Function GMP::Instance::GenerateRobustCounterpart(MP, UncertainParameters, UncertaintyConstraints, Name)

GMP::Instance::GenerateRobustCounterpart

The function GMP::Instance::GenerateRobustCounterpart generates the robust counterpart of a (linear) mathematical program.

If the deterministic model is a linear program (LP) then the robust counterpart will be a LP if the uncertainty constraints are linear, or a second-order cone program (SOCP) if some of the uncertainty constraints are ellipsoidal.

If the deterministic model is a mixed-integer program (MIP) then the robust counterpart will be a MIP if the uncertainty constraints are linear, or a mixed-integer second-order cone program (MISOCP) if some of the uncertainty constraints are ellipsoidal.

SOCP and MISOCP problems can be solved by using CPLEX or Gurobi.

GMP::Instance::GenerateRobustCounterpart(
     MP,                      ! (input) a symbolic mathematical program
     UncertainParameters,     ! (input) a set of uncertain parameters
     UncertaintyConstraints,  ! (input) a set of uncertainty constraints
     [Name]                   ! (optional) a string expression
)

Arguments

MP

A symbolic mathematical program in the set AllMathematicalPrograms. The mathematical program should have model type LP or MIP.

UncertainParameters

A subset of AllUncertainParameters.

UncertaintyConstraints

A subset of AllUncertaintyConstraints.

Name

A string that holds the name for the generated robust counterpart.

Return Value

A new element in the set AllGeneratedMathematicalPrograms with the name as specified by the name argument.

Note

• If the Name argument is not specified, or if it is the empty string, then the name of the symbolic mathematical program followed by ‘robust counterpart’ is used to create a new element in the set AllGeneratedMathematicalPrograms.

• If AIMMS detects that the robust counterpart is infeasible during the generation, AIMMS will issue a warning and the robust counterpart will not be generated.

• As part of the generation, AIMMS will check whether the uncertainty set satisfies the Slater condition (controlled by the option Slater_condition_check). To do so, AIMMS will solve a linear program (LP) or a second-order cone program (SOCP).

• The created GMP cannot be modified, e.g., it is not allowed to change row or columns in the robust counterpart.

See also

The procedure GMP::Instance::Solve.