Design Types

The following design types are supported:

Factorial Designs are used for screening vital factors, and they are available in standard design folios and robust design folios. The links below explain these designs in the context of standard design folios.

  • Two Level Factorial: You can have 2 to 15 factors; each factor must have exactly two levels. This design type can be used to investigate all main effects and their interactions (full factorial) or just a subset of them (fractional factorial).
  • Plackett-Burman Factorial: This is a special category of two level fractional factorial design, where only a few specifically chosen runs are performed to investigate the main effects (i.e., no interactions). You can have 2 to 47 factors; each factor must have exactly two levels.
  • General Full Factorial: Different factors can have different numbers of levels. All effects and their interactions are investigated. You can have 2 to 7 factors; each factor can have 2 to 20 levels.
  • Taguchi Orthogonal Array (OA) Factorial: This is a highly fractional design type. It can investigate main effects with just a few runs, and different factors can have different numbers of levels.

Response Surface Method Designs are used for system optimization and are available only in standard design folios.

  • Central Composite Design Response Surface Method: This is the most commonly used response surface methodology design. It is typically used to study the quadratic effects of factors.
  • Box-Behnken Response Surface Method: This design type allows you to consider from three to nine quantitative factors at three levels each (one level being a center point). If setting all factors at the high level at the same time carries a risk of equipment damage or violates other constraints, these designs provide a matrix that avoids setting all factors at extreme values simultaneously.

One Factor Designs are used for comparison, and they are available only in standard design folios.

  • Only one factor is investigated, and the response is compared at different factor levels. The factor levels must be qualitative. (To perform one factor analysis with quantitative levels, use the Multiple Linear Regression folio.)

Mixture Designs are used when the factors in an experiment are components in a mixture, and when you wish to determine the best proportions to use for each component.

  • Simplex Designs include:

    • Simplex Lattice: With this design, the blends in the experiment are determined by the specified number of levels of each component (i.e., the degree of design + 1). Since it includes all the reasonable combinations of components, this design is useful when the number of components is not large, and a higher polynomial equation is needed for optimization.

    • Simplex Centroid: By default, this design includes single-component blends (vertices) and all centroids up to the dimension g, where g is the number of components. Users can specify the degree of design (i.e., the dimension of centroids). Since it usually has fewer test runs than a simplex lattice design, this design is useful when the number of components is larger, but a lower polynomial equation will suffice for optimization.

    • Simplex Axial: This design includes vertices, blends with an absent component (edge points), center points and interior points between the center point and vertices (axial points). It is mainly used for screening components.

  • Extreme Vertex Designs: This design allows you to impose additional limits on the component values by specifying upper bounds on components and defining linear constraints for blends.

Related Topics and Links