Menu.FrF2levelTab1 {RcmdrPlugin.DoE}

Basic information for regular (fractional) factorial 2-level designs

Description

Basic information for regular (fractional) factorial 2-level designs

Brief statistical background

These designs are created with function FrF2 from package FrF2. Look up the documentation there for statistical details. Users who are not familiar with the terminology around regular fractional factorial designs (e.g. the concept of resolution) can browse the DoEGlossary.

Inputs on Tab Base Settings

name of design

must be a valid name. The design itself is created under this name in the R workspace.

number of runs

must be a power of 2. If this is omitted by unchecking the checkbox next to it, the function looks for the smallest possible design that fulfills all requirements.

number of factors

must always be specified. It must be smaller than the number of runs. Resolution IV designs are possible, if this number is smaller than half the number of runs.

replications

is the number of times each experimental run is conducted. If larger than 1, each run is conducted several times. If the checkbox next to the number of replications is checked, it is assumed that the experiment involves repeated measurements for one setup of the experimental run; if it is not checked, the experimental run itself is replicated with everything relevant newly set up (much more valuable than repeated measurements, unless the key driver of variability is in the measuring step). If the check box is not checked, the experiment will be randomized separately for each round of replications (first all first runs, then all second runs etc.).

blocks

is the number of blocks. Blocks are needed, if a factor relevant to the outcome but not of interest in itself is to be corrected for (e.g. batch of material). If you just give a number here, the program takes care of all details itself. If you are an expert and consider it necessary, you can provide more information on the special tab on blocking that is visible if special choices have been activated.

randomization settings

should normally not be changed; you can provide a seed if you want to exactly reproduce a randomized design created in the past. Unchecking the randomization box will produce a non-randomized experiment. This is usually NOT recommended.

Design Properties

allow to set a minimum resolution (III, IV or V+); for most purposes, IV is a good solution, i.e. it might be a good idea to increase the default choice.
In addition, a choice can be made between
MA = maximum resolution and minimum aberration, which implies usage of the design with the overall least severe degree of aliasing among effects,
OR
MaxC2 = maximum number of clear 2-factor interactions, which means that the design is chosen for which the most 2-factor interactions are not aliased with any other main effect or 2-factor interaction.

Note that, when choosing MaxC2, it is particularly important to also consider setting resolution to IV, as otherwise the design will very likely be of resolution III only, even if there are good resolution IV designs with many clear 2-factor interactions around.

Information Buttons

are available for looking at interim information on the suitable designs: The button View available designs allows to look at the properties of the designs in the catalogue, the button Show best 10 designs ... prints more detailed information on the best 10 available designs for the currently specified settings.

Note that you may want to retrieve the still open menu from somewhere in case the latter button is used, since this button releases the grab on the menu in support of scrollability of the printed results. I hope to find a better solution for this in the future.

Expert Choices

If the special choices are activated, it is possible
to specify a set of generators,
to specify a catalogue of designs different from the default one (if one is in the workspace; with package FrF2.catlg128 installed, a catalogue from that package can be loaded into the workspace using the menu item "Experts: load additional catalogue ...")
or to pick a specific design name from the current catalogue of designs.
Radio buttons activate one of these possibilities. Details for the consequences of these choices can be found in the help file for function FrF2.

Author(s)

Ulrike Groemping

See Also

See also FrF2 for the function that does the calculations, catlg for the catalogue of designs that this functionality is based on, and Menu.2level for the distinction between screening designs and regular fractional factorial designs.