摘要 |
A method for optimizing multiple process windows in a semiconductor manufacturing process is disclosed. The method comprises performing dependent variable composition on a plurality of dependent variables. Metrology data is joined with the dependent variables, and then a partial least squares regression is performed on the joined data set to obtain a prediction equation, and a variable importance prediction for each process window in a process window set. A set of product limited yield are derived, and the process window, set is adjusted, and the yields recalculated, until an optimal process window set is derived.
|