Design of Experiments Guide > Custom Designs
Publication date: 10/01/2019

Custom Designs

Construct Designs That Meet Your Needs

Use the Custom Design platform to construct optimal designs that are custom built for your specific experimental setting. Generally, a custom design is more cost-effective than a design obtained using alternative methods. You can perform the following tasks:

Enter factors of many different types.

Specify constraints on the design space.

Indicate which effects are necessary to estimate and which are desirable to estimate, if possible, given the number of runs.

Specify a number of experimental runs that matches your budget.

The Custom Design platform constructs a wide variety of designs, including these special design types: Screening, Response Surface, Mixture, Random Block, Split Plot, Split-Split-Plot, and Two-Way Split Plot.

This chapter contains a detailed example of how to use the Custom Design platform, followed by information about the platform. See also Examples of Custom Designs section.

Figure 4.1 Color Map for Absolute Correlations 

Contents

Overview of Custom Design

Example of a Custom Design

Create the Design
Analyze the Data

Custom Design Window

Responses
Factors
Define Factor Constraints
Model
Alias Terms
Design Generation
Design
Design Evaluation
Output Options

Custom Design Options

Description of Options
Simulate Responses
Save X Matrix
Number of Starts
Design Search Time
Set Delta for Power

Technical Details for Custom Designs

Designs with Randomization Restrictions
Covariates with Hard-to-Change Levels
Numbers of Whole Plots and Subplots
Optimality Criteria
D-Efficiency
Coordinate-Exchange Algorithm
Want more information? Have questions? Get answers in the JMP User Community.