Design Expert

Welcome to Design-Expert
Page 1 of 19
Welcome to Design-Expert software

Stat-Ease, Inc. welcomes you to version 9 (v9) of Design-Expert software (DX9) for design of experiments
(DOE). Use this Windows-based program to optimize your product or process. It provides many
powerful statistical tools, such as:
x Two-level factorial screening designs: Identify the vital factors that affect your
process or product so you can make breakthrough improvements.
x Multilevel Categoric (general factorial) studies: Discover the best combination of
categorical factors, such as source versus type of raw material supply.
x Response surface methods (RSM): Find the optimal process settings to achieve peak
performance.
x
Mixture design techniques: Discover the ideal recipe for your product formulation.
x Combinations of process factors, mixture components, and categorical factors: Mix

your cake (with different ingredients) and bake it too!
Your Design-Expert program offers rotatable 3D plots to easily view response surfaces from all
angles. Use your mouse to set flags and explore the contours on interactive 2D graphs. Our
numerical optimization function finds maximum desirability for dozens of responses simultaneously!
Youll find a wealth of statistical details within the program itself via various Help screens. Take
advantage of this information gold-mine that is literally at your fingertips. Also, do not overlook the
helpful annotations provided on all reports.
For a helpful collection of checklists and cheat sheets, see the Handbook for Experimenters. Its
free to all registered users. Furthermore, for quick primers on the principles of design and analysis,
we recommend you read the following two soft-cover books from Stat-Ease Principals Mark
Anderson and Pat Whitcomb published by Productivity Press of New York city:
x
x
DOE Simplified: Practical Tools for Effective Experimentation,

RSM Simplified: Optimizing Processes Using Response Surface Methods for Design of
Experiments.
Anderson and Whitcomb have also written a Primer on Mixture Design. Its posted free for all to
read via the Im a Formulator link on the Stat-Ease home page.
Go to http://www.statease.com/prodbook.html for details and ordering information on the books listed
above.
Whats New
Those of you whove used previous versions of Design-Expert software will be impressed with the
many improvements in Version 9. Here are the highlights:
Hard-to-change factors handled via split plots
Two-level, multilevel and optimal factorial split-plot designs: Make it far easier as a practical matter
to experiment when some factors cannot be easily randomized.
file:///C:/Users/p00011956/AppData/Local/Temp/~hhE5E5.htm
28-07-2016
Page 2 of 19
Half-normal selection of effects from split-plot experiments with test matrices that are balanced and
orthogonal: The vital effects, both whole-plot (created for the hard-to-change factors) and sub-plot
(factors that can be run in random order), become apparent at a glance!
Effects from split plots assessed via REML* and Kenward-Rogers approximate F test: See the
familiar p-values that tell you whats statistically significant.
Design resolution provided for two-level factorial split plots: Assess from the start whether your
choice suffices for screening main effect(Res IV) or characterizing interactions (Res V).
Power calculated for split plots versus the alternative of complete randomization: See how
accommodation of hard-to-change factors degrades the ability to detect certain effects.
Check designs with restricted randomization for REML/OLS* equivalence: Keep things simple
statistically (KISS) in the ANOVA. (Ordinary least squares)
Other new design capabilities
Definitive screening designs: If you want to cull out the vital few from many numeric process factors,
this fractional three-level DOE choice resolves main effects clear of any two-factor interactions and
squared terms (see screen shot of correlation matrixmore on that later).
On the Factorial tab select Simple Sample to generate descriptive statistics based on a mean-model
only: Take advantage of powerful features in Design-Expert software for data characterization,
diagnostics and graphicsfor example with raw outputs from a process being run at steady-state.
Much-improved capabilities to confirm or verify model predictions
New Post Analysis Node (at bottom of the handy tree structuring of Design, Analysis and
Optimization) contains Point Prediction, Confirmation and Coefficients Table reports: Old and new
28-07-2016
Page 3 of 19
features gathered in a logical place at the end of the natural progression from design through
analysis.
Entry fields for confirmation data and calculation of mean results: Makes it really easy to see if
follow-up runs fall within the sample-size-adjusted prediction intervals.
Enter verification runs embedded within blocks as controls or appended to your completed design:
Lend veracity to your ultimate model by these internal checks.
Verification points displayed on model graphs and raw residual diagnostics: See how closely these
agree to whats predicted by your model.
New and more-informative graphics

LOESS* fit line for Graph Columns: Draw a curve through a non-linear set of points as you see fit.
*(Locally weighted scatterplot smoothing.)
28-07-2016
Page 4 of 19
Color-coded correlation grid for graph columns: Identify at a glance any factors that are not
controlled independently of each other, that is, orthogonally; also useful for seeing how one response
correlates to another.*
*(Data shown in screen shot comes from historical data detailed in RSM Simplified on NFL sacks versus attributes of defensive
linemen.)
Jump to run added to Factors Tool for model graphs: For multidimensional experimental regions,
find the slice of interest (containing the point you want to see) at the press of a button.
When jumping to a run, the range expands to include the design point: Use this feature to check how
well your model fitscomparing the actual result versus what is predicted via the surface graph (in
28-07-2016
Page 5 of 19
this case very wellthe circled red point is barely beneath the surface!).
Ignored (and missing) runs can be shown on graphs: Good to be reminded that the original design
called for this, but for one reason or another, you ignored the outcome (or the response could not be
collected, or it was skipped).
Choice to do diagnostic graphs with externally-studentized residuals (now the default): This deletiondiagnostic (vs internally-studentized) provides a more sensitive view of potential abnormalities.
Many new icons, such as ones for Clear Points and Pop-Out View on the Diagnostics Tool: Jump to
features used frequently more quickly via these handy markers (also they look good!).
Three-component contour graphs in reals: Get a better view of the restrictions placed on your mixture
space by the constraints you enter on each component and the total.
28-07-2016
Page 6 of 19
Half-normal plot for one-factor categorical experiments with replicates: See at a glance if anything
significant emerges.
Greater flexibility in data display and export

Descending sort of all individual design layout columns via right-click menu (shown) or double-click
on header (toggles with ascending sortpreviously the only option): Helpful, for example, when
minimum response is desired.
Identify via Build Type the predetermined Model, Lack of Fit, Center, and Replicate points in your
design layout: Dissect the matrix laid out for optimal (I, D, etc) experiments.
Switch directly between continuous and discrete point type: Sometimes the settings for a factor
cannot be easily changed (for example, diameter of molded part)then it pays to recognize them as
28-07-2016
Page 7 of 19
discrete, thus enabling the numeric optimizer being set so it will not stray away from specific values.
Ignorable block and/or factor columns: Handy for what-if analysis, such as what would have
happened if you had not blocked your experiment.
Journal feature to export data directly to Microsoft Word or Powerpoint: Fast and formatted for you
to quickly generate a presentable report on your experimental results.
Improved copy/paste of Final Equation from the analysis of variance (ANOVA) report to Microsoft
Excel: This not only saves tedious transcription of coefficients but it also sets up a calculator for you
to plug and chug, that is, enter into the spreadsheet cells what values for the inputs youd like to
evaluate and see what the model predicts for your response.
28-07-2016
Page 8 of 19
From Evaluation and ANOVA screens, the X matrix can be viewed and exported: This is helpful, for
example, for copy and paste to R or Matlab where statisticians can do further manipulations for
research purposes.
Display full precision of F-test: If just presenting p<0.0001 is not precise enough, show all the
decimals.
New XML* script commands for exporting point predictions: Helpful for situations where one wants
to automate the transfer of vital outputs from Design-Expert to other programs.
*(Extensible Markup Language)
More powerful tools for modeling

Design model included in Fit Summary: This can be very helpful for combined designs such as
response surface optimizations that include categorical factors (in this case recommending a model
that included some cross-product terms of 3rd order, which provided a better fit of the data).
28-07-2016
Page 9 of 19
All-hierarchical model (AHM) selection: Sort through all possible models up to the one you
designed the experiment for, but all the while maintain hierarchy of terms so you do not end up with
something ill-formulated.
(PS. The alpha out is enforced after AHM is completed by doing a final sweep using backward selection, after which hierarchy is again
corrected by the program.)
Non-linear equations involving trigonometric, exponential and other functions allowed for creating
deterministic responses (for examplecosts) or simulations: This will be especially helpful for setting
up more realistic scenarios for students to solve during hands-on workshops for teaching DOE.
(PS. Simulator now provides an entry field for ratio of variance between whole and sub plots so trainers can set up split-plot exercises.)
Special quartic Scheff polynomial included in automatic selection for mixture modeling: Sometimes
this added degree (4th!) of non-linear blending helps to better shape the response surfacemaking it
better for predictive purposes.
28-07-2016
Page 10 of 19
More choices when custom-designing your experiment

Required model points set aside from optional additional ones that may be needed for adequate sizing
of the design: Prevents setting up an experiment with too few points to fit the chosen model.
Enter a single factor constraint for response surface designs: Creates a hard limit on inputs that
cannot go beyond a certain point (such as zero time) physically or operationally.
Greater flexibility in setting up models: For example you can now create an optimal model for
experiment on mixtures with varying categorical ingredients, some of which can go to zero.*
*(See presentation of Categoric Mixture Components Proportion Going to Zero by Pat Whitcomb, ENBIS-12,
Ljubljana, Slovenia. Slides available on request to stathelp@statease.com.)
Save candidate sets in actuals: More flexibility for customizing your experiment design.
More capability for numerical optimization
Include Cpk* as a goal: Meet quality goals explicitly.
*(A process capability index widely used for Six Sigma and Design
for Six Sigma programs.)
28-07-2016
Page 11 of 19
Enhanced design evaluation

Random model generator provided to generate a realistic response via a quadratic polynomial with
coefficients picked by chance: Use this to play around with how the software presents the analysis
better than just generating random numbers that only fit a mean model.
One-sided option added to FDS* graph: Size your design properly for a verification experiment done
to create a QbD** design space.
*(Fraction of design space)
**(Quality by Designa protocol promoted by the US Food & Drug Administration (FDA).)
Many things made nicer, easier, more configurable and faster

Components that do not vary in a mixture experiment can now be included in the design buildsee
them highlighted with gray in the layout: Provide a recipe sheet that encompasses the entire
formulation, not just what will be manipulated in your study.
Automatically re-sort by run order after re-randomizing: A little feature that saves users a bother.
28-07-2016
Page 12 of 19
Faster display of graphs: Great for dazzling your audience with 3D graphs in high resolution.
Pop-out views numbered: Makes it easier to distinguish and find the associated view-Tool when rearranging on your desktop.
Graph state stored with file: Restores setting to the way you liked them.
Improved graphics on Transformation screen: Looks more elegantbetter to show off your results via
live presentations or webinars.
Fonts on analysis tabs now configurable under Edit Preferences (Dialog Control): Go ahead and make
them Comic Sans if you would like to lighten things up. ; )
Safety net expandedmore mistakes caught and heads-ups given

Warning when largest effect not selected on half-normal plot: This would not make sense, but it might
happen due to, for example, not lassoing points correctly.
28-07-2016
Page 13 of 19
Hover Help added to select fields: When your mouse goes over an entry place, the program fills you
in with a bit more information on whats entailed in the feature you are specifying.
Niceties that only statisticians might truly appreciate

Mean correction for transformation bias when responses displayed in original scale: All you need to
know is that our statisticians figured out how to eliminate a tricky, little-known bias!
Propagation of error (POE) carried out to the second derivative: Makes POE more accuratethats a
good thing!
Allow averaging of categoric factors when viewing a graph: Convenient for getting the big picture of
where to find robust operating settings.
Display confidence bands with or without POE added: Easier to match output with other programs
that do not offer POE features like this.
Add unblocked results to evaluation of blocked experiments: Aids in comparing designs on the basis
of matrix measures.
28-07-2016
Page 14 of 19
Scale to largest estimable effect those normal effects that cannot be otherwise estimated: This can
happen when effects become too large compared to the error estimated by chi-square.
Technical stuff only those adept at programming will get
Automatically generate DTD* files: Now these will always be up to date.
*(Document Type Definition)
New command to export runs of a specific type: Particularly useful for verification points.
Hints and Frequently Asked Questions
Missing data
Leave it blank; a zero is treated as a numeric value.
Right-click
You can find many features in the program by right-clicking. Heres a list of where right-clicking will
pull up other options.
Design layout screen the column headers (gray squares labeled Run, Factor, etc) give access to
Editing, Sorting, Inserting, Deleting, etc.
Design layout screen the gray square to the left of each row gives access to Duplicating, Deleting,
Ignoring, etc.
Model editing screen (where you see "M" and "e" beside the list of factors) right-clicking brings
options to put a term into the model, or error.
Reports (ANOVA, Design Evaluation, etc) you can access the Journal feature, context-sensitive Help
and editing commands like Cut/Copy/Paste.
Graphs change graph options like X, Y axes, contour settings, colors, and insert flags. Export to
Word or Power Point.
Right-click exactly on a Contour line and you can set the exact level for that line.
Factors Tool set the axis for displaying a factor on plots.
Importing/Exporting Data
Data can be imported and exported by using the standard copy and paste operations from Windows.
Highlight the group of cells and select copy from the Edit menu or right-click in the highlighted area
28-07-2016
Page 15 of 19
to select "Copy" (Ctrl-C). Go to the program you want to paste to and select paste from the second
programs Edit menu (Ctrl-V).
If you are trying to import a new design from a spreadsheet, a design must be set up in Design-Expert
to receive it. Make sure that the factor levels are correctly specified. Duplicate or Delete rows in the
design layout screen to get the desired number of rows. Copy and paste the data from the other source
into the design layout over the blank spreadsheet.
Designs can be exported to tab delimited text files. These files can be edited to enter responses and
then imported by Design-Expert.
The steps for using this feature are
1) Create design in Design-Expert
2) Save the design as a *.dxpx file.
3) Export the design as a text file.
4) Open the text file using a third party text editing software
5) Enter the response data using the third party software
6) Save as a tab delimited text file from the third party software
7) Open the *.dxpx file saved above
8) Import the text file with the responses included.
This feature is most often used with automated experiment equipment.
The number, names and order of the rows and columns must exactly match the design or the import
will fail.
Analyzing historical data (data mining)
It is possible to analyze data that has been collected in other ways than by doing designed
experiments. We caution about the interpretation of the analyses because the factors are most likely
not independent from each other. But, if you want to do it, use the Historical Data Design on the
Response Surface tab and follow the instructions at the top of the opening screen. Before doing the
analysis, we recommend checking Design Evaluation to determine if there is any aliasing and how
much correlation there is between the factors.
Historical "designs" are available for response surface designs and mixture design, but not combined
designs or factorial designs. However, any design with enough rows can have data pasted over the
designs runs. Use the evaluation node to see how good of a design the dataset happens to be.
Display problems - lines breaking on reports
If the text on a report is breaking up, use the "maximize" icon on the top right corner of the screen.
There may be problems if the resolution of the monitor is set at 640 x 480 or if Windows is set up to
use "large" fonts and the monitor resolution is 800 x 600. There are two solutions:
Change to a higher resolution monitor setting.
Change the Windows display to "small" fonts.
Saving model and analysis
The analysis cannot be saved independent of the data. A way to examine multiple models is to insert a
new response column and copy/paste the response data into the new column. Then with two (or more)
28-07-2016
Page 16 of 19
columns of the same response, you can create a different model for each column and then be able to
compare them.
Format of data - decimal places
On the design layout screen, right-click on the column header for the column you want to change and
choose Edit Info and change the format.
Changes to the design
You may have gone to the effort of typing in the factor names and responses and then decide you
want to change the design type, or change some of the options. First, you can change many things on
the design layout screen by right-clicking on the Factor column headers or row headers (to the left of
each row.) If this doesnt work, then simply click on File, New Design and answer NO to the "save
changes?" question and YES to the "Use previous design info?" question. This preserves all the factor
names and levels and the response names, allowing you to easily re-build without re-typing
everything. Another option for changing factor and response information is to use the View, Column
Info Sheet from the Design Layout screen.
Messed-up coding levels
This can happen if you enter actual values while displaying coded factors or edit the highs and lows in
the design layout rather than rebuild as noted above.
Follow these steps:
1. Display the data in the design layout in actual units.(Display Options ==> Process Factors, check
Actual)
2. Click on a factor header to highlight the data in the column.
3. Respond to the Warning message. Yes will change the data set to match what was just entered.
No will only change the underlying coded values leaving the current actual entries as is.
4. Repeat steps 2 to 3 for the remaining factors.
Unfortunately, this trick won't work with mixtures as you can't edit the pseudo ranges. You will need
to rebuild the design.
Can't run one combination of factor levels
If there is only one combination that is difficult or impossible to perform, you might just leave that
run out of the experiment and treat it as missing data. Another option is to build a candidate set
containing only the possible combinations and building an optimal design. Combine factors if a
particular combination of two factors cannot be part of the design. You lose the interactions but can
still test for differences (effects) between the combinations. Optimal designs can also have
constraints applied to them if the excluded combinations only involve numeric factors.
Too many runs on Multilevel Categoric designs
On the Factorial tab, choose Optimal. This will allow you to create a multi-level design that only
estimates the terms that you are interested in. You probably dont need to estimate 3-factor and
higher-order interactions as they are likely to be pooled into error. Leaving the high order interactions
out of the designed for model reduces the number of runs needed.
28-07-2016
Page 17 of 19
Different results than by-hand calculations

Differences between program results and hand calculations are normally due to a loss of significant
digits. These differences can be particularly pronounced when a transformation is being used or when
you are using the actual prediction equation. For readability, the coefficients on the ANOVA are, by
default, only displayed with two decimal places. You can increase the number of decimal places
displayed by choosing the "Force precision with X significant digits" option under Edit ==>
Preferences/General Tab. Select "Force precision" over-rides all local display formats. This can ruin
otherwise "nice" looking output. Turning "Force precision" off will restore the number formats to
local control.
Results from two-factor experiments
If only two factors are tested, the two main effects can be estimated and tested statistically. However,
if the interaction is also estimated, all the degrees of freedom are used and statistical tests cannot be
performed. Thus none of the three effects can be tested. We recommend replicating this experiment to
provide estimation of pure error. Then the interaction and error can both be estimated and all effects
tested statistically.
View alias structure for pre-existing design
Click on the Evaluation node on the left side of the screen. For a factorial design, most likely you will
want to set the Order to 2FI (the model includes all terms up to the two-factor interactions). This will
provide the aliases for the main effects and two-factor interactions.
No significant effects
When nothing is significant, try plotting the factors individually versus the response by using the
Graph Columns node on the left side of the screen. You may pick up on some trends that are
interesting, even if they are not statistically significant.
Green triangles on Half Normal Plot
These triangles represent unused degrees of freedom that are left over from replicates. When the
design has replicates, the pure error from them can be pooled with the natural process variation to
help guide your selection of effects on these plots. Watch out for the situation when the green
triangles are located off to the upper right on the Effects Plot (where significant effects should
normally be located). This means the error between replicates is the biggest effect. Put another way,
the estimated error is much larger than the effects caused by the factor changes.
No lack-of-fit info
Most likely, there are no exact replicates in the design. The Lack-of-fit test requires replicates, plus
more design points than the number of coefficients you are trying to estimate.
Transformation of Factor Levels
If it makes more sense to use transformed factor settings then enter the transformed units and limits as
the high and low settings. To reverse the transformation create an extra response for each
transformed factor. Right-click the top of the extra response(s) and select the simulate
option. Choose the Use equation in analysis type. Enter the inverse function for the transformed
variable in the provided box. Click the Help on the dialog box for more information about what
functions are available.
28-07-2016
Page 18 of 19
Low Predicted R-squared

Things to look for:
Factor effects are small relative to the experimental error
Outliers in the data
Too many insignificant model terms
If the number of coefficients in the model approach the number of design points the predictions can
become unstable
If there are high leverage points in the data set, the problem could be due to a combination of a
potential outlier and high influence check for large Cooks distance values.
Negative Adjusted R-squared
A negative Adj. R-squared is produced when the MSmodel is less than the MSresidual. This is not a
model one would ordinarily choose. A negative Adj. R-squared is a warning that the variation
explained by the model (MSmodel) is less than the residual variation (MSresidual). If the model
selected is statistically significant (MSmodel is greater than MSresidual) then the Adj. R-squared
cannot be negative.
Zero desirability in numerical optimization
If you have responses that completely conflict with each other, i.e., the best settings for one are the
worst settings for the other, then it is possible that no region satisfies your criteria. Optimization is a
trade-off: The criteria may need to be adjusted so that some acceptable operating region can be found.
28-07-2016
Page 19 of 19
New Designs and Name Changes in V9 of Design-Expert Software

There have been many improvements in Design-Expert (DX9) version 9, notably split-plot designs,
which accommodate hard-to-change factors by restricting their randomization into groups. These
new designs can be seen at the bottom of the factorial design builder in DX9 (see the box in the
screenshot, below right). Randomized designs remain available, but some feature new, more
descriptive names, and they have been resorted for easier access. As always in Stat-Ease software,
the most commonly used designs get top priority, that is, they are listed in order of usefulness.
For a quick overview of the changes, compare the screenshots below.
28-07-2016

Design Expert

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Design Expert

Uploaded by

Copyright:

Available Formats

Welcome to Design-Expert

Welcome to Design-Expert software

x Combinations of process factors, mixture components, and categorical factors: Mix

DOE Simplified: Practical Tools for Effective Experimentation,

New and more-informative graphics

Greater flexibility in data display and export

More powerful tools for modeling

More choices when custom-designing your experiment

for Six Sigma programs.)

Enhanced design evaluation

Many things made nicer, easier, more configurable and faster

Safety net expandedmore mistakes caught and heads-ups given

Niceties that only statisticians might truly appreciate

Different results than by-hand calculations

Low Predicted R-squared

New Designs and Name Changes in V9 of Design-Expert Software

You might also like