Simplex Method Examples
Example 2: (Infinitely many solutions)
Show by simplex method that the following LPP has infinitely many
solutions
= 4 + 10
Subject to
2 + ≤ 10
2 + 5 ≤ 20
2 + 3 ≤ 18
, ≥0
Solution: We write the LPP in the standard form.
= 4 + 10
Subject to
2 + + = 10
2 + 5 + = 20
2 + 3 + = 18
, , , , ≥0
Let the initial non-basic variables be , . So, the initial basic feasible
solution is given by = 10, = 20, = 18.
4 10 0 0 0
Basis
0 2 1 1 0 0 10 10
0 2 5 0 1 0 20 4 →
0 2 3 0 0 1 18 6
0 0 0 0 0
=!
"
− 4 10 0 0 0
↑
%
% → , % → % − % , % → % − 3%
5
4 10 0 0 0
Basis
0 8/5 0 1 -1/5 0 6 15/4 →
10 2/5 1 0 1/5 0 4 10
0 4/5 0 0 -3/5 1 6 15/2
4 10 0 2 0
=!
"
− 0 0 0 -2 0
↑
All − ≤ 0, optimal solution is = 0, = 4, &'( = 40.
If among non-basic original decision variables, − = 0 implies the
existence of an alternative optima.
We have two original decision variables , , of which is non-basic
with − = 0. Therefore, alternative optima exists.
2 4
% → % )5/8+, % → % − % , % → % − %
5 5
4 10 0 0 0
Basis
4 1 0 5/8 -1/8 0 15/4
10 0 1 -1/4 1/4 0 5/2
0 0 0 -1/2 -1/2 1 6
4 10 0 2 0
=!
"
− 0 0 0 -2 0
, ,
All − ≤ 0, second optimal solution is = , = , &'( = 40.
-
Note that &'( = 40 in both the optimal solutions.
, ,
Let = )0,4,6,0,6+ /0 = 1 , , 0,0,62.
-
Let ∗
=4 + )1 − 4+ , 0 ≤ 4 ≤ 1.
15 5
⇒ ∗
= 4)0,4,6,0,6+ + )1 − 4+ 6 , , 0,0,67
4 2
15 15 5 3
=6 − 4, + 4, 64, 0,67
4 4 2 2
15 15 5 3
⇒ ) ∗ + = 4 6 − 47 + 10 6 + 47 = 15 − 154 + 25 + 154 = 40
4 4 2 2
⇒ ∗ is also an optimal solution.Note that for every 4, there is ∗
. Thus,
the given LPP has infinitely many optimal solutions.
Example 3: (A case of unbounded optima, no solution)
Solve
= 2 +3 +4 + -
8 9 : :;
+ 5 + 9 − 6 - ≥ −2
3 − + + 3 - ≤ 10
2 +3 −7 +8 - ≤0
, , , -≥0
Solution: Write the above in the standard form.
= 2 +3 +4 + -
8 9 : :;
− −5 −9 +6 - + = 2
3 − + + 3 - + = 10
2 +3 −7 +8 -+ = 0
, , , -, , , ≥ 0
Assume initial non-basic variables are , , , - . This implies the initial
basic feasible solution is given by = 2, = 10, = > ?ℎ A > >
0 CC /;8Dℎ. [Reason: If we start with = 0, solution is
degenerate. This degeneracy remains in the simples table till the end. To
remove degeneracy in the simplex table we approximate by a small but
positive (>0) number >. ]
2 3 4 1 0 0 0
CSV -
(Basis)
0 -1 -5 -9 6 1 0 0 2 −E
0 3 -1 1 3 0 1 0 10 10 →
0 2 3 -7 8 0 0 1 > −E
0 0 0 0 0 0 0
=!
"
− 2 3 4 1 0 0 0
↑
% → % + 9% , % → % + 7%
2 3 4 1 0 0 0
CSV -
(Basis)
0 8 -14 0 33 1 9 0 92 −E
4 3 -1 1 3 0 1 0 10 −E
0 23 -4 0 35 0 7 1 70 −E
+>
12 -4 4 12 0 4 0
=!
"
− -10 7 0 -11 0 -4 0
↑
Note that in the above table, is an incoming variable for the next
iteration but there is no outgoing variable as all column entries are
negative. This indicates that an optimal solution exists at the point of
infinity. That means this is the case of an unbounded optima. So, the
given LPP has no solution.
Note that value depends upon . We can take as large as possible
without violating optimality conditions.
