Modeling and Multivariate Methods - SAS

600 Visualizing, Optimizing, and Simulating Response Surfaces Chapter 24
The Simulator
Figure 24.50 Results
If unit price can’t be raised anymore, you should now investigate lowering your cost, or increasing sales, if
you want to further decrease the probability of being unprofitable.
The Defect Profiler
The defect profiler shows the probability of an out-of-spec output defect as a function of each factor, while
the other factors vary randomly. This is used to help visualize which factor’s distributional changes the
process is most sensitive to, in the quest to improve quality and decrease cost.
Specification limits define what is a defect, and random factors provide the variation to produce defects in
the simulation. Both need to be present for a Defect Profile to be meaningful.
At least one of the Factors must be declared Random for a defect simulation to be meaningful, otherwise
the simulation outputs would be constant. These are specified though the simulator Factor specifications.
Important: If you need to estimate very small defect rates, use Normal weighted instead of just Normal.
This allows defect rates of just a few parts per million to be estimated well with only a few thousand
simulation runs.
About Tolerance Design
Tolerance Design is the investigation of how defect rates on the outputs can be controlled by controlling
variability in the input factors.
The input factors have variation. Specification limits are used to tell the supplier of the input what range of
values are acceptable. These input factors then go into a process producing outputs, and the customer of the
outputs then judges if these outputs are within an acceptable range.
Sometimes, a Tolerance Design study shows that spec limits on input are unnecessarily tight, and loosening
these limits results in cheaper product without a meaningful sacrifice in quality. In these cases, Tolerance
Design can save money.
In other cases, a Tolerance Design study may find that either tighter limits or different targets result in
higher quality. In all cases, it is valuable to learn which inputs the defect rate in the outputs are most
sensitive to.
This graph shows the defect rate as a function of each factor as if it were a constant, but all the other factors
varied according to their random specification. If there are multiple outputs with Spec Limits, then there is
a defect rate curve color-coded for each output. A black curve shows the overall defect rate—this curve is
above all the colored curves.