# Linear programming using two-phase simplex and graphical method

a) When solving linear programming problems a number of problem cases can arise. Explain, with the aid of diagrams where appropriate, how you would identify each of the following cases when solving a two-variable problem using the graphical method and when using the two-phase Simplex method.

i) A non-unique solution

ii) An infeasible problem.

iii) An unbounded problem.

iv) A degenerate solution.

v) Describe the usual consequence of degeneracy and explain briefly how degeneracy can be avoided.

b) Explain how the two-phase Revised Simplex method indicates that a linear programming problem is (i) infeasible, (ii) unbounded and (iii) has infinitely many solutions.

© BrainMass Inc. brainmass.com October 10, 2019, 7:30 am ad1c9bdddfhttps://brainmass.com/math/linear-programming/linear-programming-two-phase-simplex-graphical-method-584171

#### Solution Preview

QUESTION 2

Two-phase Simplex method

A non-unique solution

If Dz=0, then we know that there is an alternate optima or the solution is non-unique. Refer to the inner productive rule for the derivation and formula for Dz.

An infeasible problem

If w'>0 in the Phase 1 LP, then the problem is infeasible

An unbounded problem

If maxin Z = +¥, then the problem is unbounded. In other words, if the optimum value, which is represented by Z in the LP, approaches infinity, then the problem is unbounded.

A degenerate solution

If w'=0 in the Phase 1 LP and one or more of the basic variables are equal to 0, then the problem has a feasible solution, but it is degenerate

Graphical method

A non-unique solution

When the obkective function line is moved parallel to the farthest or nearest boundary of the LP ...

#### Solution Summary

The expert describes the usual consequence of degeneracy and explain briefly how degeneracy can be avoided. Identify non-unique solution, degenerate solution, infeasible problem, unbounded problem using graphical method and two-phase simplex method.