Manufacturing Intelligence for Equipment Condition Monitoring in Semiconductor Manufacturing

For modern semiconductor manufacturing, a large number of interrelated equipment data are automatically collected. These data are usually used for fault detection and classification (FDC). However, unusual measurements may reflect a wafer defect or a change in equipment conditions. Early detection of the equipment condition changes assists with efficient maintenance. This study aimed to develop hierarchical indices that can be used for the conditional-based maintenance (CBM). For convenience, only the highest level index is used for real-time monitoring. Once this index decays, engineers could simply drill down to lower indices to identify the root cause. For validation, the proposed approach is conducted in a leading semiconductor foundry in Taiwan. The result shows that the highest index indicates the change of equipment conditions right after the preventive maintenance (PM).