Spatial Pattern Recognition with Adjacency-Clustering: Improved Diagnostics for Semiconductor Wafer Bin Maps
In semiconductor manufacturing, statistical quality control hinges on an effective analysis of wafer bin maps, wherein a key challenge is to detect how defective chips tend to spatially cluster on a wafer–a problem known as spatial pattern recognition. Detecting defect patterns on a wafer can deliver key diagnostics about the root causes of defects and assist production engineers in mitigating future failures. Recently, there has been a growing interest in mixed-type spatial pattern recognition–when multiple defect patterns, of different shapes, co-exist on the same wafer. Mixed-type spatial pattern recognition entails two central tasks: (1) spatial filtering, to distinguish systematic patterns from random noises; and (2) spatial clustering, to group the filtered patterns into distinct defect types. Observing that spatial filtering is instrumental to high-quality pattern recognition, we propose to use a graph-theoretic method called adjacency-clustering, which leverages spatial dependence among adjacent defective chips to effectively filter the raw wafer bin maps. Tested on real-world data and compared against a state-of-the-art approach, our proposed method achieves at least 49 terms of internal cluster validation quality (i.e., validation without external class labels), and about 6 external cluster validation metric based on external class labels. Interestingly, the margin of improvement appears to be a function of the defect pattern complexity, with larger gains achieved for more complex-shaped patterns. This superior performance is a testament to the method's promising impact to semiconductor manufacturing, as well as other contexts where mixed-type spatial patterns are prevalent.
READ FULL TEXT