Fault coverage and yield predictions: do we need more than 100% coverage?
Leendert M. Huisman · 2002
The influence of non-random defect distributions and of non-modelled faults on the behavior of the cumulative chip fallout as a function of the cumulative fault coverage is discussed. The theory is applied to published yield data and a novel explanation of the deviation of these data from the classical Williams-Brown theory is presented.>