mbcdoe.design

Properties and methods for design of experiment (doe) design objects

expand all in page

Description

Use these properties and object functions to create and examine doe design objects.

Creation

Create a mbcdoe.design object using CreateDesign.

Properties

expand all

mbcdoe.designconstraint object containing one or more constraints, specified as a 1-by-n array.

mbcdoe.generator object containing new design type parameters based on the new design generator, specified as a 1-by-n array. Design generators provide the properties for all the design types.

The properties you can set depend on the design Type.

Use getAlternativeTypes to get a list of alternative generators.

This property is read-only.

Design inputs, specified as a 1-by-n array of mbcmodel.modelinput objects. For mbcdoe.design, D.Inputs = NewInputs updates the inputs. You cannot change the number of design inputs. Many designs have Limits properties in addition to model input ranges. These properties allow you to restrict the range of the design without changing the model or losing points via a constraint.

Data Types: integer

mbcmodel.model or mbcmodel.linearmodel object containing the design model, specified as a 1-by-n array.

D.Model = NewModel changes the model for the design to NewModel.

Setting this property changes optimal designs to custom if the new model does not support optimal designs.

Design object or design constraint object name, specified as a character vector.

Data Types: char | string

This property is read-only.

Number of design object inputs, specified as a scalar.

Data Types: integer

This property is read-only.

Number of points in the design after applying the constraints, specified as a scalar.

You can specify the number of points for a design using the generator object. The NumberOfPoints property of mbcdoe.generator is the number of points before any constraints are applied. You cannot specify the number of points for all design types (e.g., it is not allowed for Central Composite, Box Behnken).

Data Types: integer

Design points, specified as an array. You can perform any valid MATLAB® operation on this matrix. The number of columns of the points matrix must be the same as the number of inputs when setting Points. If you make an assignment to the Points, the design type changes to Custom. Points are only updated in the underlying design if they have changed.

Data Types: double

Fixed and free point status, specified as an array. Each point has a type of free, fixed or data.

You can specify fixed points. free is the default. If a point is matched to data, then it is of type data.

D.PointTypes returns a cell array of PointTypes, one for each design point. You cannot change a PointType of data to something else as the data is set by the test plan when matching the design to data.

To fix all the points in a design, use the FixPoints method.

Data Types: cell

This property is read-only.

Design type style, specified as a n-by-1 array of character vectors that contains the design type style, where n is the number of designs.

The style of the design style is one of the following:

  • 'User-defined'

  • 'Optimal'

  • 'Space-filling'

  • 'Classical'

  • 'Experimental data'

Data Types: char | string

This property is read-only.

Design type, specified as an array. To set the property, use the mbcdoe.generator object.

D.Type returns the design type. You can only choose a type when you create designs. You can only set the Type of a mbcdoe.generator object after design creation, or when calling Generate or Augment.

G.Type = NewType changes the Type, where G is a mbcdoe.generator object.

To get a list of types to use as alternative designs for the current design using getAlternativeTypes, enter this command. D is an mbcdoe.design object.

Dlist = getAlternativeTypes(D)

To use the alternative designs, the design Type must be one shown in the following table. The Type property determines the Style property.

StyleType
OptimalD-Optimal
V-Optimal
A-Optimal
ClassicalBox-Behnken
Central Composite
Full Factorial
Plackett-Burman
Regular Simplex
Space-fillingLattice
Latin Hypercube Sampling
Stratified Latin Hypercube
Sobol Sequence
Halton Sequence
Experimental dataDesign points replaced by data points
CustomAny design with a mix of Types (eg an optimally augmented space-filling design)

To specify the Type while creating and then generating a design of a given size:

D = CreateDesign(model,'Type','Sobol Sequence')
D = Generate(D,128);

Data Types: char | string

Object Functions

AddConstraintAdd design constraint
AugmentAdd design points
ConstrainedGenerateGenerate constrained space-filling design of specified size
CreateCandidateSetCreate candidate set for optimal designs
CreateConstraintCreate design constraint
DiscrepancyDiscrepancy value
FixPointsFix design points
GenerateGenerate design points
getAlternativeTypesAlternative model or design types
MaximinMaximum of minimum of distance between design points
MergeMerge designs
MinimaxMinimum of maximum distance between design points
OptimalCriteriaOptimal design criteria
RemovePointsRemove all nonfixed design points
Scatter2DPlot design points

Examples

collapse all

To create a space-filling design for a test plan TP.

sfDesign = CreateDesign(TP, ...
    'Type', 'Latin Hypercube Sampling',...
    'Name', 'Space Filling');

Version History

Introduced before R2006a

See Also

Topics