Class LinearSolver.Builder

java.lang.Object

org.ojalgo.optimisation.GenericSolver.Builder<LinearSolver.Builder, LinearSolver>

org.ojalgo.optimisation.linear.LinearSolver.Builder

All Implemented Interfaces:

Optimisation, Optimisation.ProblemStructure

Enclosing class:

LinearSolver

public static final class LinearSolver.Builder
extends GenericSolver.Builder<LinearSolver.Builder, LinearSolver>

Compared to

invalid reference

LinearSolver.StandardBuilder

this builder:
1) Accepts inequality constraints
2) Has relaxed the requiremnt on the RHS to be non-negative (both equalities and inequalities)

Compared to ConvexSolver.Builder this builder:
1) Requires the objective function to be linear (or only the linear factors will be concidered)
2) Assumes (requires) variables to be non-negative

Nested Class Summary

Nested classes/interfaces inherited from interface Optimisation

Optimisation.Constraint, Optimisation.ConstraintType, Optimisation.Integration<M,S>, Optimisation.Model, Optimisation.Objective, Optimisation.Options, Optimisation.ProblemStructure, Optimisation.Result, Optimisation.Sense, Optimisation.Solver, Optimisation.State

Field Summary

Fields inherited from interface Optimisation.ProblemStructure

DEBUG

Constructor Summary

Constructors

Constructor	Description
Builder()

Method Summary

Modifier and Type	Method	Description
protected LinearSolver	doBuild(Optimisation.Options options)
LinearSolver.Builder	equalities(Access2D<?> mtrxAE, Access1D<?> mtrxBE)
LinearSolver.Builder	equality(double rhs, double... factors)
protected double[]	getLowerBounds()
LinearFunction<Double>	getObjective()
protected double[]	getUpperBounds()
LinearSolver.Builder	inequalities(Access2D<?> mtrxAI, Access1D<?> mtrxBI)
LinearSolver.Builder	inequality(double rhs, double... factors)
LinearSolver.Builder	lower(double bound)
final LinearSolver.Builder	lower(double... bounds)
(package private) <S extends SimplexStore> S	newSimplexStore(Function<LinearStructure, S> storeFactory, int... basis)
(package private) SimplexTableau	newSimplexTableau(Function<LinearStructure, SimplexTableau> tableauFactory)
final LinearSolver.Builder	objective(double... factors)
LinearSolver.Builder	objective(int index, double value)
LinearSolver.Builder	objective(MatrixStore<Double> mtrxC)
LinearSolver.Builder	toStandardForm()	Convert inequalities to equalities (adding slack variables) and make sure all RHS are non-negative.
LinearSolver.Builder	upper(double bound)
final LinearSolver.Builder	upper(double... bounds)

Methods inherited from class GenericSolver.Builder

append, append, build, build, countAdditionalConstraints, countEqualityConstraints, countInequalityConstraints, countVariables, doCountVariables, getAE, getAE, getAE, getAI, getAI, getAI, getBE, getBE, getBI, getBI, getC, getFactory, getLowerBounds, getObjective, getRowsAE, getRowsAI, getUpperBounds, reset, setNumberOfVariables, setObjective, solve, toString

Methods inherited from class Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Methods inherited from interface Optimisation.ProblemStructure

countConstraints

Constructor Details

Builder

Builder()

Method Details

equalities

public LinearSolver.Builder equalities(Access2D<?> mtrxAE,
 Access1D<?> mtrxBE)

Overrides:

equalities in class GenericSolver.Builder<LinearSolver.Builder, LinearSolver>

equality

public LinearSolver.Builder equality(double rhs,
 double... factors)

Overrides:

equality in class GenericSolver.Builder<LinearSolver.Builder, LinearSolver>

getObjective

public LinearFunction<Double> getObjective()

Overrides:

getObjective in class GenericSolver.Builder<LinearSolver.Builder, LinearSolver>

inequalities

public LinearSolver.Builder inequalities(Access2D<?> mtrxAI,
 Access1D<?> mtrxBI)

Overrides:

inequalities in class GenericSolver.Builder<LinearSolver.Builder, LinearSolver>

inequality

public LinearSolver.Builder inequality(double rhs,
 double... factors)

Overrides:

inequality in class GenericSolver.Builder<LinearSolver.Builder, LinearSolver>

lower

public final LinearSolver.Builder lower(double... bounds)

lower

public LinearSolver.Builder lower(double bound)

objective

public final LinearSolver.Builder objective(double... factors)

objective

public LinearSolver.Builder objective(int index,
 double value)

objective

public LinearSolver.Builder objective(MatrixStore<Double> mtrxC)

toStandardForm

public LinearSolver.Builder toStandardForm()

Convert inequalities to equalities (adding slack variables) and make sure all RHS are non-negative.

Defines optimisation problems on the LP standard form:

min [C]^T[X]
when [AE][X] == [BE]
and 0 invalid input: '<'= [X]
and 0 invalid input: '<'= [BE]

A Linear Program is in Standard Form if:

• All constraints are equality constraints.
• All variables have a nonnegativity sign restriction.

Further it is required that the constraint right hand sides are nonnegative (nonnegative elements in [BE]). Don't think that's an actual LP standard form requirement, but it is commonly required, and also here.

The LP standard form does not dictate if expressed on minimisation or maximisation form. Here it should be a minimisation.

upper

public final LinearSolver.Builder upper(double... bounds)

upper

public LinearSolver.Builder upper(double bound)

doBuild

protected LinearSolver doBuild(Optimisation.Options options)

Specified by:

doBuild in class GenericSolver.Builder<LinearSolver.Builder, LinearSolver>

getLowerBounds

protected double[] getLowerBounds()

getUpperBounds

protected double[] getUpperBounds()

newSimplexStore

<S extends SimplexStore> S newSimplexStore(Function<LinearStructure, S> storeFactory,
 int... basis)

newSimplexTableau

SimplexTableau newSimplexTableau(Function<LinearStructure, SimplexTableau> tableauFactory)