Docs: Design description

Mon, 01 Jan 0001 00:00:00 +0000

In the first tab Experimental factors, the characteristics of the factors involved in the experiment are described. There are two types for factors: quantitative and categorical.

To add one quantitative factor, click on the button

Similarly, to add one categorical factor, click on the button

Click on the respective buttons repeatedly until the desired number and type of factors are added.

Setting number of levels for categorical factors

For categorical factors that are added, specify the number of levels associated with each categorical factor. By default, the number of levels for each categorical factor is set to 2. The number of levels for each categorical factor can be changed by clicking on the respective cells in the column levels and entering the actual number of levels for the respective factor.

Setting reset difficulty

In some experiments, it may happen that changing the settings for certain factors is more difficult (or expensive) than for some other factors. In such situations, we make a distinction between hard-to-change factors and easy-to-change factors. In the software, by specificying the reset difficulty for each factor that is added (quantitative or categorical), the software produces designs that take this into account. By setting the reset difficulties, the designs produced by the software will guarantee that the recommended number of changes in settings for the hard-to-change factors is less than that for the easy-to-change factors.

Clicking on the cells in the column reset difficulty, allows you to change these settings and offers three options as given below.

There are three scenarios to consider:

If all factors are equally easy or hard to change, the reset difficulty option for all factors can be set to normal.
If the factors can be divided into two groups say A and B, such that the settings of all factors in a given group are equally hard to change, and the settings for all factors belonging to group A are harder to change than for all factors belonging to group B, then assign "hard" as the reset difficulty for all factors in group A and assign "normal" as the reset difficulty for all factors in group B.
If the factors can be divided into three groups say A, B and C, such that the settings of all factors in a given group are equally hard to change, and the settings for all factors belonging to group A are harder to change than for all factors belonging to group B, and all factors belonging to group B are harder to change than for all factors belonging to group C, then assign "hardest" as the reset difficulty for all factors in group A, "hard" as the reset difficulty for all factors in group B and assign "normal" as the reset difficulty for all factors in group C.

Miscellaneous notes

At this stage, the user cannot make anymore changes. In a later screen, the user will be allowed to

Change names of each factor
Set the minimum and maximum value for each quantitative factor based on the engineering units
Provide labels for each categorical level for each categorical factor

After specifying the type of each factor and their reset difficulty, the user can click on Next at the bottom of the page which will take you to the Summary tab.

After verifying all details in the Summary tab, you can then click on Optimal design to go to next page.

Alternatively, if the user prefers to explore the existing catalog of designs already available in the EFFEX software, click on Search for design in catalog. This will lead to the Catalog search page, where the user will be presented with a selection of designs that match the requirements that have been already entered in the Experimental factors tab. It is possible that there are no designs in the catalog that match the requirements specified by the user, in which case, click on Optimal design.

Docs: Optimal design

Mon, 01 Jan 0001 00:00:00 +0000

In the tab Model and design structure, you can select the model you would like to study in your experiment and the minimum and maximum limit for the number of runs for your experiment.

For Model of interest, there are three choices:

Main effects (this includes only first-order effects)
Main and interaction effects (this includes first-order effects as well as two-factor interaction effects)
Main and second-order effects (this includes first-order effects, two-factor interaction effects and quadratic effects)

For number of runs, a range can be specified either using the slider or using the two text boxes. The text box to the left can be used to set the lower limit for the number of runs, and the text box to the right can be used to specify the upper limit.

In the tab Coordinate exchange algorithm parameters, you can specify the optimality criteria that is of interest and the number of random starts.

For the Optimality criteria, there are three choices: D-optimal, A-optimal and I-optimal

All three criteria are selected by default. However, the user has the choice to uncheck the criteria that are not relevant for their experiment.

Effex app documentation – Guided DoE

Docs: Design description

Setting number of levels for categorical factors

Setting reset difficulty

Miscellaneous notes

Docs: Optimal design