Scribd
Upload
86 views

Simplex Method (LINEAR PROGRAMMING)

The document describes the revised simplex method for solving linear programming problems. It provides an example problem to minimize z = -3x1 + x2 + x3 subject to several constraints. The solution proceeds through 4 iterations of the simplex method by choosing entering and leaving variables at each step based on pivot columns and simplex multipliers. The revised simplex method requires less computation than other simplex methods and provides multipliers useful for sensitivity analysis. An exercise is provided to solve another problem using the revised simplex method.

Uploaded by

mohisdasdadas
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
86 views

Simplex Method (LINEAR PROGRAMMING)

The document describes the revised simplex method for solving linear programming problems. It provides an example problem to minimize z = -3x1 + x2 + x3 subject to several constraints. The solution proceeds through 4 iterations of the simplex method by choosing entering and leaving variables at each step based on pivot columns and simplex multipliers. The revised simplex method requires less computation than other simplex methods and provides multipliers useful for sensitivity analysis. An exercise is provided to solve another problem using the revised simplex method.

Uploaded by

mohisdasdadas
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 23

LINEAR PROGRAMMING

Lecture 11
REVISED SIMPLEX METHOD
by
Dr. KUSUM DEEP
Professor
Department of Mathematics
Indian Institute of Technology Roorkee
Roorkee – 247667, Uttaranchal, INDIA
Emails: [email protected], [email protected]
OUTLINE OF THIS TALK

• The Revised Simplex Method


• Example
• Simplex multipliers
• Advantages of the Revised Simplex
Method
• Exercise
Example
Minimize z = – 3x1 + x2 + x3
Subject to
x1 – 2x2 + x3 < 11
– 4x1 + x2 + 2x3 > 3
2x1 – x3 = –1
xi > 0 , i = 1, 2, 3.
Maximize z = 3x1 – x2 – x3
Subject to x1 – 2x2 + x3 + x4 = 11
– 4x1 + x2 + 2x3 – x5 = 3
– 2x1 + x3 = 1
xi > 0 , i = 1, 2, 3, 4, 5.
Maximize z = 3x1 – x2 – x3 – Mx6 – Mx7
where M → ∞
Subject to
x1 – 2x2 + x3 + x4 = 11
– 4x1 + x2 + 2x3 – x5 + x6 = 3
– 2x1 + x3 + x7 = 1
xi > 0 , i = 1, 2, 3, 4, 5 and xi > 0 for i = 6, 7
3 –1 –1 0 0 –M –M
Iter C0 Basis x1 x2 x3 x4 x5 x6 x7 RHS
1 0 x4 1 –2 1 1 0 0 0 11
–M x6 –4 1 2 0 –1 1 0 3
–M x7 –2 0 1 0 0 0 1 1

2 0 x4 3 –2 0 1 0 0 –1 10
–M x6 0 1 0 0 –1 1 –2 1
–1 x3 –2 0 1 0 0 0 1 1

3 0 X4 3 0 0 1 –2 2 –5 12
–1 x2 0 1 0 0 –1 1 –2 1
–1 x3 –2 0 1 0 0 0 1 1

4 3 x1 1 0 0 1/3 –2/3 2/3 –5/3 4


–1 x2 0 1 0 0 –1 1 –2 1
–1 x3 0 0 1 2/3 –4/3 4/3 –7/3 9
Observations

1 −2 1 1/3 2/3 −5/3 1 0 0


�−4 1 2� � 0 1 −2 � = �0 1 0�
−2 0 1 2/3 4/3 −7/3 0 0 1
Maximize z = 3x1 – x2 – x3 – Mx6 – Mx7
s. t. x1 – 2x2 + x3 + x4 = 11
– 4x1 + x2 + 2x3 – x5 + x6 = 3
– 2x1 + x3 + x7 = 1
1 −2 1 1 0
P1=�−4�, P2=� 1 �, P3=�2�, P4 =�0�, P5=�−1�,
−2 0 1 0 0
0 0 11
P6= �1�, P7 = �0� and b = � 3 �
0 1 1
Since x4, x6, x7 is basis in initial table
1 0 0
[B]3X3 = (𝑃𝑃4 𝑃𝑃6 𝑃𝑃7 ) = �0 1 0� = I
0 0 1
Hence B-1 = I and 𝑏𝑏� = B-1b = b
Initial Table:
B-1 b
x4 1 0 0 11
x6 0 1 0 3
x7 0 0 1 1

Simplex multipliers = (π1 π2 π3) = cBB-1


1 0 0
=(0 −𝑀𝑀 −𝑀𝑀) �0 1 0� =(0 −𝑀𝑀 −𝑀𝑀)
0 0 1
𝑐𝑐�𝑗𝑗 = 𝑐𝑐𝑗𝑗 − 𝜋𝜋𝑃𝑃𝑗𝑗 for non-basic variables,
i.e. for j=1, 2, 3 and 5.
1
𝑐𝑐�1 = 𝑐𝑐1 − 𝜋𝜋𝑃𝑃1 =3 - (0 −𝑀𝑀 −𝑀𝑀) �−4� = 3 - 6M
−2
−2
𝑐𝑐�2 = 𝑐𝑐2 − 𝜋𝜋𝑃𝑃2 =1 - (0 −𝑀𝑀 −𝑀𝑀) � 1 � = M - 1
0
1
𝑐𝑐�3 = 𝑐𝑐3 − 𝜋𝜋𝑃𝑃3 =1 - (0 −𝑀𝑀 −𝑀𝑀) �2�= 3M - 1
1
0
𝑐𝑐�5 = 𝑐𝑐5 − 𝜋𝜋𝑃𝑃5 =0 - (0 −𝑀𝑀 −𝑀𝑀) �−1� = - M
0
Since 𝑐𝑐�3 is most positive, x3 enters the basis.
���
The pivot column is: 𝑃𝑃3 = B 𝑃𝑃3
-1

1 0 0 1 1
= �0 1 0� �2� = �2�
0 0 1 1 1
The initial Table is completed as:

Entering Pivot
B-1 b variable Column

x4 1 0 0 11 1
x6 0 1 0 3 x3 2
x7 0 0 1 1 1
From Table 1
Entering Pivot
B-1 b variable Column

x4 1 0 0 11 1
x6 0 1 0 3 x3 2
x7 0 0 1 1 1
Minimum ratio test is = min(11/1, 3/2, 1/1) = 1
So, leaving variable is x7 and Pivot is : 1
R1 is replaced by R1 – R3 and
R2 is replaced by R2 – 2R3
Table 2:
B-1 b
x4 1 0 -1 10
x6 0 1 -2 1
x3 0 0 1 1
Simplex multipliers =(π1 π2 π3) = cBB-1
1 0 −1
= (0 −𝑀𝑀 −1) �0 1 −2�
0 0 1
= (0 −𝑀𝑀 2𝑀𝑀 − 1)
𝑐𝑐�1 = 1, 𝑐𝑐�2 = 𝑀𝑀 − 1 and 𝑐𝑐�5 = −𝑀𝑀
Note: Need not calculate 𝑐𝑐�7
as x7 is artificial variable.
𝑐𝑐�2 is most positive,
so entering variable is x2.
���2 = B-1 𝑃𝑃2
The pivot column is: 𝑃𝑃
1 0 −1 −2 −2
= �0 1 −2� � 1 � = � 1 �
0 0 1 0 0
Table 2 is completed as:
Entering Pivot
B-1 b variable Column

x4 1 0 -1 10 -2
x6 0 1 -2 1 x2 1
x3 0 0 1 1 0

Leaving variable is x6 and Pivot is : 1


R1 is replaced by R1 – R3 and
R2 is replaced by R2 – 2R3
Table 3
B-1 b
x4 1 2 -5 12
x2 0 1 -2 1
x3 0 0 1 1
Simplex multipliers =(π1 π2 π3) = cBB -1

1 2 −5
=(0 −1 −1) �0 1 −2�=(0 −1 1)
0 0 1
𝑐𝑐�1 = 1, 𝑐𝑐�3 = −1, i.e. 𝑐𝑐�1 is most positive
so entering variable is x1
�
The pivot column is: 𝑃𝑃1 = B 𝑃𝑃1
-1

1 2 −5 1 3
= �0 1 −2� �−4� = � 0 �
0 0 1 −2 −2
Table 3 is completed as:

Entering Pivot
B-1 b variable Column

x4 1 2 -5 12 3
x6 0 1 -2 1 x1 0
x3 0 0 1 1 -2
Table 4
B-1 b
x1 1/3 2/3 -5/3 4
x2 0 1 -2 1
x3 2/3 4/3 -7/3 9
Simplex multipliers =(π1 π2 π3) = cBB-1
1/3 2/3 −5/3
=(3 −1 −1) � 0 1 −2 �
2/3 4/3 −7/3
=(1/3 −1/3 2/3)
1
𝑐𝑐�4 = 0 - (1/3 −1/3 −2/3) �0� = -1/3
0
0
𝑐𝑐�5 = 0 - (1/3 −1/3 −2/3) �−1� =-1/3
0
Table 4 is optimal.
Solution is:x1=4, x2=1, x3=9 & x4=0, x5=0.
Optimum value is:2
Iter CB Basis B-1 RHS Simplex multipliers
= CB B-1
1 0 x4 1 0 0 11 1 0 0
–M x6 0 1 0 3 (0 –M –M) �0 1 0�
–M x7 0 0 1 1 0 0 1
= (0 –M –M)
2 0 x4 1 0 –1 10 1 0 −1
–M x6 0 1 –2 1 (0 –M –1) �0 1 −2�
0 0 1
–1 x3 0 0 1 1 = (0 –M 2M–1)
3 0 x4 1 2 –5 12 1 2 −5
(0 –1 –1) �0 1 −2�
–1 x2 0 1 –2 1
0 0 1
–1 x3 0 0 1 1 = (0 –1 1)
4 3 x1 1/3 2/3 –5/3 4 1/3 2/3 −5/3
–1 x2 0 1 –2 1 (3 –1 –1) � 0 1 −2 �
–1 x3 2/3 4/3 –7/3 9 2/3 4/3 −7/3
= (1/3 –1/3 2/3)
Advantages of the
Revised Simplex Method
• It needs lesser computations.

• It provides simplex multipliers at each


iterations, which are useful for duality
theory and sensitivity analysis.
Exercise
Solve the following problem by the
Revised Simplex Method:

Minimize x1 + 2x2 + x3
s. t. 2x1 + x2 + x3 < 2
3x1 + 4x2 + 2x3 > 16
x1 , x2 , x3 > 0.
THANK YOU
2/15/2016 23

You might also like