Page 2 of 3
(continued...) These dimensions can then be toleranced or redefined as more appropriate geometric tolerances.
The current version only supports bilateral and symmetric tolerances so any other variants need to be redefined to suit.
At this point the tolerance scheme can be checked for any inconsistencies before initiating an analysis.
Both worst case and statistical analyses can be run.
The worst case analysis assumes that all variations occur in the worst possible way and finds the combination of defined tolerances that result in the response function being at the upper and lower limits.
Since the worst case is almost always too conservative giving results that are very pessimistic, statistical analysis can be used to take into account the low probability of a worst-case combination occurring.
MonteCarlo simulation, the Quadrature Technique and Linear root sum square stack-up methods can be selected to balance accuracy against computing time.
The MonteCarlo method can be particularly time consuming as a large number of iterations are required to give high confidence of an accurate result.
The results are saved out to disk and can be reviewed using the post-processor window.
By selecting the available response function and analysis from scroll down lists, options such as nominal, upper and lower conditions can be selected by radio button.