Identifying wafer fabrication defect signatures

We introduce a system called IDS for diagnosing semiconductor fabrication process. IDS can identify the carry-over defect that contributed to a defective cluster captured by an in-line inspection tool during a specific fabrication processing step. Let R
_x
be a defective cluster captured by an in-line inspection tool during the fabrication processing step ps
_i
. First, IDS would identify the defect type of R
_x
by comparing the image of R
_x
with the images in reference wafer maps of previous defect clusters captured at psi. Let dj be the identified defect type. IDS would use rules of inferences for identifying the root cause of dj. It will utilize d
_j
as a trigger for the appropriate defect specification rules using the standard rules of inferences. By applying the defect specification rules recursively using the inference rules, IDS can identify the root cause of d
_j
. We experimentally evaluated the quality of IDS. Results revealed marked prediction accuracy.