A cluster-modified Poisson model for estimating defect density and yield

A simple modification of the Poisson model that accounts for defect clustering when estimating the defect density and yield of a future product is described. The Poisson yield equation is easy to use and interpret, and the composite model can be extended to a layered model without the difficulties associated with, for example, the negative binomial model. The advantage of the method is that the Poisson model, to which all yield models reduce as yields increase, can continue to be used with less effort than, more convenience than, and at least as good accuracy as other more complicated models