To create an experiment with all categorical factors,
Quick Designs asks about the characteristics of the factors to study. Often, you select
an answer for the maximum number of levels for a factor in the experiment.
- When all factors have 2 levels, select Estimate main and
interaction effects when all factors have two
levels.
- When at least one factor has 3 or more levels, select Estimate main and
interaction effects when at least one factor has more than two
levels.
Consider whether you need 1 of 2 types of designs for more specific cases. One case is a
hard-to-change factor. If you have a hard-to-change factor, select Estimate main
and interaction effects when all factors have two levels, and one is
hard to change.
The other case is an experiment that includes noise factors that you can control in the
experiment but not in the production environment. In this case, the goal of the
experiment is to determine the best settings for the controllable factors as the
uncontrollable factors vary. If you have this specific goal, select Find the
optimal factor settings to achieve robustness to uncontrollable
noise.
Decision details
The following information defines levels for categorical variables, defines main
effects, defines interaction effects, defines a hard-to-change factor, and
describes a design to find the optimal factor settings to achieve robustness to
uncontrollable noise.
- What are factor levels for categorical variables?
- For categorical variables, factor levels are the values of the factor to
study in the experiment. For example, in an experiment about plastic
strength, you decide to include the type of hardening additive as a factor.
You have 3 types of additives that you label A, B, and C. The factor has 3
levels.
- What is a main effect?
- A main effect is an estimate of the effect of a single factor. For example,
fertilizer company B is comparing the plant growth rate measured in plants
treated with their product compared to plants treated by company A's
fertilizer. In the experiment, fertilizer B has a higher plant growth rate
mean than fertilizer A. The difference in the means is the main effect of
the fertilizer factor.
- What is an interaction effect?
- An interaction effect is an estimate of the way that the effect of one
factor depends on the value of one or more other factors. For example, if
the levels are wide enough, the effect of time on the quality of a baked
product depends on temperature. When the temperature is so low that the
product is under cooked, then an increase in time increases the quality.
When the temperature is in an acceptable range, an increase in time
decreases the quality because the product burns. The effect of time depends
on the value of temperature.
- What is a hard-to-change factor?
- A hard-to-change factor is a factor that is difficult to randomize
completely because of time or cost constraints. For example, temperature is
a common hard-to-change factor because adjusting temperature often requires
significant time to stabilize. A split-plot design is a designed experiment
that includes at least one hard-to-change factor. In a split-plot
experiment, levels of the hard-to-change factor are held constant for
several experimental runs.
- When would I find the optimal factor settings to achieve robustness to
uncontrollable noise?
- A Taguchi design is a designed experiment that lets you choose a product or
process that functions more consistently in the operating environment.
Taguchi designs recognize that not all factors that cause variability can be
controlled. These uncontrollable factors are called noise factors. Taguchi
designs try to identify controllable factors (control factors) that minimize
the effect of the noise factors. During experimentation, you manipulate
noise factors to force variability to occur and then determine optimal
control factor settings that make the process or product robust, or
resistant to variation from the noise factors. A process designed with this
goal will produce more consistent output. A product designed with this goal
will deliver more consistent performance regardless of the environment in
which it is used.
- A well-known example of Taguchi designs is from the Ina Tile Company of
Japan in the 1950s. The company was manufacturing too many tiles outside
specified dimensions. A quality team discovered that the temperature in the
kiln used to bake the tiles varied, causing nonuniform tile dimension. They
could not eliminate the temperature variation because building a new kiln
was too costly. Thus, temperature was a noise factor. Using Taguchi designed
experiments, the team found that by increasing the clay's lime content, a
control factor, the tiles became more resistant, or robust, to the
temperature variation in the kiln, letting them manufacture more uniform
tiles.
