|
| The first step is to identify those ET parameters which demonstrate the phenomenon of bimodality. To perform this step you have to do the following:
The phenomenon of bimodality of ET parameters can be presented graphically.
|
|
| The second step is to extract the hidden "good" and "bad" distribution patterns of the ET parameter and check their quality.
|
|
|
The third step is near real-time analysis of yield. This analysis consists of comparing the currently measured ET parameters with previously created "good" and "bad" patterns. The result of these comparisons is two distributions of P-values One distribution of P-value will indicate that theET parameter belongs to the "bad" pattern, the other that the distribution of P-values indicate that the ET parameter belongs to the "good" pattern.
|