com.numericalmethod.suanshu.optimization.constrained.linearprogramming.simplex
Class Tableau

java.lang.Object
  com.numericalmethod.suanshu.optimization.constrained.linearprogramming.simplex.Tableau

public class Tableau
extends java.lang.Object

This class implements a "tableau" and its operations used in the Simplex method when solving an LP problem.

Nested Class Summary

static class	Tableau.Label
static class	Tableau.LabelType

Field Summary

double	epsilon a precision parameter: when a number |x| ≤ ε, it is considered 0
static Tableau.Label	UNDEFINED

Constructor Summary

Tableau(Matrix T, double epsilon)
Construct a simplex tableau, assuming the standard form.

Tableau(StandardLpProblem2 problem)
Construct a simplex tableau from a standard problem.

Tableau(Tableau.Label[] rowLabel, Tableau.Label[] colLabel, Matrix T, double epsilon)
Construct a simplex tableau.

Tableau(Tableau table)
Copy ctor.

Method Summary

static DenseMatrix	buildTable(StandardLpProblem2 problem) phase 1: (feasible) tableau initialization
double	get(int row, int col) Get the table entry at [row, col].
double	getBCol(int row)
Tableau.Label	getColLabel(int i)
double	getCostRow(int col)
Tableau.Label	getRowLabel(int i)
boolean	isFeasible() Check if this table is feasible.
double	min() Get the minimum inferred by this table.
ImmutableVector	minimizer() Get the minimizer inferred by this table.
int	nCols() Get the number of non-basic variables.
int	nRows() Get the number of basic variables.
int	problemSize() Get the number of variables in the problem, the cost/objective function.
void	swap(int r, int s) Perform a Jordan Exchange to swap row r with column s.
java.lang.String	toString()

Methods inherited from class java.lang.Object

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

Field Detail

UNDEFINED

public static final Tableau.Label UNDEFINED

epsilon

public final double epsilon

a precision parameter: when a number |x| ≤ ε, it is considered 0

Constructor Detail

Tableau

public Tableau(Tableau.Label[] rowLabel,
               Tableau.Label[] colLabel,
               Matrix T,
               double epsilon)

Construct a simplex tableau.

Parameters:

rowLabel - the labeling of the basic variables

colLabel - the labeling of the non-basic variables

T - the table represented in matrix form

epsilon - a precision parameter: when a number |x| ≤ ε, it is considered 0

Tableau

public Tableau(Matrix T,
               double epsilon)

Construct a simplex tableau, assuming the standard form.

Parameters:

T - the table represented in matrix form

epsilon - a precision parameter: when a number |x| ≤ ε, it is considered 0

Tableau

public Tableau(StandardLpProblem2 problem)

Construct a simplex tableau from a standard problem.

Parameters:

problem - a standard linear programming problem

Tableau

public Tableau(Tableau table)

Copy ctor.

Parameters:

table - another Tableau

Method Detail

nRows

public int nRows()

Get the number of basic variables.

Returns:

number of basic variables

nCols

public int nCols()

Get the number of non-basic variables.

Returns:

number of non-basic variables

get

public double get(int row,
                  int col)

Get the table entry at [row, col].

Parameters:

row - the row index

col - the column index

Returns:

T[row, col]

getCostRow

public double getCostRow(int col)

getBCol

public double getBCol(int row)

getColLabel

public Tableau.Label getColLabel(int i)

getRowLabel

public Tableau.Label getRowLabel(int i)

isFeasible

public boolean isFeasible()

Check if this table is feasible.

Returns:

true iff the table is feasible

problemSize

public int problemSize()

Get the number of variables in the problem, the cost/objective function.

Returns:

the number of variables

min

public double min()

Get the minimum inferred by this table.

Returns:

the minimum

minimizer

public ImmutableVector minimizer()

Get the minimizer inferred by this table.

Returns:

the minimizer

swap

public void swap(int r,
                 int s)

Perform a Jordan Exchange to swap row r with column s.

Parameters:

r - the index to the entering variable (row)

s - the index to the leaving variable (column)

buildTable

public static DenseMatrix buildTable(StandardLpProblem2 problem)

phase 1: (feasible) tableau initialization

Parameters:

problem - a linear programming problem

Returns:

the tableau representation of a linear programming problem | A -b | T = | c' 0 |

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object