A neural-network approach to recognize defect spatial pattern in semiconductor fabrication

Yield enhancement in semiconductor fabrication is important. Even though IC yield loss may be attributed to many problems, the existence of defects on the wafer is one of the main causes. When the defects on the wafer form sp atial patterns, it is usually a clue for the identification of equipment pr oblems or process variations. This research intends to develop an intellige nt system, which will recognize defect spatial patterns to aid in the diagn osis of failure causes. The neural-network architecture named adaptive reso nance theory network 1 (ART1) was adopted for this purpose, Actual data obt ained from a semiconductor manufacturing company in Taiwan were used in exp eriments with the proposed system. Comparison between ART1 and another unsu pervised neural network, self-organizing map (SOM), was also conducted. The results show that ART1 architecture can recognize the similar defect spati al patterns more easily and correctly.