48‐4: A Study on Reducing Transistor Electrical Characteristic Inspection Processing Time Using Machine Learning

With growing competition in display manufacturing, ensuring the quality of TFT components is crucial. This study uses machine learning to reduce the Tact Time of TFT electrical inspections while maintaining accuracy. By analyzing data from 47,160 glass substrates, clustering and predictive modeling techniques were applied. These methods improved inspection efficiency, enhanced production yield, and reduced costs, providing a scalable solution for modern manufacturing.