I have been through many engineering changes in my life and many had to do with stupid tolerancing during design often related to tolerance stack-up. I'm not familiar with Statistical Tolerancing but look forward to any and all replies. Good topic.
A Search is a terrible thing to waste! One Test is Worth 1000 Expert Opinions - The plural of anecdote is not data - Correlation does not imply Causation
We can't solve problems by using the same kind of thinking we used when we created them. - Unknown
Marc, I second that statement. I look forward to lots of discussion on statistcal tolerancing.
IMO statistical tolerancing is one of the first steps in the design phase. Without it a six sigma process will fail.
One using monte carlo simulation and predicting the resultant tolerance stack ups (typically of assemblies) based on individual (component) tolerances.
The second category was the tolerance design based on Design of Experiments and analysis of variance... deliberate change in component dimensions to study their sensitivity on the stacked up tolerance.
The more I read, the more I think appropriate tolerancing is one of the major keys to Quality improvement and cost reduction.
We call the topic "Realistic Tolerancing" and it forms part of our six sigma projects.
We determine the input tolerance (x) from the output tolerance (y) by using regression analysis and it scares the **** out of designers and their dreams.
If you have a good correlation between x and y its easy if you have statistical software like Minitab.
(I also remember covering the subject at college many years ago, but if I recall the formula was really basic and based upon Cp values. I'll try to find my old notes and blow the dust off them can't promise though).