In this paper, we propose a model-based approach to estimating production parameters of semiconductor FAB
equipment. For FAB scheduling, for example, we need to know equipment’s production parameters such as flow
time, tact time, setup time, and down time. However, these data are not available, and they have to be estimated
from material move data such as loading times and unloading times that are automatically collected in modern
automated semiconductor FAB. The proposed estimation method may be regarded as a Bayes estimation method
because we use additional information about the production parameters. Namely, it is assumed that the technical
ranges of production parameters are known. The proposed estimation method has been applied to a LCD FAB,
and found to be valid and useful.