Comprehensive Selection of DOE Design Types
Traditional DOE Design Types
DOE++ makes it easy to build, modify and evaluate any of the following design types. In Version 10, we've added the ability ignore specific results for the analysis without removing them from the data sheet and now you can also enter up to 50 measurements for every test run!
One Factor Designs (also known as one-way ANOVAs) for determining whether a particular factor has an effect on a specified output or response.
Factorial Designs for determining which factors have a significant effect on the output of the response and identifying interactions between factors.
- Full Factorial Designs test all possible combinations
of factors at each level. Available full factorial designs include:
- Two Level Full Factorial
- General Full Factorial
- Fractional Factorial Designs test a subset of the possible
combinations of factors at the levels in question. Available fractional
factorial designs include:
- Two Level Fractional Factorial
- Taguchi Orthogonal Array
Response Surface Method Designs for studying the quadratic effects of the factors, which makes these designs well-suited to predictive modeling and optimization. Available RSM designs include Central Composite and Box-Behnken. For central composite designs in Version 10, you can now specify a range and alpha value to ensure that the factor always stays within specified limits.
Robust Parameter Designs that aim to minimize the variability of the response in spite of noise factors by combining an inner array of control factors with an outer array of noise factors.
Mixture Designs Folio
New in Version 10, the Mixture Designs folio is used when the factors in the experiment are proportions of components in a mixture (e.g., the proportions of iron, copper, nickel and chromium in a stainless steel alloy). Mixture designs can also be used to investigate the effect of process factors (e.g., the temperature, pressure and curing time). This allows you to optimize both the proportions of components and the way the mixture is processed.
Reliability DOE Provides the Proper Treatment for Life Data — Only in DOE++!
Available only in DOE++, reliability DOE (R-DOE) is a special category of DOE in which traditional experiment design types are combined with reliability analysis methods to investigate the effects of different factors on the life of a product.
Instead of using the F ratio (as in traditional DOE techniques), R-DOE uses the likelihood ratio, which provides the appropriate treatment for interval and right censored data and supports analysis with skewed distributions (e.g., Weibull, lognormal and exponential) — no adjustments or transformations required!
In the Synthesis version:
- All of the design types provided for traditional DOE analysis are available for R-DOE as well (including general full factorial, Taguchi OA and response surface methodology designs).
- You can create multi-response designs that use R-DOE for some responses and traditional DOE for others.
- Now in Version 10, you can manually set the shape parameter of the life distribution. This can reduce the uncertainty in your calculations by basing the regression model on a trusted historical estimate.