DOE: Design an experiment, fit a model to experimental data, and find best factor settings that optimize responses.

Purpose

This DOE app helps you determine the relationship between factors affecting a process and the output of that process.

You can use it to design an experiment, analyze experimental data and optimize output.

Features

• Various types of designed experiments to help you create a design.
• Customization of design table by replication, randomization, blocking, etc.
• Fit of mathematical model to data from test runs.
• Diagnostic statistics and graphs to help you interpret the results.
• Optimization of response after defining factor constraints.

Installation

The app requires R software (minimum required version 3.4.1) and packages (rsm, FrF2, alr3, qpcR).

For Origin version older than 2020b:

1. Install R or upgrade it before installing the App (download R from here).
2. Download the file DOE.opx, and drag-and-drop the file onto Origin workspace to install the App.
3. The App will start downloading dependent R packages. Wait a few minutes until it is completed.

For Origin 2020b and later versions:

1. Run Origin as administrator. Download the file DOE.opx and install it by choosing Tools->Package Manager->Tools->Install a Package.
2. After App installation, a pop-up dialog will ask if you agree to download and install R software before it proceeds. (If automatic downloading fails, you can install the App again or download R from here.)
3. For the first time you launch the App, it will start downloading dependent R packages. Wait a few minutes until it is completed.

Note: If downloading packages fails, a pop-up dialog will ask you to copy 2 lines of commands from Results Log and run them in R to complete downloading packages. 
          If you reinstall or upgrade R after installing the app, it would not work properly. Please reinstall the app.

Operation

Click the app icon to open the dialog with 4 buttons. Click one of them to open a dialog. 

• Power and Sample Size

  Determine sample sizes before creating a design. Calculate power for a given sample size or sample sizes for given power values.

• Create Design

  Identify factors and levels for each factor. Use techniques of the design to create a design table that makes the experiment cost-effective. Conduct a series of experiments and collect response data for each run in the table. The following types of design are supported.

  • Factorial Design
    • 2-Level Factorial
    • Plackett-Burman
    • General Full Factorial
  • Response Surface Design
    • Central Composite
    • Box-Behnken

• Analyze Design

  Fit a model to data collected using the design. Evaluate which factors have a significant impact on the response. You can interpret the relationship between factors and response through a variety of analytical tools and graph tools.

• Optimize Response

  Determine what levels of factors should be to achieve a desired response. You can choose goals like minimize, maximize and target to optimize. Put constraints on factor levels before optimization.

Help File

This app provides a help file that introduces design of experiments and explains dialog settings. Click ? button in dialog or press F1 to open it.

Sample OPJU File

This app provides a sample OPJU file and a data file. Right click on the app icon in the Apps Gallery window, and choose Show Samples Folder from the short-cut menu. A folder will open. Drag-and-drop the project file DOE Sample.opju from the folder onto Origin. The Notes window in the project shows detailed steps.
Note: If you wish to save the OPJU after changing, it is recommended that you save to a different folder location (e.g. User Files Folder).

v1.26:5/24/2022 Fix ANOVA table bugs. Fix crash issue caused by R v4.2.
v1.22: Fix optimization plot issue.
v1.21:Fix bugs of response values in contour and surface plots.
v1.2: fix alr3 package download issue.
v1.1: Automatically download and install R software after App installation. Need Origin 2020b or later version.