Kathmandu University School of Management

(KUSOM)

Subject:
Optimization Technique for Managerial Decisions

Group Assignment:
Linear Programming Problem

Submitted to:
Dr. Umesh Rajopadhyaya

Submitted by:
Pradeep Poudel(16331)
Pawan Sapkota (16332)
Shivani Shah (16333)
Eman Shakya (16334)
Merina Shrestha (16336)
Kritika Thapa (16337)

Date of Submission:19th July, 2016

QUESTION:
Maximize Z= 20x1 + 30x2
Subject to constraints
2x1 + x2 40.......................................................................................................................(i)
4x1 x2 20......................................................................................................................(ii)
x1 30..............................................................................................................................(iii)
x1, x2 0

SOLUTION,
Converting the above constraints (i), (ii) and (iii) into equations form, we have,
From (i),
2x1 + x2 + S1 = 40
From (ii),
4x1 x2 + S2 = 20
From (iii),
x1 S3 + A1 = 30
where, S1, S2 and S3 are slack variables and A1 is artificial variable.
x1, x2, S1, S2, S3, A1 0
Maintaining equal number of variables in the above equations,
The objective function becomes,

Z = 20x1 + 30x2 + 0S3 + 0S1+ 0S2 MA1

2x1 + x2 + 0S3 + 1S1 + 0S2 + 0A1 = 40

4x1 x2 + 0S3 + 0S1 + 1S2 + 0A1 = 20
x1 + 0x2 1S3 + 0S1 + 0S2 + 1A1 = 30
Representing the above equations in the matrix form of AX= B,

2 1 0 10 0
4 1 0 0 1 0
1 0 1 0 0 1

[]
x1
x2
S3
S1
S2
A1

[]
40
20
30

Constructing the first simplex tableau

CB
0
0
-M

Cj
Basis

20
x1

30
x2

0
S3

0
S1

0
S2

-M
A1

S1
S2
A1
Zj
Cj Z j

2
4
1
-M
20 + M

1
-1
0
0
30

0
0
-1
M
-M

1
0
0
0
0

0
1
0
0
0

0
0
1
-M
0

Ration
bi
40
20
30
-30M

Colum
n
20
5
30

Since all Cj-Zj are neither zero nor negative, the optimal solution is not obtained, so we look
for the next simplex tableau.
Since (20 + M) is the greatest number, so the corresponding column is the key column and
the corresponding variable is the entering variable.
Calculating the values of ratio column by dividing the values in bi column by the
corresponding numbers in the key column.

Since 5 is the smallest number among the non-negative numbers, the corresponding row is
the pivot row and the corresponding variable is the leaving variable.
Pivot number = 4
Entering variable = x1
Leaving variable = S2
For the 2nd simplex tableau,
New values for the main row are calculated by dividing the old values of the row by the pivot
number.
4
4

= 1,

1
4

= -0.25,

0
4

= 0,

0
4

= 0,

1
4

= 0.25,

0
4

=0

Values for column S1 are,

Number
in new

row
2
1
0
1
0
0
40

Number above or
below pivot number
2
2
2
2
2
2
2

Corresponding number in
new pivot row

Number
=

is new
row
0
1.5
0
1
-0.5
0
30

1
-0.25
0
0
0.25
0
5

Values for column S2 are,

Number
in new
row
1
0
-1
0
0
1
30

Number above or
below pivot number
1
1
1
1
1
1
1

Corresponding number in
new pivot row
1
-0.25
0
0
0.25
0
5

Number
=

is new
row
0
0.25
-1
0
-0.25
1
25

Constructing next simplex tableau,

Cj
Basis

20
x1

30
x2

0
S3

0
S1

0
S2

-M
A1

S1
x1
A1
Zj
Cj Zj

0
1
0
20
0

1.5
-0.25
0.25
-5 - 0.25M
35 + 0.25M

0
0
-1
M
-M

1
0
0
0
0

-0.5
0.25
-0.25
5 + 0.25M
-5 - 0.25M

0
0
1
-M
0

CB
0
20
-M

Ration
bi
30
5
25
100 - 25M

Column
20
-20
100

Since all Cj-Zj are neither zero nor negative, the optimal solution is not obtained, so we look
for the next simplex tableau.
Since (35 + 0.25M) is the greatest number among (Cj Zj), the corresponding column is the
key column and the corresponding variable is the entering variable.
Since 20 is the smallest non-negative number in the ratio column, the corresponding row is
the pivot row and the corresponding variable is the leaving variable.
Pivot number = 1.5
Entering variable = x2
Leaving variable = S1
For next simplex tableau,
New values for the main row are calculated by dividing the old values of the row by the pivot
number.
0
1.5

= 0,

1.5
1.5

= 1,

0
1.5

= 0,

1
1.5

2
0.5
3 , 1.5

1
= 3 ,

0
1.5

= 0,

30
1.5

= 20
Values for column x1 are,
Number
in new

Number above or
below pivot number

Corresponding number in

new pivot row

= Number
is new

row
1
-0.25
0
0

-0.25
-0.25
-0.25
-0.25

0
1
0
2
3

row
1
0
0
1
6

0.25

-0.25

1
3

1
6

0
5

-0.25
-0.25

0
20

0
10

Values for column A1 are,

Number
in new

Number above or

below pivot number

row
0
0.25
-1
0

Corresponding number in
new pivot row

Number
=

is new

0.25
0.25
0.25
0.25

0
1
0
2
3

row
0
0
-1
1
6

-0.25

0.25

1
3

1
6

1
25

0.25
0.25

0
20

1
20

Constructing next simplex tableau,

CB

Cj
Basis

20
x1

30
x2

0
S3

0
S1

0
S2

-M
A1

bi

Ration

30

x2

2
3

1
3

20

20

x1

1
6

1
6

10

Column

-M

A1

-1

1
6

1
6

20

Zj

20

30

140+ M
6

40+ M
6

-M

780 - 20M

Cj Zj

-M

140+ M
6

40+ M
6

Since all Cj-Zj are either zero or negative, the optimal solution is achieved.
x1 = 20
x2 = 10
and profit Z = 780 20M

Type of solution
Since the basic variables in the final simplex tableau contains the artificial variable A1, this
linear programming problem has infeasible solution.

Implications
Feasible solution cannot be obtained for this problem due to the constraints which are
mutually exclusive to each other. To obtain proper optimum solution, the constraints need to
be revised.