Advanced Methods for the Analysis of Semiconductor Manufacturing Process Data - Advanced Techniques in Knowledge Discovery and Data Mining - page 35

Fig. 2.5. Automatic transport system loading up a fully automatic wafer storage
(Stocker).
analyze a process failure may already have been deleted before the anomaly
is recognized and triggers a detailed investigation.
Collecting and evaluating all data for even a single lot are a formidable
tasks. The resulting very large multivariate data set must therefore be anal-
ysed for deviations from allowed parameters or operation ranges, i.e., anomaly
or novelty detection, and nonobvious multivariate dependencies of the in-
volved parameters and the structure in the data must be disclosed for im-
proved process control. Here, appropriate methods, e.g., from soft computing,
for online observation and o ine interactive analysis employing novelty clas-