Dakota Reference Manual  Version 6.12
Explore and Predict with Confidence
 All Pages

Characteristic values to scale each objective function


Alias: objective_function_scales

Argument(s): REALLIST

Default: 1.0 (no scaling)


Each real value in primary_scales is a nonzero characteristic value to be used in scaling each objective function. They only have effect when the associated method specifies scaling.

This keyword is required for responses-objective_functions-scale_types of 'value' and optional for 'log'. When specified in conjunction with log, scale values are applied prior to the logarithm, to permit log scaling of strictly negative response functions.

Length: When specified, primary_scales must have length one of:

  • One (the single value will be applied to each objective function); or
  • Number of response groups (objective_functions), so the number of scalars plus the number of fields when fields are present; or
  • Total number of response elements, so the number of scalar responses plus the sum of the lengths of the fields.

Usage Tips:

When primary_scales are specified, but not responses-objective_functions-scale_types, the scaling type is assumed to be 'value' for all primary response functions.

Use scale values of 1.0 to selectively avoid scaling a subset of response functions.

When scaling is active, it is applied to objective functions prior to any weights and multi-objective sum formation. See the equations in objective_functions.

See the scaling information under specific methods, e.g., method-*-scaling for details on how to use this keyword.