5.3. Choosing and running an experimental design
5.3.3. How do you select an experimental design?
188.8.131.52. Fractional factorial designs
The basic purpose of a fractional factorial design is to economically investigate cause-and-effect relationships of significance in a given experimental setting. This does not differ in essence from the purpose of any experimental design. However, because we are now able to choose fractions of a full design, and hence be more economical than ever, we also have to be aware that different factorial designs serve different purposes.
Broadly speaking, designs of resolution three, and sometimes four, seek to screen out the few important main effects from the many non-significant others. For this reason these designs are often termed main effects designs, or screening designs.
On the other hand, designs of resolution five, and higher, are used for focusing on more than just main effects in an experimental situation. These designs allow us to estimate interaction effects and investigate whether there are quadratic effects. This gives us an idea of the (local) shape of the response surface we are dealing with. For this reason they are termed response-surface method (RSM) designs.
Within the screening/RSM strategy of design there are a number of functional purposes that designs are put to. For example, an experiment might be designed to determine how better to make a product or process robust against the influence of external and non-controllable influences such as the weather. Experiments might be designed to troubleshoot a process, to determine bottlenecks, or to specify which component(s) of a product are most in need of improvement. Experiments might also be designed to optimize yield, or to minimize defect levels, or to move a process away from an unstable operating zone. All these aims and purposes can be achieved using fractional factorial designs and their appropriate design enhancements.