Digital transformation of wafer fabrication process: a case of semiconductor manufacturing

Semiconductors are essential for the digital revolution, modern electronics, and global economic growth. However, the growing demand, geopolitical tensions, and supply chain disruptions have accelerated the adoption of artificial intelligence (AI) and digital technology (DT) in semiconductor manufacturing. To address this gap, we studied five use cases from leading semiconductor manufacturers (Intel, Micron, Sony, Samsung, and SK Hynix). We investigate the application of AI & DT in the semiconductor wafer fabrication process (WFP). We propose a three-layer framework for WFP, consisting of supply chain, manufacturing process, and AI & DT layers, with enabling technologies. The framework incorporates AI & DT into manufacturing process-specific applications, including pattern classification, smart sensors & automation, real-time monitoring, yield prediction, automated material handling systems, predictive maintenance, data integrity, and fault & anomaly detection in semiconductor manufacturing. The implementation of AI & DT in the semiconductor industry has demonstrated significant improvements, including a 35% reduction in quality events (defects) and 25-40% reductions in cycle time. The proposed framework provides a strategy to improve operational efficiency at WFP while overcoming the limitations of the traditional approach. Key research findings highlight the potential impact of leveraging AI & DT in semiconductor WFP.