Jg. Ramirez et B. Cantell, AN ANALYSIS OF A SEMICONDUCTOR EXPERIMENT USING YIELD AND SPATIAL INFORMATION, Quality and reliability engineering international, 13(1), 1997, pp. 35-46
Semiconductor yield data is binary by nature since an integrated circu
it (IC), or die, can only pass or fail a particular test. Still, many
people rely on simple linear regression to analyse this type of data,
which can produce incorrect results, e.g. negative yield predictions.
We discuss the use of logistic regression for the analysis of yield da
ta, and show that even this approach has to be modified to take into a
ccount the phenomenon of overdispersion. We present several approaches
to accommodate overdispersion. We also discuss a statistic for measur
ing the spatial dependence of ICs within a wafer, the spatial log-odds
ratio, which provides additional information to complement a yield an
alysis. These ideas are demonstrated using an example from our manufac
turing area. (C) 1997 by John Wiley & Sons, Ltd.