiconductor manufacturing using patent data and identifies promising application areas. A technology-function matrix analysis reveals that AI is primarily applied to process control and defect detection through image analysis. Additionally, a generalized linear mixed model (GLMM) analysis shows that technologies related to error correction, scheduling, and advanced metrology have recently demonstrated high growth rates. These findings offer practical implications for managers and practitioners seeking to leverage AI in semiconductor manufacturing.