Sensitivity Analysis

A linear programming problem is characterised by

following:
• Objective Function
• RHS value
• Co-efficient matrix

Maximize Z= 4X1 + 3 X2 + 5X3

s.t.
X1 + 2 X2 + 3X3 ≤ 9
2X1 + 3 X2 + X3 ≤ 12
X 1 , X 2 , X3 ≤ 0
 After solving LPP, we can analyse Maximize Z = 4 X 1 + 3 X 2 + 5 X 3
s.t.
following change in resources for X1 + 2 X 2 + 3 X 3  9
above cases. 2 X 1 + 3 X 2 + X 3  12
X1, X 2 , X 3  0
1. Change in objective function for non-basic variables
2. Change in objective function for basic variables
3. Change in RHS value
4. Change in constraint coefficient – non-basic variable
5. Change in constraint coefficient – basic variable
6. Add a variable
7. Add a constraint
 Example
Maximize Z = 4 X 1 + 3 X 2 + 5 X 3
s.t.
X1 + 2 X 2 + 3 X 3  9
2 X 1 + 3 X 2 + X 3  12
X1, X 2 , X 3  0
Maximize Z = 4 X 1 + 3 X 2 + 5 X 3 + 0 S1 + 0 S 2
Changing s.t.
Constraint X 1 + 2 X 2 + 3 X 3 + S1 = 9
into
equation, 2 X 1 + 3 X 2 + X 3 + S 2 = 12
we have X1, X 2 , X 3  0
 Cj 4 3 5 0 0
CB
Soluti
BV X1 X2 X3 S1 S2 Ratio
on
0 S1 1 2 3 1 0 9 3
0 S2 2 3 1 0 1 12 12

Cj - Zj 4 3 5 0 0 Zmax=0
 Cj 4 3 5 0 0
CB
BV X1 X2 X3 S1 S2 Solution Ratio

5 X3 1/3 2/3 1 1/3 0 3 9

0 S2 5/3 7/3 0 -1/3 1 9 27/5

Cj - Zj 7/3 -1/3 0 -5/3 0 Zmax=15

 Cj 4 3 5 0 0
CB
BV X1 X2 X3 S1 S2 Solution Ratio

5 X3 0 1/5 1 2/5 -1/5 6/5

4 X1 1 7/5 0 -1/5 3/5 27/5

Cj - Zj 0 -18/5 0 -6/5 -7/5 Zmax=138/5

 Case 1: Change in objective function for non-
basic variables
Cj 4 3 5 0 0
CB
BV X1 X2 X3 S1 S2 Solution Ratio
5 X3 0 1/5 1 2/5 -1/5 6/5
4 X1 1 7/5 0 -1/5 3/5 27/5
Cj - Zj 0 -18/5 0 -6/5 -7/5 Zmax=138/5

• Basic Variable: X1, X3

• Non Basic Variable : X2, S1, S2
Here non basic variables are X2, S1 and S2. Let Co-efficient of X2
Changes from 3 to C2

C2 – Z2 = C2 – (1 + 28/5) = C2 – 33/5 ≤ 0

or, C2 ≤ 33/5 for optimal solution. So, we should take C2 > 33/5
or 6.6
Cj 4 C2 5 0 0
CB
BV X1 X2 X3 S1 S2 Solution Ratio
5 X3 0 1/5 1 2/5 -1/5 6/5
4 X1 1 7/5 0 -1/5 3/5 27/5
Cj - Zj 0 C2 – 33/5 0 -6/5 -7/5 Zmax=138/5
 So we consider value of C2 as 7.

Cj 4 7 5 0 0
CB
BV X1 X2 X3 S1 S2 Solution Ratio
5 X3 0 1/5 1 2/5 -1/5 6/5 6
4 X1 1 7/5 0 -1/5 3/5 27/5 27/7
Cj - Zj 0 2/5 0 -6/5 -7/5

Cj 4 7 5 0 0
CB
BV X1 X2 X3 S1 S2 Solution Ratio
5 X3 -1/7 0 1 3/7 -2/7 3/7
7 X2 5/7 1 0 -1/7 3/7 27/7
Cj - Zj -2/7 0 0 -8/7 -11/7 Zmax=204/7
 With C2 = 3; Zmax=138/5

With C2 = 7; Zmax=204/7

Solution optimized by Zmax=204/7 - 138/5= 54/35

 Case 2: Change in objective function for basic
variables
Cj 4 3 5 0 0
CB
BV X1 X2 X3 S1 S2 Solution Ratio
5 X3 0 1/5 1 2/5 -1/5 6/5
4 X1 1 7/5 0 -1/5 3/5 27/5
Cj - Zj 0 -18/5 0 -6/5 -7/5 Zmax=138/5

• Non Basic Variable : X2, S1, S2

• Basic Variable: X1, X3
 Co-efficient of Basic variable X1 changes to C1, Causing
change in value of Cj - Zj

Cj C1 3 5 0 0
CB
BV X1 X2 X3 S1 S2 Solution Ratio
5 X3 0 1/5 1 2/5 -1/5 6/5
C1 X1 1 7/5 0 -1/5 3/5 27/5
Cj - Zj 0 -18/5 0 -6/5 -7/5
C2 – Z2 C4 – Z4 C5 – Z5
C2 – Z2 = 3 – (1 + 7 C1/5) = 2 - 7 C1/5 ≤ 0 or C1 ≥ 10/7
C4 – Z4 = 0 – (2 – C1/5) = C1 /5 – 2 ≤ 0 or C1 ≤ 10
C5 – Z5 = 0 – (-1 + 3C1/5) = 1 – 3C1/5 ≤ 0 or C1 ≥ 5/3
Therefore, optimal condition will not be violated for
10/7 ≤ C1 ≤ 10
So, Let C1=12, which will make S1 having positive value of Cj-Zj.
 Cj 12 3 5 0 0
CB
BV X1 X2 X3 S1 S2 Solution Ratio
5 X3 0 1/5 1 2/5 -1/5 6/5
12 X1 1 7/5 0 -1/5 3/5 27/5
New New New
Cj - Zj 0 0
Value Value Value
C2 – Z2 C4 – Z4 C5 – Z5

We get X1=6, S1=3 and Zmax=72

Case 3: Change in RHS value of constraints
Maximize Z = 4 X 1 + 3 X 2 + 5 X 3
Given Problem is
s.t.
X1 + 2 X 2 + 3 X 3  9
2 X 1 + 3 X 2 + X 3  12
X1, X 2 , X 3  0
 Cj 4 3 5 0 0
CB
Solutio
BV X1 X2 X3 S1 S2 Ratio
n
5 X3 0 1/5 1 2/5 -1/5 6/5
4 X1 1 7/5 0 -1/5 3/5 27/5
Cj - Zj 0 -18/5 0 -6/5 -7/5 138/5
 Cj 4 3 5 0 0
CB
BV X1 X2 X3 S1 S2 Solution Ratio
5 X3 0 1/5 1 2/5 -1/5 68/5
4 X1 1 7/5 0 -1/5 3/5 -4/5
Cj - Z j 0 -18/5 0 -6/5 -7/5
Negative value in RHS means we have an infeasible solution. i.e. primal (RHS) is infeasible but
the duel (Cj - Zj) is feasible. Therefore, we have to perform duel simplex iteration till we get
basic feasible solution.
So we leave out infeasible RHS and compute Ratio as positive value of ( j j ) infeasible row value
C −Z

Cj 4 3 5 0 0
CB
BV X1 X2 X3 S1 S2 Solution Ratio
5 X3 0 1/5 1 2/5 -1/5 68/5
4 X1 1 7/5 0 -1/5 3/5 -4/5
Cj - Z j 0 -18/5 0 -6/5 -7/5
Ratio - 6 -
So Outgoing variable is X1and incoming variable is S1.
Solution is X3=12, S1=4, and Zmax =48.
Case 4: Change in constraint coefficient – non-
basic variable

Maximize Z = 4 X 1 + 3 X 2 + 5 X 3
s.t.
X1 + 2 X 2 + 3 X 3  9
2 X 1 + 3 X 2 + X 3  12
X1, X 2 , X 3  0
 Cj 4 3 5 0 0
CB
BV X1 X2 X3 S1 S2 Solution Ratio

5 X3 0 1/5 1 2/5 -1/5 6/5

4 X1 1 7/5 0 -1/5 3/5 27/5

Cj - Z j 0 -18/5 0 -6/5 -7/5 138/5

• Non basic Variables are X2, S1, S2

• Let Co-efficient of X2 in 2nd constraint changes from 3 to a.
 Cj 4 3 5 0 0
CB
BV X1 X2 X3 S1 S2 Solution Ratio
5 X3 0 4/5 1 2/5 -1/5 6/5
4 X1 1 -2/5 0 -1/5 3/5 27/5
Cj - Zj 0 3/5 0 -6/5 -7/5

The solution obtained is X1=6, X2=3/2 and Zmax=57/2

 Case 5: Change in constraint coefficient – basic
variable

Solve the new problem again considering it a fresh new Linear

Programming Problem.
Case 6: Add a new variable (Addition of new product)
Maximize Z = 4 X 1 + 3 X 2 + 5 X 3
s.t.
X1 + 2 X 2 + 3 X 3  9
2 X 1 + 3 X 2 + X 3  12
X1, X 2 , X 3  0

Cj 4 3 5 0 0
CB
BV X1 X2 X3 S1 S2 Solution Ratio
5 X3 0 1/5 1 2/5 -1/5 6/5
4 X1 1 7/5 0 -1/5 3/5 27/5
C -Z 0 -18/5 0 -6/5 -7/5 138/5
The original problem has 3 variables. Let suppose a new product has been
introduced/added. The product added is X4 and its coefficients in objective
function is 6 and value of P is 1 & 3 respectively.

1
i.e. C4=6 and P4 =  
3

Maximize Z = 4 X 1 + 3 X 2 + 5 X 3 + 6 X 4
s.t.
X1 + 2 X 2 + 3 X 3 + X 4  9
2 X 1 + 3 X 2 + X 3 + 3 X 4  12
X1, X 2 , X 3 , X 4  0
 Cj 4 3 5 6 0 0
CB
BV X1 X2 X3 X4 S1 S2 Solution Ratio
5 X3 0 1/5 1 2/5 -1/5 6/5
4 X1 1 7/5 0 -1/5 3/5 27/5
Cj - Zj 0 -18/5 0 -6/5 -7/5 138/5
 Cj 4 3 5 6 0 0
CB
BV X1 X2 X3 X4 S1 S2 Solution Ratio
5 X3 0 1/5 1 -1/5 2/5 -1/5 6/5 -
4 X1 1 7/5 0 8/5 -1/5 3/5 27/5 27/5
Cj - Zj 0 -18/5 0 3/5 -6/5 -7/5 138/5

The optimal solution obtained is X3=15/8, X4=27/8, Zmax=237/8

 Case 7: Add a new constraint
Maximize Z = 4 X 1 + 3 X 2 + 5 X 3
s.t.
X1 + 2 X 2 + 3 X 3  9
2 X 1 + 3 X 2 + X 3  12
X1, X 2 , X 3  0

Cj 4 3 5 0 0
CB
BV X1 X2 X3 S1 S2 Solution Ratio
5 X3 0 1/5 1 2/5 -1/5 6/5
4 X1 1 7/5 0 -1/5 3/5 27/5
Cj - Zj 0 -18/5 0 -6/5 -7/5 138/5
 Maximize Z = 4 X 1 + 3 X 2 + 5 X 3
s.t.
X1 + 2 X 2 + 3 X 3  9
2 X 1 + 3 X 2 + X 3  12
X1 + X 2 + X 3  8
X1, X 2 , X 3  0

In this condition, if we see the optimal solution obtained when X1=27/5 and X3=6/5

We have X1 + X2 + X3 = 27/5 + 0 + 6/5 = 33/5 which is less than 8.

So we don’t need to evaluate further for this addition of constraint, it will not have
impact on optimal solution and it will not provide us better optimal solution.
 Maximize Z = 4 X 1 + 3 X 2 + 5 X 3
s.t.
X1 + 2 X 2 + 3 X 3  9
2 X 1 + 3 X 2 + X 3  12
X1 + X 2 + X 3  6
X1, X 2 , X 3  0

In this condition, if we see the optimal solution obtained when X1=27/5 and X3=6/5

We have X1 + X2 + X3 = 27/5 + 0 + 6/5 = 33/5 which is not less than 6.

So we need to evaluate further for this addition of constraint, as it will have impact
on optimal solution.
 Cj 4 3 5 0 0
CB
BV X1 X2 X3 S1 S2 Solution Ratio
5 X3 0 1/5 1 2/5 -1/5 6/5
4 X1 1 7/5 0 -1/5 3/5 27/5
Cj - Zj 0 -18/5 0 -6/5 -7/5 138/5

we need to write basic variable in terms of non basic variable.

i.e. we have write equation from the old optimal table obtained.
X1 + X 2 + X 3  6
 27 7 1 3  6 1 2 1 
  − X 2 + S1 − S 2  + X 2 +  − X 2 − S1 + S 2   6
5 5 5 5  5 5 5 5 
3 1 2 3
 − X 2 − S1 − S 2  −
5 5 5 5
 3 1 2 3
− X 2 − S1 − S 2 + S3 = −
5 5 5 5
Cj 4 3 5 0 0 0
CB
BV X1 X2 X3 S1 S2 S3 Solution Ratio
5 X3 0 1/5 1 2/5 -1/5 0 6/5
4 X1 1 7/5 0 -1/5 3/5 0 27/5
0 S3 0 -3/5 0 -1/5 -2/5 1 -3/5
Cj - Z j 0 -18/5 0 -6/5 -7/5 0 138/5
Ratio 6 6 7/2

The optimal solution is obtained in one iteration and is given by X1=9/2,

X3=3/2, S2=3/2, Zmax=51/2