Jm. Hughesoliver et al., ACHIEVING UNIFORMITY IN A SEMICONDUCTOR FABRICATION PROCESS USING SPATIAL MODELING, Journal of the American Statistical Association, 93(441), 1998, pp. 36-45
Material is deposited onto the wafer surface during several steps of w
afer fabrication. This material must be deposited evenly across the en
tire wafer surface, close to the targeted thickness, and with little w
afer-to-wafer variability. But unequal variances across the wafer and
under different process conditions, as well as nonstationary correlati
on across a wafer, make these goals difficult to achieve, because trad
itional methods for optimizing deposition processes assume homogeneity
and independence. We avoid these assumptions and determine the best s
ettings of process variables using physically motivated statistical mo
dels for the mean response. unequal variances, and nonstationary spati
al correlation structure. Data from a rapid thermal chemical vapor dep
osition process is used to illustrate the approach. A simulation exerc
ise demonstrates the advantages of fitting flexible variance models an
d using appropriate performance measures.