Plackett Burman designs

Table 5.10. Plackett-Burman design matrix lor N 8 experiments and consequently m = 7 factors (including dummy variables) at two levels... 

If nonlinear effects are expected the variables must be varied at more than two levels. A screening plan comparable to the Plackett Burman design but on three levels is that of Box and Behnken [I960]. 

TABLE 7 A Plackett-Burman Design in Which Eight Factors are Studied Simultaneously in 12 Experiments... 

ASTM El 169-89 Standard Guide for Conducting Ruggedness Tests (Plackett-Burman Design), American Society for Testing and Materials (ASTM), 100 Barr Harbor Drive, West Conshohocken, PA 19428-2959. 

TABLE 7 Plackett-Burman Design to Examine up to Seven Factors in Eight Experiments... 

TABLE 9 Results of the Eight-Experiment Plackett-Burman Design ... 

Plackett-Burman design. (----) least-squares line through all effects (-) assumed line through... 

Vander Heyden, Y., Hartmann, C., Massart, D. L., Michel, L., Kiechle, P., and Ernl, R (1995). Ruggedness tests for a high-performance liquid chromatographic assay comparison of an evaluation at two and three levels by using two-level Plackett-Burman designs. Anal. Chim. Acta 316, 15-26. 

The relationship of saturated fractional factorial and Plackett-Burman designs. 

The foldover design for the Plackett-Burman design in Table 14.9. 

An 11-factor Plackett-Burman design with foldover. 

Using row and column operations, convert the following 7-factor Plackett-Burman design to the saturated fractional factorial design shown in Table 14.7. 

When all of the variables are quantitative, an estimate of the experimental error can be obtained by adding to the full factorial, fractional factorial or Plackett-Burman design, a number of runs at the center of the design. The center of the design is the midpoint between the low and high settings of the two-level factors in the experiment. Thus, if there are p variables, and the levels of the variables have been coded (-1, +1), then the center of the design is (Xj, = (0, 0,. .., 0). If the... 

It is clear that if the first-order model (equation (5)) is assumed to be valid, then a resolution III design or a Plackett-Burman design can be used since this design will estimate all of the yff without bias. 

For this reason one prefers to apply an experimental design. In the literature a number of different designs are described, such as saturated fractional factorial designs and Plackett-Burman designs, full and fractional factorial designs, central composite designs and Box-Behnken designs [5]. 

---