Quantitative Analysis for

Management
Goa Institute of Management
PGDM-FT 2024-2025

Dr. Deepti Mohan

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

Tie for entering variable
• Tie for entering variable – break arbitrarily
𝒄𝒋 10 10 0 0 0
𝒄𝑩 𝒙𝑩 𝒙𝟏 𝒙𝟐 𝒔𝟏 𝒔𝟐 𝒔𝟑 RHS

0 𝒔𝟏 2 1 1 0 0 50
0 𝒔𝟐 2 5 0 1 0 100
0 𝒔𝟑 2 3 0 0 1 90
𝑧𝑗 0 0 0 0 0 0
𝑐𝑗 − 𝑧𝑗 10 10 0 0 0 0

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

Tie for leaving variable

• Tie for leaving variable - again break arbitrarily!

𝒄𝒋 -1 -1 4 0 0 0
𝒄𝑩 𝒙𝑩 𝒙𝟏 𝒙𝟐 𝒙𝟑 𝒔𝟏 𝒔𝟐 𝒔𝟑 RHS Ratio

0 𝒔𝟏 1 1 2 1 0 0 8 4
0 𝒔𝟐 1 1 -1 0 1 0 2 -
0 𝒔𝟑 -1 1 1 0 0 1 4 4
𝑧𝑗 0 0 0 0 0 0 0
𝒄𝒋 -𝑧𝑗 -1 -1 4 0 0 0

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

LPP cases for Simplex Method
Case How to identify in Simplex Method
Unique optimal solution For a maximization (minimization) objective, the algorithm terminates with negative
(positive) cj-zj values for all non-basic variables
Multiple optimal solutions Algorithm terminates with one or more non-basic variable(s) having zero cj-zj value.
Unbounded Algorithm does not terminate but no basic variable is able to leave. (i.e. there will be
an entering variable but no leaving variable)
No feasible solution ?

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

 Unique optimal solution
𝒄𝒋 40 10 0 0 0
𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒
𝑍 = 40𝑥1 + 10𝑥2 + 0𝑠1 + 0𝑠2 + 0𝑠3 𝒄𝑩 𝒙𝑩 𝒙𝟏 𝒙𝟐 𝒔𝟏 𝒔𝟐 𝒔𝟑 RHS Ratio
𝑠. 𝑡.
2𝑥1 + 𝑥2 + 𝑠1 = 50
N N B B B
2𝑥1 + 5𝑥2 + 𝑠2 = 100 0 𝒔𝟏 2 1 1 0 0 50 25
2𝑥1 + 3𝑥2 + 𝑠3 = 90
& 𝑥1, 𝑥2 , 𝑠1 , 𝑠2 , 𝑠3 ≥ 0 0 𝒔𝟐 2 5 0 1 0 100 50
0 𝒔𝟑 2 3 0 0 1 90 45
𝑧𝑗 0 0 0 0 0 0
𝑐𝑗 − 𝑧𝑗 40 10 0 0 0 0
𝒄𝒋 40 10 0 0 0
𝒄𝑩 𝒙𝑩 𝒙𝟏 𝒙𝟐 𝒔𝟏 𝒔𝟐 𝒔𝟑 RHS Ratio
B N N B B
40 𝒙𝟏 1 1/2 1/2 0 0 25
0 𝒔𝟐 0 4 -1 1 0 50
0 𝒔𝟑 0 2 -1 0 1 40
𝑧𝑗 40 20 20 0 0 1000
𝑐𝑗 − 𝑧𝑗 0 -10 -20 0 0
05-11-2024 Dr. Deepti Mohan | Goa Institute of Management
Unique optimal solution to the LPP

There is only single (unique) optimal solution to the LPP if the

algorithm terminates and none of the non-basic variable have
zero cj-zj value when the algorithm terminates (all are positive for
minimization problem and all are negative for maximization
problem).

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

 Multiple optimal solutions
𝒄𝒋 20 10 0 0 0
𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒
𝑍 = 20𝑥1 + 10𝑥2 + 0𝑠1 + 0𝑠2 + 0𝑠3 𝒄𝑩 𝒙𝑩 𝒙𝟏 𝒙𝟐 𝒔𝟏 𝒔𝟐 𝒔𝟑 RHS Ratio
𝑠. 𝑡.
2𝑥1 + 𝑥2 + 𝑠1 = 50
N N B B B
2𝑥1 + 5𝑥2 + 𝑠2 = 100 0 𝒔𝟏 2 1 1 0 0 50 25
2𝑥1 + 3𝑥2 + 𝑠3 = 90
& 𝑥1, 𝑥2 , 𝑠1 , 𝑠2 , 𝑠3 ≥ 0 0 𝒔𝟐 2 5 0 1 0 100 50
0 𝒔𝟑 2 3 0 0 1 90 45
𝑧𝑗 0 0 0 0 0 0
𝑐𝑗 − 𝑧𝑗 20 10 0 0 0 0
𝒄𝒋 20 10 0 0 0
𝒄𝑩 𝒙𝑩 𝒙𝟏 𝒙𝟐 𝒔𝟏 𝒔𝟐 𝒔𝟑 RHS Ratio
B N N B B
20 𝒙𝟏 1 1/2 1/2 0 0 25
0 𝒔𝟐 0 4 -1 1 0 50
0 𝒔𝟑 0 2 -1 0 1 40
𝑧𝑗 20 10 10 0 0 500
𝑐𝑗 − 𝑧𝑗 0 0 -10 0 0
05-11-2024 Dr. Deepti Mohan | Goa Institute of Management
Multiple solutions/Alternate optima

Multiple solutions/Alternate optima exists when there is at least one

non-basic variable with zero cj-zj value when the algorithm
terminates.

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

 Unbounded optimal solution
Iteration-1 𝒄𝒋 5 4 0 0
Maximize
𝒄Maximize
𝑩 𝒙𝑩 𝒙𝟏 𝒙𝟐 𝒔𝟏 𝒔𝟐 RHS Ratio
Z = 5x1 + 4x2 + 0s1 + 0s2 N N B B
Z = 5x1 + 4x2 + 0s1 + 0s2
subject to 0 to 𝒔𝟏 1 0 1 0 7 7
subject
x1 + s1 =7 0 + s1 =7 𝒔𝟐 1 -1 0 1 8 8
x1
x1 - x2 + s2 = 8 x1 - x2 + s2 = 8 𝑧𝑗 0 0 0 0 0
& x1, x2, s1, s2 ≥0 & x1, x2, s1, s2𝑐𝑗≥0 5 4 0 0
-𝑧𝑗
Iteration-2 𝒄𝒋 5 4 0 0
In the second iteration,
𝒄𝑩 𝒙𝑩 𝒙𝟏 𝒙𝟐 𝒔𝟏 𝒔𝟐 RHS Ratio
𝑥2 is the entering variable but B N N B
there is no leaving variable!
5 𝒙𝟏 1 0 1 0 7 -
In the pivot column, 0 𝒔𝟐 0 -1 -1 1 1 -
all the coefficients 𝑧𝑗 5 0 5 0 35
are zero or negative
𝑐𝑗 -𝑧𝑗 0 4 -5 0

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

Unbounded optimal solution

The algorithm will not terminate. But, it reaches an iteration

where there is an entering variable but no corresponding leaving
variable (in the column against the entering variable, all the
coefficients will be zero or negative and hence no ratio can be
computed).

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

LPP cases for Simplex Method
Case How to identify in Simplex Method
Unique optimal solution For a maximization (minimization) objective, the algorithm terminates with negative
(positive) cj-zj values for all non-basic variables
Multiple optimal solutions Algorithm terminates with one or more non-basic variable(s) having zero cj-zj value.
Unbounded Algorithm does not terminate but no basic variable is able to leave. (i.e. there will be
an entering variable but no leaving variable)
No feasible solution Occurs only in BigM type LPP problems

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

 Big M Method

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

Example-1

Minimize Z = 4x1 + x2
S.t.
3x1 + x2 = 3
4x1 + 3x2 ≥ 6
x1 + 2x2 ≤ 3
& x1, x2 ≥0

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

Example-1
You need to introduce an artificial
Minimize Z = 4x1 + x2 Standard form: variable in each equation without a
slack variable.
S.t. Minimize Z = 4x1 + x2
3x1 + x2 = 3 S.t. Minimize Z = ?
4x1 + 3x2 ≥ 6 3x1 + x2 = 3 S.t.
x1 + 2x2 ≤ 3 4x1 + 3x2 –u = 6 3x1 + x2 + a1 = 3
& x1, x2 ≥0 x1 + 2x2 +s = 3 4x1 + 3x2 –u + a2 = 6
& x1, x2 , u, s ≥0 x1 + 2x2 + s = 3
& x1, x2 , u, s, a1, a2 ≥0

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

• A feasible solution to the newly created problem is a feasible solution
to the original problem only if the values of artificial variables are
zero.
• Since it is a minimization problem, we heavily penalize any non-zero
value of artificial variable by making its coefficient +M in the objective
function, where M is a very large positive number.
• If it is a maximization problem, we heavily penalize any non-zero
value of artificial variable by making its coefficient -M in the objective
function, where M is a very large positive number.

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

Example-1Introducing artificial variables
Minimize Z = 4x1 + x2 Standard form:
After introducing artificial variables for
S.t. Minimize Z = 4x1 + x2 applying Big M method:
3x1 + x2 = 3 S.t. Minimize Z = 4x1 + x2 +M a1 +M a2

4x1 + 3x2 ≥ 6 3x1 + x2 = 3 S.t.

3x1 + x2 + a1 = 3
x1 + 2x2 ≤ 3 4x1 + 3x2 –u = 6
4x1 + 3x2 –u + a2 = 6
& x1, x2 ≥0 x1 + 2x2 +s = 3
x1 + 2x2 + s = 3
& x1, x2 , u, s ≥0
& x1, x2 , u, s, a1, a2 ≥0

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

Example -1
Minimize Z = 4x1 + x2 Standard form:
After introducing artificial variables for
S.t. Minimize Z = 4x1 + x2 applying Big M method:
3x1 + x2 = 3 S.t. Minimize Z = 4x1 + x2 +M a1 +M a2

4x1 + 3x2 ≥ 6 3x1 + x2 = 3 S.t.

3x1 + x2 + a1 = 3
x1 + 2x2 ≤ 3 4x1 + 3x2 –u = 6
4x1 + 3x2 –u + a2 = 6
& x1, x2 ≥0 x1 + 2x2 +s = 3
x1 + 2x2 + s = 3
& x1, x2 , u, s ≥0
& x1, x2 , u, s, a1, a2 ≥0

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

 Minimize Z = 4x1 + x2 +M a1 +M a2

S.t.

3x1 + x2 + a1 = 3

4x1 + 3x2 –u + a2 = 6

x1 + 2x2 + s = 3
• All artificial
& x1, x2 , u, s, a1, a2 ≥0
variables and
slack variables
Iteration-1 𝒄𝒋
will be the basic
variables in the 𝒄𝑩 𝒙𝑩 𝒙𝟏 𝒙𝟐 𝒖 𝒔 𝒂𝟏 𝒂𝟐 RHS Ratio
first iteration. N N N B B B
𝒂𝟏
𝒂𝟐
𝒔
𝑧𝑗
𝑐𝑗 -𝑧𝑗

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

 Minimize Z = 4x1 + x2 +M a1 +M a2

S.t.

3x1 + x2 + a1 = 3

4x1 + 3x2 –u + a2 = 6

• Iteration 1: x1 + 2x2 + s = 3

• Solution? & x1, x2 , u, s, a1, a2 ≥0

• Entering Iteration-1 𝒄𝒋 4 1 0 0 M M
variable =?
𝒄𝑩 𝒙𝑩 𝒙𝟏 𝒙𝟐 𝒖 𝒔 𝒂𝟏 𝒂𝟐 RHS Ratio
Leaving
variable =? M 𝒂𝟏 3 1 0 0 1 0 3 1
M 𝒂𝟐 4 3 -1 0 0 1 6 1.5
0 𝒔 1 2 0 1 0 0 3 3
𝑧𝑗 7M 4M -M 0
𝑐𝑗 -𝑧𝑗 4-7M 1-4M M 0

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

 Minimize Z = 4x1 + x2 +M a1 +M a2 Iteration-1 𝒄𝒋 4 1 0 0 M M
S.t.
𝒄𝑩 𝒙𝑩 𝒙𝟏 𝒙𝟐 𝒖 𝒔 𝒂𝟏 𝒂𝟐 RHS Ratio
3x1 + x2 + a1 = 3

4x1 + 3x2 –u + a2 = 6 M 𝒂𝟏 3 1 0 0 1 0 3 1
x1 + 2x2 + s = 3 M 𝒂𝟐 4 3 -1 0 0 1 6 1.5
& x1, x2 , u, s, a1, a2 ≥0
0 𝒔 1 2 0 1 0 0 3 3
𝑧𝑗 7M 4M -M 0
• Iteration 1:
• Entering variable =x1 𝑐𝑗 -𝑧𝑗 4-7M 1-4M M 0
• Leaving variable =a1 Iteration-2 𝒄𝒋 4 1 0 0 M M
𝒄𝑩 𝒙𝑩 𝒙𝟏 𝒙𝟐 𝒖 𝒔 𝒂𝟏 𝒂𝟐 RHS Ratio

• Iteration 2:
4 𝒙𝟏
• Entering variable =?
• Leaving variable =?
M 𝒂𝟐
0 𝒔
𝑧𝑗
𝑐𝑗 -𝑧𝑗

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

 Minimize Z = 4x1 + x2 +M a1 +M a2 Iteration-1 𝒄𝒋 4 1 0 0 M M
S.t.
𝒄𝑩 𝒙𝑩 𝒙𝟏 𝒙𝟐 𝒖 𝒔 𝒂𝟏 𝒂𝟐 RHS Ratio
3x1 + x2 + a1 = 3

4x1 + 3x2 –u + a2 = 6 M 𝒂𝟏 3 1 0 0 1 0 3 1
x1 + 2x2 + s = 3 M 𝒂𝟐 4 3 -1 0 0 1 6 1.5
& x1, x2 , u, s, a1, a2 ≥0
0 𝒔 1 2 0 1 0 0 3 3
𝑧𝑗 7M 4M -M 0
• Iteration 1:
• Entering variable =x1 𝑐𝑗 -𝑧𝑗 4-7M 1-4M M 0
• Leaving variable =a1 Iteration-2 𝒄𝒋 4 1 0 0 M M
𝒄𝑩 𝒙𝑩 𝒙𝟏 𝒙𝟐 𝒖 𝒔 𝒂𝟏 𝒂𝟐 RHS Ratio

• Iteration 2:
4 𝒙𝟏 1 1/3 0 0 1/3 0 1 3
• Entering variable =?
• Leaving variable =?
M 𝒂𝟐 0 5/3 -1 0 -4/3 1 2 6/5
0 𝒔 0 5/3 0 1 -1/3 0 2 6/5
𝑧𝑗 4 4/3+(5/3)M -M 0 -
𝑐𝑗 -𝑧𝑗 0 -1/3-(5/3)M M 0

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

Tie between a slack and an artificial variable

When there is a tie between slack and artificial variables to

leave the basis, the preference shall be given to the artificial
variable to leave the basis.

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

 Minimize Z = 4x1 + x2 +M a1 +M a2 Iteration-1 𝒄𝒋 4 1 0 0 M M
S.t.
𝒄𝑩 𝒙𝑩 𝒙𝟏 𝒙𝟐 𝒖 𝒔 𝒂𝟏 𝒂𝟐 RHS Ratio
3x1 + x2 + a1 = 3

4x1 + 3x2 –u + a2 = 6 M 𝒂𝟏 3 1 0 0 1 0 3 1
x1 + 2x2 + s = 3 M 𝒂𝟐 4 3 -1 0 0 1 6 1.5
& x1, x2 , u, s, a1, a2 ≥0
0 𝒔 1 2 0 1 0 0 3 3
𝑧𝑗 7M 4M -M 0
• Iteration 1:
• Entering variable =x1 𝑐𝑗 -𝑧𝑗 4-7M 1-4M M 0
• Leaving variable =a1 Iteration-2 𝒄𝒋 4 1 0 0 M M
𝒄𝑩 𝒙𝑩 𝒙𝟏 𝒙𝟐 𝒖 𝒔 𝒂𝟏 𝒂𝟐 RHS Ratio

• Iteration 2:
4 𝒙𝟏 1 1/3 0 0 1 3
• Entering variable =?
• Leaving variable =?
M 𝒂𝟐 0 5/3 -1 0 2 6/5
0 𝒔 0 5/3 0 1 2 6/5
𝑧𝑗 4 4/3+(5/3)M -M 0 -
𝑐𝑗 -𝑧𝑗 0 -1/3-(5/3)M M 0

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

 Minimize Z = 4x1 + x2 +M a1 +M a2
Iteration-2 𝒄𝒋 4 1 0 0 M M
S.t.

3x1 + x2 + a1 = 3 𝒄𝑩 𝒙𝑩 𝒙𝟏 𝒙𝟐 𝒖 𝒔 𝒂𝟏 𝒂𝟐 RHS Ratio

4x1 + 3x2 –u + a2 = 6 4 𝒙𝟏 1 1/3 0 0 1 3
x1 + 2x2 + s = 3
M 𝒂𝟐 0 5/3 -1 0 2 6/5
& x1, x2 , u, s, a1, a2 ≥0
0 𝒔 0 5/3 0 1 2 6/5
• Iteration 2:
• Entering variable =x2
𝑧𝑗 4 4/3+(5/3)M -M 0 -
• Leaving variable =a2 𝑐𝑗 -𝑧𝑗 0 -1/3-(5/3)M M 0

Iteration-3 𝒄𝒋 4 1 0 0 M M
𝒄𝑩 𝒙𝑩 𝒙𝟏 𝒙𝟐 𝒖 𝒔 𝒂𝟏 𝒂𝟐 RHS Ratio
• Iteration 3:
• Entering variable =? 4 𝒙𝟏
• Leaving variable =?
1 𝒙𝟐
0 𝒔
𝑧𝑗
𝑐𝑗 -𝑧𝑗

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

 Minimize Z = 4x1 + x2 +M a1 +M a2
Iteration-2 𝒄𝒋 4 1 0 0 M M
S.t.

3x1 + x2 + a1 = 3 𝒄𝑩 𝒙𝑩 𝒙𝟏 𝒙𝟐 𝒖 𝒔 𝒂𝟏 𝒂𝟐 RHS Ratio

4x1 + 3x2 –u + a2 = 6 4 𝒙𝟏 1 1/3 0 0 1 3
x1 + 2x2 + s = 3
M 𝒂𝟐 0 5/3 -1 0 2 6/5
& x1, x2 , u, s, a1, a2 ≥0
0 𝒔 0 5/3 0 1 2 6/5
• Iteration 2:
• Entering variable =x2
𝑧𝑗 4 4/3+(5/3)M -M 0 -
• Leaving variable =a2 𝑐𝑗 -𝑧𝑗 0 -1/3-(5/3)M M 0

Iteration-3 𝒄𝒋 4 1 0 0 M M
𝒄𝑩 𝒙𝑩 𝒙𝟏 𝒙𝟐 𝒖 𝒔 𝒂𝟏 𝒂𝟐 RHS Ratio
• Iteration 3:
• Entering variable =u 4 𝒙𝟏 1 0 1/5 0 3/5 3
• Leaving variable =s
1 𝒙𝟐 0 1 -3/5 0 6/5
0 𝒔 0 0 1 1 0 0
𝑧𝑗 4 1 1/5 0 18/5
𝑐𝑗 -𝑧𝑗 0 0 -1/5 0

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

 Minimize Z = 4x1 + x2 +M a1 +M a2
Iteration-3 𝒄𝒋 4 1 0 0 M M
S.t.

3x1 + x2 + a1 = 3 𝒄𝑩 𝒙𝑩 𝒙𝟏 𝒙𝟐 𝒖 𝒔 𝒂𝟏 𝒂𝟐 RHS Ratio

4x1 + 3x2 –u + a2 = 6 4 𝒙𝟏 1 0 1/5 0 3/5 3
x1 + 2x2 + s = 3
1 𝒙𝟐 0 1 -3/5 0 6/5
& x1, x2 , u, s, a1, a2 ≥0
0 𝒔 0 0 1 1 0 0
• Iteration 3:
𝑧𝑗 4 1 1/5 0 18/5
• Entering variable =u
• Leaving variable =s
𝑐𝑗 -𝑧𝑗 0 0 -1/5 0

Iteration-4 𝒄𝒋 4 1 0 0 M M
𝒄𝑩 𝒙𝑩 𝒙𝟏 𝒙𝟐 𝒖 𝒔 𝒂𝟏 𝒂𝟐 RHS Ratio
• Iteration 4:
• Entering variable =? 4 𝒙𝟏
• Leaving variable =?
1 𝒙𝟐
0 𝒖
𝑧𝑗
𝑐𝑗 -𝑧𝑗

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

 Minimize Z = 4x1 + x2 +M a1 +M a2
Iteration-3 𝒄𝒋 4 1 0 0 M M
S.t.

3x1 + x2 + a1 = 3 𝒄𝑩 𝒙𝑩 𝒙𝟏 𝒙𝟐 𝒖 𝒔 𝒂𝟏 𝒂𝟐 RHS Ratio

4x1 + 3x2 –u + a2 = 6 4 𝒙𝟏 1 0 1/5 0 3/5 3
x1 + 2x2 + s = 3
1 𝒙𝟐 0 1 -3/5 0 6/5
& x1, x2 , u, s, a1, a2 ≥0
0 𝒔 0 0 1 1 0 0
• Iteration 3:
𝑧𝑗 4 1 1/5 0 18/5
• Entering variable =u
• Leaving variable =s
𝑐𝑗 -𝑧𝑗 0 0 -1/5 0

Iteration-4 𝒄𝒋 4 1 0 0 M M
• Iteration 4:
• Algorithm terminates
𝒄𝑩 𝒙𝑩 𝒙𝟏 𝒙𝟐 𝒖 𝒔 𝒂𝟏 𝒂𝟐 RHS Ratio

4 𝒙𝟏 1 0 0 -1/5 3/5
Optimal Solution: 1 𝒙𝟐 0 1 0 3/5 6/5
𝑥1 = 3/5, 𝑥2 = 6/5
0 𝒖 0 0 1 1 0
Optimal objective
function value, 𝑍 ∗ = 18/5 𝑧𝑗 4 1 0 -1/5 18/5
𝑐𝑗 -𝑧𝑗 0 0 0 1/5

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

Example 2
Solve the following LPP using Big M method (Unbounded Optimal Solution - Maximization Case)
Maximize Z = 6x1 + 8x2
S.t.
2x1 + 3x2 ≥ 15
4x1 + 2x2 ≥ 8
& x1, x2 ≥0

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

Example 2
Solve the following LPP using Big M method (Unbounded Optimal Solution - Maximization Case)
Maximize Z = 6x1 + 8x2
S.t.
2x1 + 3x2 ≥ 15
4x1 + 2x2 ≥ 8
& x1, x2 ≥0

• 𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍 = 6𝑥1 + 8𝑥2 + 0𝑢1 + 0𝑢2 − 𝑀𝑎1 − 𝑀𝑎2

• S.t.

• 2𝑥1 + 3𝑥2 – 𝑢1 + 𝑎1 = 15

• 4𝑥1 + 2𝑥2 − 𝑢2 + 𝑎2 = 8

• & 𝑥1, 𝑥2 , 𝑢1 , 𝑢2 , 𝑎1 , 𝑎2 ≥ 0

The algorithm will not terminate. There will be no leaving variable in the last iteration.

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

Example 3
Solve the following LPP using Big M method (Multiple Optimum Solution Case)
Maximize Z = 2000x1 + 4000x2
s.t.
1x1 - 2x2 ≥ 10
1x1 + 2x2 ≤ 30
& x1, x2 ≥0

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

Example 3
Solve the following LPP using Big M method (Multiple Optimum Solution Case)
Maximize Z = 2000x1 + 4000x2
s.t.
1x1 - 2x2 ≥ 10
1x1 + 2x2 ≤ 30
& x1, x2 ≥0

𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍 = 2000𝑥1 + 4000𝑥2 + 0𝑢 + 0𝑠 − 𝑀𝑎

𝑠. 𝑡.
1𝑥1 − 2𝑥2 − 𝑢 + 𝑎 = 10
1𝑥1 + 2𝑥2 + 𝑠 = 30
& 𝑥1, 𝑥2 , 𝑢, 𝑠, 𝑎 ≥ 0

The algorithm will terminate but one of the non-basic variables will have zero value.

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

Example 4
Solve the following LPP (No optimum solution case)
Maximize Z = 20x1 + 30x2
s.t.
2x1 + x2 ≤ 40
4x1 - x2 ≤ 20
x1 ≥ 30
& x1, x2 ≥0

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

Example 4
Maximize
Solve the following LPP (No optimum Z = 20x1 + 30x2+ 0s1+ 0s2+ 0u - Ma
solution case)
S.t.
Maximize Z = 20x1 + 30x2
2x1 + x2 + s1 =40
S.t.
4x1 - x2 + s2 =20
2x1 + x2 ≤ 40
x1 –u + a = 30
4x1 - x2 ≤ 20
& x1, x2, s2, s2,u, a ≥0
x1 ≥ 30
& x1, x2 ≥0

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

Example 4 Iteration-1
𝒄𝑩
𝒄𝒋
𝒙𝑩
20
𝒙𝟏
30
𝒙𝟐 𝒔𝟏
0 0
𝒔𝟐
0
𝒖
-M
𝒂 RHS Ratio
Maximize
Z = 20x1 + 30x2+ 0s1+ 0 𝒔𝟏 2 1 1 0 0 0 40 20
0s2+ 0u - Ma 0 𝒔𝟐 4 -1 0 1 0 0 20 5
S.t. -M 𝒂 1 0 0 0 -1 1 30 30
2x1 + x2 + s1 =40 𝑧𝑗 -M 0 0 0 M -M
4x1 - x2 + s2 =20 𝑐𝑗 -𝑧𝑗 20+M 30 0 0 -M 0
x1 –u + a = 30 Iteration-2 𝒄𝒋 20 30 0 0 0 -M
& x1, x2, s2, s2,u, a ≥0 𝒄𝑩 𝒙𝑩 𝒙𝟏 𝒙𝟐 𝒔𝟏 𝒔𝟐 𝒖 𝒂 RHS Ratio

0 𝒔𝟏 0 3/2 1 -1/2 0 0 30 20
20 𝒙𝟏 1 -1/4 0 1/4 0 0 5 -
-M 𝒂 0 1/4 0 -1/4 -1 1 25 100
𝑧𝑗 20 -M/4-5 0 M/4+5 M -M
𝑐𝑗 -𝑧𝑗 0 M/4+35 0 -M/4-5 -M 0
05-11-2024 Dr. Deepti Mohan | Goa Institute of Management
 Example 4 Iteration-2 𝒄𝒋 20 30 0 0 0 -M
Maximize 𝒄𝑩 𝒙𝑩 𝒙𝟏 𝒙𝟐 𝒔𝟏 𝒔𝟐 𝒖 𝒂 RHS Ratio
Z = 20x1 + 30x2+ 0s1+
0s2+ 0u - Ma 0 𝒔𝟏 0 3/2 1 -1/2 0 0 30 20
S.t. 20 𝒙𝟏 1 -1/4 0 1/4 0 0 5 -
-M 𝒂 0 1/4 0 -1/4 -1 1 25 100
2x1 + x2 + s1 =40
𝑧𝑗 20 -M/4-5 0 M/4+5 M
4x1 - x2 + s2 =20
𝑐𝑗 -𝑧𝑗 0 M/4+35 0 -M/4-5 -M
x1 –u + a = 30
Iteration-3 𝒄𝒋 20 30 0 0 0 -M
& x1, x2, s2, s2,u, a ≥0
𝒄𝑩 𝒙𝑩 𝒙𝟏 𝒙𝟐 𝒔𝟏 𝒔𝟐 𝒖 𝒂 RHS Ratio

Algorithm terminated 30 𝒙𝟐 0 1 2/3 -1/3 0 0 20

but artificial variable
20 𝒙𝟏 1 0 1/6 1/6 0 0 10
remains in the basic
variable set with a non- -M 𝒂 0 0 -1/6 -1/6 -1 1 20
zero value. 𝑧𝑗 20 30 M/6+ M/6+ M
𝑐𝑗 -𝑧𝑗 0 0 -M/6- -M/6- -M
05-11-2024 Dr. Deepti Mohan | Goa Institute of Management
Infeasible LPP
• If any artificial variable is a basic variable with non-zero value when
the BigM algorithm is terminating, then the original problem is
infeasible (has no feasible solution).

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management

Summary of the LPP cases in Simplex/BigM
Case How to identify in Simplex/Big M
Unique optimal solution Algorithm terminates with non-zero cj-zj values for all non-basic variables
Multiple optimal solutions Algorithm terminates with one or more non-basic variable(s) having zero cj-zj value.
Unbounded Algorithm does not terminate but no basic variable is able to leave.
No feasible solution Big M terminates, but artificial variable(s) with non-zero value remains as a basic
variable when the algorithm terminates.

05-11-2024 Dr. Deepti Mohan | Goa Institute of Management