Scribd
Upload
3 views8 pages

STUDY MATERIAL ON LINEAR PROGRAMMING PART 10

The document discusses slack and surplus variables used in linear programming to convert inequalities into equalities for optimization problems. It outlines the standard form of linear programming problems and details the Simplex method for solving a specific linear programming problem, including the construction of tables and calculations for optimal solutions. The final optimal solution is presented as x1 = 200, x2 = 400, with a maximum value of z = 3600.

Uploaded by

balram12122000
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
3 views8 pages

STUDY MATERIAL ON LINEAR PROGRAMMING PART 10

The document discusses slack and surplus variables used in linear programming to convert inequalities into equalities for optimization problems. It outlines the standard form of linear programming problems and details the Simplex method for solving a specific linear programming problem, including the construction of tables and calculations for optimal solutions. The final optimal solution is presented as x1 = 200, x2 = 400, with a maximum value of z = 3600.

Uploaded by

balram12122000
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 8

STUDY MATERIAL ON LINEAR PROGRAMMING PART 10

Slack variable and Surplus variable

If the constraint in the L. P. P. be in the form of a “less than”


type of inequality as
a 11 x1 + a 12 x2 + ... + a 1n xn ≤ b1 ,

then this will be transformed into an equality by adding to the


left a non-negative quantity xn + 1 , called the slack variable as

a 11 x1 + a 12 x2 + ... + a 1n xn + xn + 1 = b1.

Similarly, if the constraint in the L. P. P. be in the form of a


“greater than” type of inequality as

a 21 x1 + a 22 x2 + ... + a 2n xn ≥ b2 ,
then this will be transformed into an equality by subtracting
from the left a non-negative quantity xn+2, called the surplus
variable, as

a 21 x1 + a 22 x2 + ... + a 2n xn − xn + 2 = b2.
Recasting a linear programming problem

A general linear programming problem can always be put


in a standard form whose characteristics are the following:
1. All constraints are equations except for the non-negativity
constraints which remain inequalities ( ≥ 0).
2. The right hand side element of each constraint equation is
non - negative.
3. All variables are non-negative.
4. The objective function is of the maximization type.

Simplex Method:

Problem: Solve the following L.P.P. by Simplex method:


Maximize z = 4x1 + 7x2,
subject to 2x1 + x2 ≤ 1000,
10x1 + 10x2 ≤ 6000,
2x1 + 4x2 ≤ 2000 ,
x1, x2 ≥ 0.
Solution: Introducing slack variables x3 , x4 , x5, the above
problem can be rewritten in the standard form as

Maximize z = 4x1 + 7x2 + 0.x3 + 0.x4 + 0.x5

subject to 2x1 + x2 + x3 + 0.x4 + 0.x5 = 1000,

10x1 + 10x2 + 0.x3 + x4 + 0.x5 = 6000

2x1 + 4x2 + 0.x3 + 0.x4 + x5 = 2000,

x1, x2 , x3 , x4, x5 ≥ 0.

The initial basis matrix is the identity matrix which is obtained


by the coefficients of x3 , x4, x5.

We put the information in the table below.

Table 1

cj 4 7 0 0 0
𝐜B B 𝐱B 𝐛 𝐚1 𝐚2 𝐚3 𝐚4 𝐚5
0 𝐚3 x3 1000 2 1 1 0 0
0 𝐚4 x4 6000 10 10 0 1 0
0 𝐚5 x5 2000 2 44 0 0 1
zj–cj –4 –7 0 0 0
↑ ↓
Index number

= ∑ (numbers in each column × corresponding number in


the column 𝐜B ) – number in the objective row

To test the optimality, we have computed in the table all the


index numbers, that is (zj – cj) for j = 1, 2, 3, 4, 5 as

z1 – c1 = (2 × 0 + 10 × 0 + 2 × 0) – 4 = – 4;

z2 – c2 = (1 × 0 + 10 × 0 + 4 × 0) – 7 = – 7;

z3 – c3 = (1 × 0 + 0 × 0 + 0 × 0) – 0 = 0;

z4 – c4 = (0 × 0 + 1 × 0 + 0 × 0) – 0 = 0;

z5 – c5 = (0 × 0 + 0 × 0 + 1 × 0) – 0 = 0;

In this table, at least one (zj – cj) < 0 and hence this table will
not give the optimal solution. To construct a new table, we
proceed as below.

Minimum- most number in the index row is (– 7) and hence


the column corresponding to this number is the key column and
𝐚2 is thus the entering vector which is shown by an upward
arrow in the table.
To find the key row, we divide the numbers in the constant
column (𝐛) by the corresponding positive numbers in the key
column.

1000
Thus, for the 1st row, = 1000,
1

6000
for the 2nd row, = 600,
10

2000
for the 3rd row, = 500.
4

The third being the smallest, 3rd row is the key row and 𝐚5
under B is the departing vector, which is shown by a downward
arrow in the table. Then 4, being the intersection of key column
and key row, is the key number.

The entering vector is 𝐚2 corresponding to the variable x2 with


price parameter 7 in the objective function. We then replace
from the key row 0, 𝐚5 and x5 under the columns 𝐜B , B and 𝐱B
respectively by 7, 𝐚2 and x2 respectively, the corresponding
elements of the key column and construct a new table.

We divide the elements of the key row by the key number 4 to


get the elements of the new 3rd row of the new table.
To get the elements of the new 1st row of the new table, we
multiply the elements of the new 3rd row by 1 and subtract them
from the elements of the old 1st row.

To get the elements of the new 2nd row of the new table, we
multiply the elements of the new 3rd row by 10 and subtract
them from the elements of the old 2nd row.

Table 2

cj 4 7 0 0 0
𝐜B B 𝐱B 𝐛 𝐚1 𝐚2 𝐚3 𝐚4 𝐚5

0 𝐚3 x3 500 3/2 0 1 0 – 1/4


0 𝐚4 x4 1000 5 0 0 1 – 5/2
7 𝐚2 x2 500 1/2 1 0 0 1/4
zj–cj – 1/2 0 0 0 7/4
↑ ↓

In the table 2, the minimum- most number in the index row is


(– 1/2), which corresponds to the column of vector 𝐚1 . Hence 𝐚1
is the entering vector.
500 1000 500
Min , , 1/2 = 200, which corresponds to
3/2 5

the 2nd row and 𝐚4 is the departing vector. Hence 5 is the key
number.

We divide the elements of the key row by the key number 5 to


get the elements of the new 2nd row of the new table.

To get the elements of the new 1st row of the new table, we
multiply the elements of the new 2nd row by 3/2 and subtract
them from the elements of the old 1st row.

To get the elements of the new 3rd row of the new table, we
multiply the elements of the new 2nd row by 1/2 and subtract
them from the elements of the old 3rd row.

Table 3

cj 4 7 0 0 0
𝐜B B 𝐱B 𝐛 𝐚1 𝐚2 𝐚3 𝐚4 𝐚5

0 𝐚3 x3 200 0 0 1 – 3/10 1/2


4 𝐚1 x1 200 1 0 0 1/5 – 1/2
7 𝐚2 x2 400 0 1 0 – 1/10 3/2
zj–cj 0 0 0 1/10 3/2
Here zj – cj ≥ 0 for all j. Hence this table gives the optimal
solution.
The required optimal solution is thus
x1 = 200, x2 = 400 and zmax = 4 × 200 + 7 × 400 = 3600.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy