Screening factor effects estimation

When such time effect is present, the design responses are corrected relative to the (nominal) experiment performed at the beginning of the experimental design (Eq. 2.10) (Figure 2.16) (5,16,106). These corrected responses are then used to estimate the factor effects from screening designs or to build the model from response surface designs (see further). From both the estimated effects and the model coefficients then the time effect has been removed ... 

In this chapter we explore factorial-based experimental designs in more detail. We will show how these designs can be used in their full factorial form how factorial designs can be taken apart into blocks to minimize the effect of (or, if desired, to estimate the effect of) an additional factor and how only a portion of the full factorial design (a fractional replicate) can be used to screen many potentially useful factors in a very small number of experiments. Finally, we will illustrate the use of a Latin square design, a special type of fractionalized design. 

The next design in degree of complexity is the two-factor, two-level factorial. This design is sometimes used to screen certain variables to determine whether their effect is important, and to estimate its direction, prior to larger scale quantification. It can be of advantage in determining what ranges of the variables should be studied. 

A minimal risk level (MRL) is an estimate of the daily human exposure to a hazardous substance that is likely to be without appreciable risk of adverse noncancer health effects over a specified duration and route of exposure. These substance-specific estimates, which are intended to serve as screening levels, are used by ATSDR health assessors and others to identify contaminants and potential health effects thai may be of concern at hazardous waste sites. The chronic-duration oral MRL of 1 picogram/kilogram/day or pg/kg/day for TCDD, or total TEQs, (ATSDR 1999) was based on neurobehavioral effects in monkeys. Based on this value, an EMEG of 50 ppt (0.05 ppb) TCDD, which is equivalent to 50 ppt (0.05 ppb) TEQs, was derived for exposure from contaminated soil. Uncertainty factors of 90 (total) were used in the calculations of the MRL (for further details, see ATSDR 1999). Based on a review of more recent literature, ATSDR scientists conclude that the MRL of 1 pg/kg/day is approximately two orders of magnitude below the noncancer health effect levels observed in recent studies. This is also true for cancer effect levels. 

The objective of a factor screening experiment is to investigate, efficiently and effectively, the factors of a system that possibly may be important to its performance and to identify those factors that have important effects. Once the important factors have been identified from the screening experiment, the experimenter will typically move the region of experimentation from the initial location towards one more likely to contain the optimum. The method of steepest ascent (Myers and Montgomery, 2002, Chapter 5) is the procedure most widely employed for this activity. Finally, once near the optimum, the experimenter will usually conduct one or more experiments in order to obtain a fairly precise description of the response surface and an estimate of the optimum conditions. 

---