Sensitivity Analysis in Linear Programming

10-04-2016
SENSITIVITY ANALYSIS IN LINEAR PROGRAMMING
In all LP models the coefficient of the objective function and the constraints are supplied as input
data or as parameters to the model.
The optimal solution obtained by the simplex method is based on the values of these coefficients.
In practice the values of these coefficients are seldom known with absolute certainty, because
many of them are functions of some uncontrollable parameters.
For instance, future demands, the cost of raw materials, or the cost of energy resources cannot be
predicted with complete accuracy before the problem is solved.
Hence the solution of a practical problem is not complete with the mere determination of the
optimal solution.
Each variation in the values of the data coefficients changes the LP problem, which may in turn
affect the optimal solution found earlier.
In order to develop an overall strategy to meet the various contingencies, one has to study how the
optimal solution will change with changes in the input (data) coefficients.
This is known as sensitivity analysis or post-optimality analysis.

Some data coefficient or parameters of the linear program may be controllable; for
example, availability of capital, raw material, or machine capacities. Sensitivity analysis
enables us to study the effect of changes in these parameters on the optimal solution. If it
turns out that the optimal value (profit / cost) changes (in our favor) by a considerable
amount for a small change in the given parameters, then it may be worthwhile to
implement some of these changes. For example, if increasing the availability of labor by
allowing overtime contributes to a greater increase in the maximum return, as compared
to the increased cost of overtime labor, then we might want to allow overtime production.
In many cases, the values of the data coefficients are obtained by carrying out statistical
estimation procedures on past figures, as in the case of sales forecasts, price estimates,
and cost data. These estimates, in general, may not be very accurate. If we can identify
which of the parameters affect the objective value most, then we can obtain better
estimates of these parameters. This will increase the reliability of our model and the
solution.
10-04-2016
Sensitivity analysis
Here we attempt to address issues such as:
What happens to the solution if the values of the objective
function coefficient change?
What happens when the values of RHS change?
What happens when the coefficients in the constraints
change?
What happens when we add a new variable?
What happens when we add a new constraint?
Maximize, Z=4X1+3X2+5X3
Subject to:
X1+2X2+3X39
2X1+3X2+X312
X1,X2 ,X30
4
1
2
2
3
3
1
1
0
0
1
9
12
1/3
5/3
2/3
7/3
1
0
1/3
-1/3
0
1
3
9
7/3
-1/3
-5/3
15
0
1
1/5
7/5
1
0
2/5
-1/5
-1/5
3/5
6/5
27/5
-18/5
-6/5
-7/5
138/5
RHS
0
0
5
0
5
4
The optimal solution is, X1=275, X3=65, Z=138/5.
3
12
9
27/5
10-04-2016
Changes in values of objective function coefficients

(C )
Two cases:
Changes in C value of a non basic variable.
Changes in C value of a basic variable.
Changes in
value of a non-basic variable
In our example, variable is a non-basic optimum. Let us consider changes in C

values. Let us assume a value C instead of the given value of 3. This will mean that all
alone will change. We compute

= C -33/5.
The present value of C =3 results in becoming negative. A value of C >33/5
would make
take a positive value resulting in variable X entering the basis.
Let us consider C =7. This would make
=2/5 and the variable enters the
basis. We perform simplex iterations till we reach the optimum. The present optimal
table (with modified values of C and
is shown and the subsequent iterations
are shown in table 3.11)
10-04-2016
0
RHS
5
4
5
7
0
1
1/5
7/5
1
0
2/5
-1/5
-1/5
3/5
6/5
27/5
2/5
-6/5
-7/5
138/5
-1/7
5/7
0
1
1
0
3/7
-1/7
-2/7
3/7
3/7
27/7
-2/7
-8/7
-11/7
204/7
= ,
The new optimal solution is,
Changes in
6
27/7
= 204/7.
value of a basic variable
Let us consider a change in the objective function coefficient of variable . Let us

call it . The change will affect all the non-basic values. We compute
=3-1-7 /5
= 02+
=0+13
/5
/5
For the present value of =4, the values are -18/5, -6/5 and -7/5. All the values
are negative. We also observe that for
<10/7, can become positive. For
> 10 and
>0
for >3/5,
>0.
In the range 6/5
consider
2/5 and
=12.
10, the present set of basic variable will be optimal. Let us
=3-[12x7/5+1] = -28/5,
= 1 = 31/5. Variable
iterations till it reaches the optimum.
The modified optimum (for
3.12
= 2 +
enters the basis and we perform simplex
= 12) and the final solutions are shown in Table
10-04-2016
12
0
RHS
5
12
0
12
0
1
1/5
7/5
1
0
2/5
-1/5
-1/5
3/5
6/5
27/5
-74/5 0
2/5
-31/5 354/5
0
1
3/2
5/2
1
0
-1/2
3
6
-15
-1
-6
72
Here also the optimum was found within one iteration. The optimum solution
is
= 12,
=3
= 72.
2 Changes in RHs values

For change in the RHs value, say
RHS =
.the change results in change in the RHS values.
2/5 1/5 (2
=
(2
1/5 3/5
12)/5
12)/5
(36-b1)/5
Here b, is such that (2 12)/5 and (36 )/5 are 0, the current basis will be
optimal with the values given by
= (2 12)/5 and
= (36 )/5. This
can happen when b 36. When b is outside this range, one of the RHS values
will be negative. When = 40,
values become
RHS=
2/5 1/5 40 68/5

=
1/5 3/5 12 4/5
We carry out dual Simplex iterations with the starting solution where one of the RHS
values is negative as shown in Table 3.13
10-04-2016
0
RHS
5
4
0
1
1/5
7/5
1
0
2/5
-1/5
-1/5
3/5
-18/5
-6/5
-7/5
68/5
-4/5
6
4
0
2
-5
3
-7
1
10
0
1
1
-3
12
4
-4
48
0
-4
-9
=12, = 4, = 48. This was also obtained from one

iteration from the modified simplex table.
The optimum solution is
Changes in coefficient of constraints (of a non-basic variable)

We consider change in the constraint coefficient of the non basic variable .
2
Considering, column vector corresponding to = . The corresponding column in
the optimum table will be
2/5 1/5 2
(4 )/5
=
1/5 3/5
(3 2)/5
(4 )/5
=3 - 5 4
= (3 7 )/5
(3 2)/5
If the change is such that 0 , the present solution will be optimal. If
b=1/3, then = 2/15 and variable will enter the basis. The value of the
entering column will be ,
2/5 1/5 2
11/15
=
=
=
1/5 3/5 1/3
1/5
=
10-04-2016
0
RHS
5
4
3
4
0
1
11/15
-1/5
1
0
2/5
-1/5
-1/5
3/5
6/5
27/5
2/15
-6/5
-7/5
138/5
0
1
1
0
15/11
3/11
6/11
-1/11
-3/11
6/11
18/11
63/11
-2/11
-14/11 -15/11 306/11
18/11
=18/11, = 63/11 , with Z=306/11. Here

too the optimum solution in obtained in one iteration.
The new optimum solution is given by
Adding a new product

Let us consider a new product to the problem. This is incorporated in the formulation
by the introduction of new variable . assuming this product requires1 unit of
resource 2 per product and fetches a profit of Rs. 6/unit.
This is same as verifying whether
Variable
-y
can enter the basis. Now,

=
1
and
3
=6-27/5=3/5
enters the basis. The modified optimum table is shown and the simplex
iterations are shown in the table below.
10-04-2016
6
RHS
5
4
0
X1
1/5
2/5
-1/5
7/5
-1/5
3/5
8/5
27/5
-18/5 0
-6/5
-7/5
3/5
138/5
1/8
5/8
3/8
7/8
1
0
3/8
-1/8
-1/8
3/8
0
1
15/8
27/8
-3/8
-33/8 0
-9/8
-13/8 0
-1/5 6/5
5
6
237/8
The optimum solution is given by
=15/8,
= 237/8.
Adding a New Constraint

let the new constraint be
8.
The solution = 27/5,

= 6/5 satisfies the constraint and the present solution
continues to be optimal even with the additional constraint.
If the constraint was + +
6, the present optimal solution violates the
constraint. We rewrite the constraint as an equation by adding a slack variable and
write the basic variables in terms of the non basic variables.
+ + +
=8
From the optimum Simplex table, we have
+7
/5 1
/5 + 3
/5 = 27/5
/5
/5 = 6/5
And
+1
/5 +2
Substituting we get
3
1
2
+
5
5
5
+
3
5
=6
line gayab
10-04-2016
(Observe that if the constraint is binding, as in this case, the final equation
after substitution will have a negative value of the RHS when the slack
variable has +1 coefficient). We include this constraint into the optimal
table and proceed with dual simplex iterations till the new optimum is
reached. This is shown in table below.
5
4
0
5
4
0
X1
X6
X5
0
1
0
1/5
7/5
-3/5
1
0
0
2/5
-1/5
-1/5
-1/5
3/5
-2/5
0
0
1
0
0
1
0
0
-18/5
3/2
-3/2
0
1
0
0
0
-6/5
-1/2
1/2
-1/2
-7/5
0
0
1
0
3/5
-1/2
3/2
-5/2
-7/2
RHS
6/5
27/5
-3/5
138/5
3/2
9/2
3/2
51/2

Sensitivity Analysis in Linear Programming

Uploaded by

Copyright:

Available Formats

Sensitivity Analysis in Linear Programming

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Sensitivity Analysis in Linear Programming

Uploaded by

Copyright:

Available Formats

10-04-2016

SENSITIVITY ANALYSIS IN LINEAR PROGRAMMING

This is known as sensitivity analysis or post-optimality analysis.

The optimal solution is, X1=275, X3=65, Z=138/5.

Changes in values of objective function coefficients

value of a non-basic variable

In our example, variable is a non-basic optimum. Let us consider changes in C

alone will change. We compute

The new optimal solution is,

value of a basic variable

Let us consider a change in the objective function coefficient of variable . Let us

10, the present set of basic variable will be optimal. Let us

enters the basis and we perform simplex

= 12) and the final solutions are shown in Table

2 Changes in RHs values

.the change results in change in the RHS values.

2/5 1/5 40 68/5

=12, = 4, = 48. This was also obtained from one

Changes in coefficient of constraints (of a non-basic variable)

-14/11 -15/11 306/11

=18/11, = 63/11 , with Z=306/11. Here

Adding a new product

This is same as verifying whether

can enter the basis. Now,

The optimum solution is given by

Adding a New Constraint

The solution = 27/5,

You might also like