Engineering Seminar Series 5/2_2010: Fast and Accurate Automatic Defect Cluster Segmentation and Detection for Semiconductor Wafers using Data Mining

Ms Melanie Ooi Po-Leen

Date: 2010-04-09
Time: 12:30 to 13:00
Venue: 9-3-03


Reduction in integrated circuit (IC) half-pitch dimensions pushes the physical limits of semiconductor test technology, which will no longer be sustainable by traditional fault isolation and failure analysis techniques. There is an urgent need for diagnostic software tools with respect to fault isolation. Because systematic failures (which manifest as clusters) observed from manufacturing defects can be traced back to a specific process, equipment or technology, a novel data mining algorithm is proposed to extract cluster information on these defects from test data logs. This algorithm executes at an average speed of 3secs/wafer on a normal processor and provides accurate detection of 99%.

About the Speaker

Ms Melanie Ooi received both her BE(Hons) & MEngSc (Research) from Monash University Sunway campus in 2003 & 2006 respectively. She is currently lecturing & pursuing her PhD study in School of Engineering. Melanie’s research & professional interest include semiconductor manufacturing and test technology Image Processing.