最新產品‎ > ‎

Design-Expert

Design-Expert
Design-Expert
® V8 Software for Design of Experiments (DOE) 
http://www.statease.com/dx8descr.html

台灣代理商 採購,技術支援請來電 02-29299388 分機16 ,
來信service@orderble.com,或點我


Stat-Ease, Inc. is proud to announce Design-Expert, Version 8. Use this Windows®-based program to optimize your product or process.  It provides many powerful statistical tools, such as:

  • Two-level factorial screening designs: Identify the vital factors that affect your process or product so you can make breakthrough improvements
  • General factorial studies: Discover the best combination of categorical factors, such as source versus type of raw material supply
  • Response surface methods (RSM): Find the optimal process settings to achieve peak performance
  • Mixture design techniques: Discover the ideal recipe for your product formulation
  • Combinations of process factors, mixture components, and categorical factors: Mix your cake (with different ingredients) and bake it too!

Easily view response surfaces from all angles with rotatable 3D plots. Set flags and explore contours on interactive 2D graphs; and use the numerical optimization function to find maximum desirability for dozens of responses simultaneously.

This significant upgrade offers powerful new statistical tools, such as upfront power calculation for factorial designs and the Fraction of Design Space (FDS) graph for design evaluation. Other new features for ease-of-use, functionality, and power add extra appeal to a long-standing and well-loved program.  Use Design-Expert software to make breakthrough improvements to your product or a process. Not only screen for vital factors, but also locate ideal process settings for top performance and discover optimal product formulations. Try it, you are sure to like it! (Download the free 45-day trial athttp://www.statease.com/soft_ftp.html or take an Online Tour.)

Click here to view the Software Overview sheet (softoverview.pdf—109KB).



What's New in Version 8

New graphics and improved interface

  • Half-normal selection of important effects on all factorial designs*: Simple and robust method for selecting important effectsformerly available only for two-level designs.  For example, the screen shot to the right is from an experiment on 5 woods glued with 5 adhesives, using 2 applicators with 4 clamps at 2 pressures.  The vital effects become apparent at a glance!
    *(Detailed in “Graphical Selection of Effects in General Factorials”—winner of the Shewell Award for best presentation at the 2007 Fall Technical Conferenceco-sponsored by the American Society for Quality and the American Statistical Association.)

  • Half-normal selection of important effects on factorial designs

  • Smoother color gradations on 2D contours: More impressive for presentations to management, clients, or colleagues.

    Smoother Color Gradations on 2D Contours

  • Rounded contour values: More presentable defaults requiring less ‘fiddling’ for reporting purposes.
  • Plant flags on 3D surfaces: Previously, you could only put flags on 2D contour plots.  To the right we see a flag planted by numerical optimization on turbidity of a detergent formulation via mixture designa specialized application of response surface methods (RSM).

    Plant flags on 3D surfaces

  • New and fully configurable mesh option that reflects smooth, lighted colors off your 3D surface: Dazzle your customers and colleagues while providing highly-informative graphics showing how responses will react to process changes.(Mesh can be turned off if you like.)

  • 3D graphs that you can spin with your mouse: When you see your cursor turn into a hand (I), simply grab and rotate!  Double-click the graph to go back to the starting angle.

  • Push-button averaging on the factors tool: Provides far easier main effects plotting and makes interactions more meaningful.  Previously, the only option to average factors came via a hidden drop-list. The screen shot series below shows the result of simply pressing the "Avg" for 5 woods glued with 5 adhesives using 2 applicators at 2 pressures. This causes the least significant difference (LSD) bars to shrink, revealing an important difference between two particular clamps.

    Push-Button AveragingPush-Button Averaging 2

  • More-interactive cube plots: Click on design points to see factor levels and response predictions on graph legends, as below. 

    Interactive Cube Plot
  • Direct setting of discrete (fixed) numeric levels in response surface designs: Limit factor settings to reasonable levels but still produce continuous models.  The example below shows that 3 battery types must be tested at 3 discrete temperatures. Previously, this would have been possible but very tricky via a work-around. Now it's easy!

    Discrete (Fixed) Numeric Levels in RSM Designs

  • Discrete factor levels adhered to in numeric optimization: Find the most desirable setting for factors that are not continuous, such as the number of passes through a spray coater.

  • Enter input variables vertically (as shown above)When entering many levels, this may be more convenient than the horizontal layout.

  • Reference lines on plots: Horizontal, vertical, and free style-lines enhance plots. Below it becomes completely clear that four clamps tested for a wood-adhesive applications fall into two distinct groups—acceptable versus not acceptable, based on a cutoff of 50. 

    Reference Lines on Plots

  • Predicted vs. Actual graph availability in Model Graphs, not just in Diagnostics: This is useful when a response has been transformed because in Model Graphs mode, you can view the more relevant original scale.

    Predicted vs. Actual Graph Availability

  • Confidence, prediction, and tolerance intervals (CI, PI & TI) plotted with configurable colors in one-factor response plots: Convey prediction uncertainties via bands around the best fit.  The screen shot at right shows actual run results represented as red circles.  The solid line is the predicted value based on the polynomial model.  The bands are the CI (narrowest), PI, and TI (widest).

    Confidence, Prediction, & Tolerance Intervals Plotted with Color

  • Color-coded response surface graphs show where standard error increases: This makes it easier to understand why a predicted response will get you in trouble by extrapolating beyond actual experimentation regions.  The example at right shows a flag set beyond the axial points of a central composite designmaking the prediction meaningless.

    Color-coded RSM Graphs Show Where Standard Error Increases

Better mixture design and modeling tools

  • Partial quadratic mixture (PQM) analysis: Model non-linear blending behavior most effectively. The example below shows an orange drink formulated using artificial flavorings. Primary taste intensity, as measured by a sensory panel, proves to be non-linear in a way that is modeled best using PQM.

    Partial Quadratic Mixture (PQM) Analysis

  • Design for linear plus squared terms in mixture models: Reduce the number of blends required for optimally-designed experiments that reveal non-linear blending.

  • Design for special and full quartic mixture models: Capture extremely non-linear relationships among all components.

  • Blocking expanded to simplex mixture designs: For example, blend your cakes and bake them in two oven batches.

  • Trace plot options show end points as actual values when building designs using U-pseudo coding: The upper (“U”) bounded approach is advantageous when inverting regions in certain constrained mixture situations.  However, due to axis flipping, it’s easy to misinterpret trends when viewing a trace plot without this new feature.

  • Increased limit on components for screening and historical* designs.  Design-Expert now handles up to 50 individual ingredientsup from 40 and 24, respectively.
    *(An example is happenstance data collected by assaying retained samples from a period of material production.)

More choices when custom-designing your experiment

  • D-, IV-, and A-optimal design selection: New and expanded criteria when crafting experiments to models of choice within realistic constraints.

    D-, IV-, and A_Optimal_Design_Selection
  • Constraints calculator: Simplifies derivation of constraint inequalities. Below, food scientists cooking starch must bake it longer at low temperatures.  With program Help guidance, the design space’s lower left corner can be excluded using a multilinear constraint equation generated from a few user inputs.  An optimal design is then fitted to this region.

    Constraints Calculator

  • Tolerance-interval-based design sizing: Enhances your fraction of design space (FDS) plots to assess whether your planned experiment is large enough, given the underlying variability (noise), to establish tolerances within the acceptable range.

    FDS Graph

Additional statistics and more concise reporting of vital results

  • Improved curvature testing for factorials with center points: All design points are now fitted to the polynomial model used for predictions.  This provides a more realistic impact of significant non-linear response behavior.  Diagnostics can be done for the model adjusted for curvature or, via a view option, unadjusted.  Models without a term for curvature (unadjusted) are used for model graph and point predictions.
  • Coefficients summary: After modeling your response(s), see a concise table of coefficients that’s color-coded by relative significance.  Below, the second response is modeled only by main effets, two being significant at the p<0.1 level.

    Coefficients Summary

  • Condensed “Fit Summary” table: See vital details on model choices before delving into all the particulars.  Below you can see why the program recommends one model over others (note the superior R-squared values for quadratic).

    Condensed "Fit Summary" Table

  • Tolerance interval (TI) estimates on point prediction: This is important for verification studies to ensure your process stays within manufacturing specifications.  For example, the TI shown below provides assurance that thickness will remain within a required range of 4400 to 4600.

    Tolerance Interval (TI) Estimates on Point Prediction

Increased visibility and versatility of tools and features

  • Many new, high-visibility toolsOptions previously available via hidden View menu options are now easily seen and capitalized upon. The Design Tool shown 'floating' on the screen shot below is one example.

  • Design layout column widths now adjust automatically by double-clicking column-header boundaries: Multiple columns adjust simultaneously!

  • Attach row comments by right-clicking on row headers. View them by right clicking on the Select box and selecting “Comments” as a column header: A handy way to record important observations, as shown below.

    Attach Row Comments on Row Headers

  • The new comments are also visible when a point is highlighted on a plot.  See how the “starch sticky” comment from above is carried over to the plot, alerting the experimenter to the unique observations about this treatment.

    Comments Are Visible When a Point is Highlighted On a Plot

  • Topic Help, Tutorials, and Sample Files now also reside in the main Help menu: Follow these alternate paths for getting timely program advice.

  • In addition, Topic Help (F1) content has been enhanced.  Context specific information is right at your fingertips.  Just press the F1 key.

  • Response surface method (RSM) models can be fitted with factors in their actual levels:  This enables no-intercept model functionality.

  • Screen Tips is now a main menu item (“Tips”): Great visibility and easy access to very useful just-in-time advice, shown below. 

    Screen Tips is Now a Main Menu Item ("Tips") 

Enhanced design evaluation

  • Several new matrix measures are now provided: Most notable is the G-efficiency.  (This criterion, expressed on a 0 to 100 percent scale with higher being better, leads to designs that generate more consistent variance of your predicted response.  However, like any other single measure, it may not accurately reflect the overall effectiveness of a particular matrix.  That’s why Design-Expert provides an array of matrix statistics and graphics for overall design evaluation.)

  • Fraction of paired design space (FPDS): This resourceful tool lets you assess the power of RSM or mixture designs to detect specified signals (response differences judged important) in the presence of noise (system-standard deviation). Below, less than half the design space reveals the difference of interest. Ideally, this exceeds 80 percent, so here the experimenter should consider adding more runs to the design.

    Fraction of Paired Design Space (FPDS)

  • New, powerful tools for multiple response optimization: Options include standard error models. All else equal, choose system settings in regions predicted to exhibit the highest precision.

Many things made nicer, easier faster throughout the program

  • One-click updates: Check for free releases with one press (shown below) and download them directly.

    One-Click Updates
  • Better defaults and tick marks: Nicely rounded values provide presentable graphs straight away.

  • Zoom up graphs with your mouse wheel (a right-click resets to original size): Quickly zero in on regions of interest.

  • Hold down your left mouse button to drag graphs into various positions (a right-click resets original placement): It’s a fast way to situate the region of interest where you want it in the coordinate space. Components G and H in the mixture trace plot at right are constrained to very tight ranges relative to other ingredients. They are hardly visible without first zooming and then dragging the intersection (the overall centroid of the formulation space) to the middle.

    Drag Graphs Into Various Positions on Mixture Trace Plot

  • Separate preference tabs for X-Y versus surface graphs: Design-Expert version 8 delivers plotting and graphing simplicity.

  • Reduced graph-updating flicker: Now it’s less distracting when you redraw responses at varying input-variable levels.

  • Categoric factors (established via general factorials, for example) are now convertible to discrete numerics: This lets you apply response surface methodologies while adhering to processes that run most conveniently only at specific settings.

    Categoric Factors Are Now Convertible to Discrete Numerics

  • Keyboard shortcut for preferences: Press Ctrl + F8 to get a box allowing you to adjust all of the program preferences with one click, a convenient way to reset all of the default settings. 

    Keyboard Shortcut for Preferences
  • Color-by-point-type added to graph columns: Very useful addition to scatter-plots, such as this one below for a central composite design (CCD).

    Color-By-Point Type Added to Graph Columns

  • Ability to clear an analysis for any given response with a simple right-click: Enables a “do-over” with a minimum of hassle.

    Clear an Analysis for any Response with a Right Click

Technical stuff only the programmers will appreciate

  • Upgraded MFC (Microsoft Foundation Class) common controls: This new application framework provides an improved look and feel.

  • XML utility offers new script feature that lists all possible commands. You can parse files with extensions other than .xml. It also provides new import/export/reset-preference commands: More power to operate Design-Expert programmatically.

Appendix: Features that come along with the free update to the latest version

  • Graphical optimization frames the “design space” where all modeled responses fall within confidence, prediction or tolerance intervals (user choice): This feature is vital for quality-by-design (QbD).

  • Additional coloring option for graphical optimization that shades outside the limits, but inside the constraints: As seen pictured below, this snaps out the sweet spot for in-spec operations.

    Coloring Option for Graphical Optimization

  • Confidence interval (CI) added to numeric optimization: This facilitates finding a desirable setup within a quality-by-design (QbD) space.

  • If available, propagation of error (POE)—error transmitted from factor variation—is now included in intervals employed in graphs (LSD bars, for example) and numerical optimization: Develop more robust operating conditions by being more aware of potential sources of error.

  • Confirmation node (under optimization branch): Enter in the sample size (n) of your confirmation runs to generate the appropriate prediction interval.

    Confirmation Node

  • Improved auto-scaling, clearer design-summary display, etc.: Easier than ever to use.

  • Added advanced preferences: Provides more control over what features get enabled, etc.

  • An XML “self-test” to validate that the software installed OK: Helpful for satisfying FDA
Comments