Manufacturing intelligence and smart production for industry 3.5 and empirical study of decision-based virtual metrology for controlling overlay errors

Focusing on the challenges of wafer fabs with multiple tools and high-mix products, this study aims to propose a decision-based virtual metrology framework to reduce the metrology cost and enhance productivity. An empirical study was conducted in a leading semiconductor company in Taiwan to validate the effectiveness of proposed approach for controlling overlay errors in lithography processes while ensuring the quality level. The results have shown practical viability of the proposed approach for reducing sampling rate and metrology cost.