Implement capability properties

The algorithm capability properties are used by the Algorithm Selection Wizard to help the user select algorithms appropriate for their problem definition.

• MaxNumberOfObjectives - the maximum number of objectives that the algorithm can handle. If the algorithm has no limit to the number of objectives, then return -1.

• NumModelEvaluations - this is a general evaluation of how many model evaluations are required for convergence and it is of the AlgorithmNumModelEvaluations enum type.

• SupportsConstraints - whether the algorithm supports constraints. The property is of the FuzzyBoolean enum type. The algorithm can specify that it does or does not support discrete variables, or that it can handle constraints, but not very well.

• SupportsDiscreteVariables - whether the algorithm supports discrete variables. The property is of the FuzzyBoolean enum type. The algorithm can specify that it does or does not support discrete variables, or that it can handle discrete variables, but not very well.

• SupportsMinMax - whether the algorithm supports the minimizing or maximizing the objective. The property is of the FuzzyBoolean enum type. The algorithm can specify that it does or does not support discrete variables, or that it can handle minimizing and maximizing objectives, but not very well.

• SupportsMultipleObjectives - whether the algorithm supports multiple objectives. The algorithm can specify whether or not it directly supports multiple objectives. If the algorithm does not support multiple objectives, the infrastructure will create a single objective as a weighted sum of all of the objectives.

• SupportsNonSmoothResponses - whether the algorithm supports non-smooth responses (objectives and constraints). The property is of the FuzzyBoolean enum type. The algorithm can specify that it does or does not support non-smooth responses, or that it can handle non-smooth responses, but not very well.

• SupportsSolveFor - whether the algorithm supports solving an objective for a specific value. The property is of the FuzzyBoolean enum type. The algorithm can specify that it does or does not support discrete variables, or that it can handle solving for a value, but not very well. If the algorithm does not support solving for a value, the infrastructure will create an objective to minimize the difference between the objective and the solve for value.