Você está na página 1de 30

Operations Research Unit 3

Unit 3 Simplex Method


Structure:
3.1 Introduction
Learning objectives
3.2 Standard Form of LPP
The standard form of LPP
Fundamental theorem of LPP
3.3 Solution of LPP – Simplex Method
Initial basic feasible solution of a LPP
To solve problem by Simplex Method
3.4 The Simplex Algorithm
Steps
3.5 Penalty Cost Method or Big M-method
3.6 Two Phase Method
3.7 Solved Problems on Minimisation
3.8 Summary
3.9 Terminal Questions
3.10 Answers to SAQs & TQs
Answers to Self Assessment Questions
Answers to Terminal Questions
3.11 References

3.1 Introduction
Welcome to the third unit of Operations Research Management on the
simplex method. The simplex method focuses on solving LPP of any
enormity involving two or more decision variables.
The simplex algorithm is an iterative procedure for finding the optimal
solution to a linear programming problem. The objective function controls
the development and evaluation of each feasible solution to the problem. If a
feasible solution exists, it is located at a corner point of the feasible region
determined by the constraints of the system.
The simplex method simply selects the optimal solution amongst the set of
feasible solutions of the problem. The efficiency of this algorithm is because
it considers only those feasible solutions which are provided by the corner

Sikkim Manipal University Page No. 46


Operations Research Unit 3

points, and that too not all of them. You can consider obtaining an optimal
solution based on a minimum number of feasible solutions.
Learning objectives
By the end of this unit, you should be able to:
 Create a standard form of LPP from the given hypothesis
 Apply the simplex algorithm to the system of equations
 Interpret the big M-technique
 Discuss the importance of the two phase method
 Construct the dual from the primal (and vice versa)

3.2 Standard Form of LPP


The characteristics of the standard form of LPP are:
1. All constraints are equations except for the non-negativity condition,
which remain inequalities (, 0) only.
2. The right-hand side element of each constraint equation is non-negative.
3. All the variables are non-negative.
4. The objective function is of maximisation or minimisation type.
You can change the inequality constraints of equations by adding or
subtracting the left hand side of each such constraint by a non-negative
variable. The non-negative variable that has to be added to a constraint
inequality of the form  to change it to an equation is called a slack
variable. The non-negative variable subtracted from a constraint inequality
of the form  to change it to an equation is called a surplus variable.
To make the right hand side of a constraint equation positive, multiply both
the sides of the resulting equation by (-1). Use the elementary
transformations introduced with the canonical form to achieve the remaining
characteristics.
3.2.1 The standard form of the LPP
Any standard form of the LPP is given by
n
Maximise or Minimise z   Ci x i
i1

Sikkim Manipal University Page No. 47


Operations Research Unit 3

Subject to:
n

a ij x j  Si  bi (bi  0) i  1, 2 .......... . m.
j 1
& xj  0, j = 1, 2, --- n.
S i  0, i = 1, 2, --- m

3.2.2 Fundamental theorem of LPP


A set of m simultaneous linear equations in n unknowns/variables, n  m,
AX = b, with r (A) = m.
If there is a feasible solution X  0, then there exists a basic feasible
solution.

Self Assessment Questions


State True or False
1. We add surplus variable for “≤” of constraint.
2. The right hand side element of each constraint is non-negative.

3.3 Solution of the Linear Programming Program – Simplex


Method
Consider a LPP given in the standard form
To optimise z = c1 x1 + c2 x2 + ---+ cn xn
Subject to
a11 x1 + a12 x2 + -- + an x n  S1 = b1
a21 x1 + a22 x2 + ----+ a2n xn  S2 = b2
………………………………………………………………….

am1 x1 + am2 x2 + -- + amn xn  Sm = bm


x1, x2, --- xn, S1, S2 ---, Sm  0.
To each of the constraint equations, add a new variable called an artificial
variable on the left hand side of every equation which does not contain a
slack variable. Subsequently every constraint equation will contain either a
slack variable or an artificial variable.
The introduction of slack and surplus variables does not alter either the
constraints or the objective function. Therefore, you can incorporate such

Sikkim Manipal University Page No. 48


Operations Research Unit 3

variables in the objective function with zero coefficients. However, the


artificial variables do change the constraints as these are added only to the
left hand side of the equations.
The newly derived constraint equation is equivalent to the original equation,
only if all the artificial variables have value zero. Artificial variables are
incorporated into the objective function with very large positive coefficient M
in the minimisation program and very large negative coefficient–M in the
maximisation program guaranteeing optimal solutions. The large positive
and negative coefficients represent the penalty incurred for making a unit
assignment to the artificial variable.
Thus the standard form of LPP can be given as follows:
Optimise Z = CT X
Subject to AX = B,
And X0
Where X is a column vector with decision, slack, surplus and artificial
variables, C is the vector corresponding to the costs; A is the coefficient
matrix of the constraint equations and B is the column vector of the right
hand side of the constraint equations.

Sikkim Manipal University Page No. 49


Operations Research Unit 3

Solved Problem 1
Consider the following LPP
Minimise Z = 4x1 + x2
Subject to 3x1 + x2 = 3
4x1 + 3x2  6
x1 + 2x2  3,
x1, x2  0

Solution: Rewriting in the standard form,


Minimise Z = 4x1 + x2 + 0.S1 + 0.S2 + M (A1 + A2)
Subject to 3x1+ x2 + A1 = 3
4x1 + 3x2 – S1 + A2 = 6
x1 + 2x2 + S2 = 3
x1, x2, S1, S2, A1, A2  0
When S2 is slack variable, S1 is a surplus variable and A1 & A2 an artificial
variables. Representing this program in matrixes, you have
 x1 
 
 x2 
Minimise Z = (4 1 0 0 M M) S 
 1
 S2 
 
 A1 
A 
 2
 x1 
  3
 3 1 0 0 1 0
Subject to    x2 
S  =  6 
 4 3  1 0 0 1  1
 1 2 0 1 0 0  S2  3
     
 A1 
A 
 2

 x1 
 
And  x2  0
S 
 1
 S2 
 
 A1 
A 
 2

Sikkim Manipal University Page No. 50


Operations Research Unit 3

3.3.1 Initial basic feasible solution of a LPP


Consider a system of m equations in n unknowns x1, x2 - - an,
a11 x1 + a12 x2 + - - + a1n xn = b1

a21 x1 + a22 x2 + - - + a2n xn = b2


………………………………………………………
am1 x1 + am2 x2 + - - + amn xn = bn
Where m  n
To solve this system of equations, you must first assign any of n – m
variables with value zero. The variables assigned the value zero are called
non-basic variables, while the remaining variables are called basic
variables. Then, solve the equation to obtain the values of the basic
variables. If one or more values of the basic variables are valued at zero,
then solution is said to degenerate, whereas if all basic variable have non-
zero values, then the solution is called a non-degenerate solution. A basic
solution satisfying all constraints is said to be feasible.

Solved Problem 2
Consider the system of equations
2x1 + x2 – x3 = 2
3x1 + 2x2 + x3 = 3
Where x1, x2, x3  0
Solution: At the onset, assign 3 – 2 = 1 variable, the value zero since
there are three variables and two equations.
Case (i):
Let x3 = 0 i.e., x3 be a non-basic variable, then equation becomes
2x1 + x2 = 2
3x1 + 2x2 = 3
Solving, we get
x1 = 1, x2 = 0.
 The solution degenerates, but is feasible.
Case (ii): Let x2 be a non-basic variable i.e., x2 = 0, then solution is
x1 = 1 and x3 = 0
Here also the solution degenerates but is feasible

Sikkim Manipal University Page No. 51


Operations Research Unit 3

Case (iii): Let x1 be non-basic i.e., x1= 0


1 1
Solution is x2 = , x3 = – .
3 3
The solution non-degenerates, but is not feasible.
Consider a LPP given in the standard form
Optimise Z = CTX
Subject to AX = B,
X  0.
To arrive at the initial solution of such a problem denoted by X 0, treat all
decision and surplus variables as non-basic variables, Assign value zero;
all slack and artificial variables as basic variables are assigned values,
which are on right-hand side of the corresponding constraint equations.

3.3.2 To solve problem by simplex method


To solve a problem by the simplex method, follow the steps below.
1. Introduce stack variables (Si’s) for “” type of constraint.
2. Introduce surplus variables (Si’s) and artificial variables (Ai) for “” type
of constraint.
3. Introduce only Artificial variable for “=” type of constraint.
4. Cost (Cj) of slack and surplus variables will be zero and that of artificial
variable will be “M”
5. Find Zj - Cj for each variable.
6. Slack and artificial variables will form basic variable for the first simplex
table. Surplus variable will never become basic variable for the first
simplex table.
7. Zj = sum of [cost of variable x its coefficients in the constraints – Profit
or cost coefficient of the variable].
8. Select the most negative value of Zj - Cj. That column is called key
column. The variable corresponding to the column will become basic
variable for the next table.
9. Divide the quantities by the corresponding values of the key column to
get ratios; select the minimum ratio. This becomes the key row. The
basic variable corresponding to this row will be replaced by the
variable found in step 6.

Sikkim Manipal University Page No. 52


Operations Research Unit 3

10. The element that lies both on key column and key row is called Pivotal
element.
11. Ratios with negative and “” value are not considered for determining
key row.
12. Once an artificial variable is removed as basic variable, its column will
be deleted from next iteration.
13. For maximisation problems, decision variables coefficient will be same
as in the objective function. For minimisation problems, decision
variables coefficients will have opposite signs as compared to
objective function.
14. Values of artificial variables will always is – M for both maximisation
and minimisation problems.
15. The process is continued till all Zj - Cj  0.

Self Assessment Questions


State True or False
3. A basic solution is said to be a feasible solution if it satisfies all
constraints.
4. If one or more values of basic variable are zero then solution is said to
be degenerate.
5. The right hand side element of each constraint is non-negative.

3.4 The Simplex Algorithm


To test for optimality of the current basic feasible solution of the LPP, use
the following algorithm called simplex algorithm. Let’s also assume that
there are no artificial variables existing in the program.
3.4.1 Steps
Perform the following steps to solve the simplex algorithm.

Sikkim Manipal University Page No. 53


Operations Research Unit 3

Figure 3.1: Steps to solve simple algorithm


1) Locate the negative number in the last row of the simplex table. Do not
include the last column. The column that has negative number is called
the work column.
2) Next, form ratios by dividing each positive number in the work column,
excluding the last row into the element in the same row and last column.
Assign that element to the work column to yield the smallest ratio as the
pivot element. If more than one element yields the same smallest ratio,
choose the elements randomly. The program has no solution, if none of
the element in the work column is non-negative.
3) To convert the pivot element to unity (1) and then reduce all other
elements in the work column to zero, use elementary row operations.
4) Replace the x -variable in the pivot row and first column by x-variable in
the first row pivot column. The variable to be replaced is called the
outgoing variable and the variable that replaces it is called the incoming
variable. This new first column is the current set of basic variables.
5) Repeat steps 1 through 4 until all the negative numbers in the last row
excluding the last column are exhausted.
6) You can obtain the optimal solution by assigning the value to each
variable in the first column corresponding to the row and last column. All
other variables are considered as non-basic and have assigned value
zero. The associated optimal value of the objective function is the
number in the last row and last column for a maximisation program, but
the negative of this number for a minimisation problem.
Sikkim Manipal University Page No. 54
Operations Research Unit 3

Caselet
A manufacturing company discontinued production of an unprofitable
product line. This created excess production capacity. The company’s
management is considering devoting this excess capacity to one or
more of the other ongoing products.
Knowing the capacity of the machines, the number of machine hours
required to produce one unit of each product and the unit profit per
product, the management needs to find out how much of each product
the company should produce to maximise profit.
The management can use the simplex method to arrive at the required
solution.

Solved Problem 3
Maximise z = x1+ 9x2 + x3
Subject to x1 + 2x2 + 3x3  9
3x1 + 2x2 + 2x3  15
x1, x2, x3  0.
Rewriting in the standard form
Maximise z = x1 + 9x2 + x3 + 0.S1 + 0.S2
Subject to the conditions
x1 + 2x2 + 3x3 + S1 = 9
3x1 + 2x2 + 2x3 + S2 = 15
x1, x2, x3, S1, S2  0.
Where S1 and S2 are the slack variables.
Solution: The initial basic solution is S1 = 9, S2 = 15

 S1  0
 X0 =   , C0 =  
S2  0

Sikkim Manipal University Page No. 55


Operations Research Unit 3

The initial simplex table is given below.


Table 3.1: Initial simplex table
x1 x2 x3 S1 S2 Ratio
1 9 1 0 0
S1 0 1 2* 3 1 0 9 9
S2 0 3 2 2 0 1 15 = 4.5 
2
15
= 7.5
2
Zj – cj –1 –9 –1 0 0

Work column* – pivot element.
S1 – outgoing variable, x2 incoming variable.
Since the three Zj – Cj are negative, the solution is not optimal. Choose
the maximum negative value that is – 9. The corresponding column
vector x2 enters the basis replacing S1, since ratio is at minimum. You
can use the elementary row operations to reduce the pivot element to 1
and other elements of work column to zero.
First iteration – The variable x1 becomes a basic variable replacing S1
and you obtain the following table.
Table 3.2: First iteration table
x1 x2 x3 S1 S2
1 9 1 0 0
X2 9 1 1 3 1 0 9
2 2 2 2
S2 0 2 0 –1 –1 1 6
9 0 25 9 0 81
Zj –Cj
2 2 2 2
Since all elements of the last row are non-negative, the optimal solution
81
is obtained. The maximum value of the objective function Z is ,
2
9
achieved for x2 = S2 = 6, are the basic variables. All other variables
2
are non-basic.

Sikkim Manipal University Page No. 56


Operations Research Unit 3

Solved Problem 4
Use simplex method to solve the LPP
Maximise Z = 2x1 + 4x2 + x3 + x4
Subject to x1 + 3x2 + x4  4
2x1 + x2  3
x2 + 4x3 + x4  3
x1, x2, x3, x4  0
Rewriting in the standard form
Maximise Z = 2x1 + 4x2 + x3 + x4 + 0.S1 + 0.S2 + 0.S3
Subject to x1 + 3x2 + x4 + S1 = 4
2x1 + x2 + S2 = 3
x2 + 4x3 + x4 + S3 = 3
x1, x2, x3, x4,
S1, S2, S3  0

Solution:
The initial basic solution is S1 = 4, S2 = 3, S3 = 3
 S1  4 0
     
X0 =  S 2  =  3  C0 = 0
S  3  0
 3    
The initial table is given by
Table 3.3: The initial table
X1 x2 x3 x4 S1 S2 S3 Ratio
2 4 1 1 0 0 0
S1 0 1 3* 0 1 1 0 0 4 4

3
S2 0 2 1 0 0 0 1 0 3 3
1
S3 0 0 1 4 1 0 0 1 3 3
1
Zj –Cj –2 –4 –1 –1 0 0 0 0

work column * pivot element

Sikkim Manipal University Page No. 57


Operations Research Unit 3

S1 is the outgoing variable, x2 is the incoming variable to the basic set.


The first iteration gives the following table:

Table 3.4: The first iteration table


x1 x2 x3 x4 S1 S2 S3 Ratio
2 4 1 1 0 0 0
x2 4 1 1 0 1 1 0 0 4
3 3 3 3
S2 0 5 0 0 1 1 1 0 5
– –
3 3 3 3
S3 0 1 0 4* 2 1 0 1 5 5
– – 
3 3 3 3 12
2 0 –1 1 4 0 0 16
Zj –Cj –
3 3 3 3

Work column

x3 enters the new basic set replacing S3, the second iteration gives the
following table :

Table 3.5: Second iteration table

x1 x2 x3 x4 S1 S2 S3 Ratio
2 4 1 1 0 0 0
x2 4 1 1 0 1 1 0 0 4 4
3 3 3 3
S2 0 5 0 0 1 1 1 0 5 1
* – –
3 3 3 3
x3 1 1 0 1 1 1 0 1 5
– –
12 6 12 4 12
3 0 0 1 5 0 1 23
Zj –Cj –
4 2 4 4 4


Work column

Sikkim Manipal University Page No. 58


Operations Research Unit 3

x1 enters the new basic set replacing S2, the third iteration gives the
following table:
Table 3.6: Third iteration table

x1 x2 x3 x4 S1 S2 S3
2 4 1 1 0 0 0
x2 4 0 1 0 2 2 1 0 1

5 5 5
x1 2 1 0 0 1 1 3 0 1
– –
5 5 5
x3 1 0 0 1 3 1 1 1 1

20 10 20 4 2
0 0 0 7 11 9 1 13
Zj –Cj 20 20 2
10 4

Since all elements of the last row are non-negative, the optimal solution
13 1
is Z = which is achieved for x2 = 1, x1 = 1, x3 = and x4 = 0.
2 2

Sikkim Manipal University Page No. 59


Operations Research Unit 3

Case Study
A manufacturing firm has discontinued production of a certain
unprofitable product line. This created considerable excess of
production capacity. Management is considering devoting this excess
capacity to one or more of three products, called product 1, 2 and 3.
The available capacity on the machines, which might limit output is
given below:

Table 3.7: Capacity of the machine for production


Machine Type Available Time
(in machine hours per week)
Milling Machine 250
Lathe 150
Grinder 50
The number of machine-hours required for each unit of the respective
product is given below:
Table 3.8: Machine hours for each unit
Productivity (in Machine
hours/Unit)
Machine Type Product 1 Product 2 Product 3
Milling 8 2 3
Machine
Lathe 4 3 0
Grinder 2 – 1

The unit profit would be Rs. 20, Rs. 6 and Rs. 8 for products 1, 2 and 3.
Find how much of each product the firm should produce in order to
maximise profit?
Let x1, x2, x3 units of products 1, 2 and 3 are produced in a week.
Then total profit from these units is
Z = 20 x1 + 6 x2 + 8 x3
To produce these units the management requires
8x1 + 2x2 + 3x3 machine hours of Milling Machine

Sikkim Manipal University Page No. 60


Operations Research Unit 3

4x1 + 3x2 + 0 x3 machine hours of Lathe


And 2x1 + x3 machine hours of Grinder
Since time available for these three machines are 250, 150 and 50
hours respectively, we have
8x1 +2x2 + 3x3  250
4x1 + 3x2  150
2x1 + x3  50.
Obviously x1, x2, x3  0
Thus the problem is to
Maximise Z = 20x1 + 6x2 + 8x3
Subject to 8x1 + 2x2 + 3x3  250
4x1 + 3x2  150
2x1 + x3  50,
x1, x2, x3  0
Rewriting in the standard form,
Maximise Z = 20x1 + 6x2 + 8x3 + 0S1 + 0S 2 + 0S 3
Subject to 8x1 + 2x2 + 3x3 + S1 = 250
4x1 + 3x2 + S2 = 150
2x1 + x3 + S3 = 50,
x1, x2, x3, S1, S2, S 3  0
Solution: The initial basic solution is
 S1   250 
   
X0 =  S 2  =  150 
S   50 
 3  
The initial simplex table is given by

Sikkim Manipal University Page No. 61


Operations Research Unit 3

Table 3.9: The initial simplex table

x1 x2 x3 S1 S2 S3 Ratio

20 6 8 0 0 0

S1 0 8 2 3 1 0 0 250 250
 31
8

S2 0 4 3 0 0 1 0 150 150
= 37.5
4

S3 0 2* 0 1 0 0 1 50 50
= 25 
2

ZJ – Cj – –6 –8 0 0 0 0
20


Work column * pivot element
x1 enters the basic set of variables replacing the variable s3. The first
iteration gives the following table:
Table 3.10: The first iteration table

x1 x2 x3 s1 s2 s3 Ratio

20 6 8 0 0 0

s1 0 0 2 –1 1 0 –4 50 50
= 25
2
s2 0 0 3* –2 0 1 –2 50 50
3
x1 20 1 0 1 0 0 1 25
2 2
Z j – Cj 0 –6 2 0 0 10 500


Work column * pivot element

Sikkim Manipal University Page No. 62


Operations Research Unit 3

x2 enters the basic set of variables replacing the variable s2. The
second iteration gives the following table:
Table 3.11: The second iteration table
x1 x2 x3 S1 S2 S3 Ratio

20 6 8 0 0 0

S1 0 0 0 1 1 2 8 50 50

3 3 3 3

X2 6 0 1 2 0 1 2 50 –
– –
3 3 3 3

X1 20 1 0 1 0 0 1 25 50
* 2
2
Zj – Cj 0 0 –2 0 2 6 600


Work column * pivot element.
x3 enters the basic set of variables replacing the variable x1. The third
iteration yields the following table:
Table 3.12: The third iteration table
x1 x2 x3 S1 S2 S3

20 6 8 0 0 0

S1 0 2 0 0 1 2 –3 0
– –
3 3
X2 6 4 1 0 0 1 0 50
3 3
X3 8 2 0 1 0 0 1 50

Zj – cj 4 0 0 0 2 8 700

Since all zj – cj  0 in the last row, the optimum solution is 700.


i.e., the maximum profit is Rs. 700/- which is achieved by producing 50
units of product 2 and 50 units of product 3.

Sikkim Manipal University Page No. 63


Operations Research Unit 3

Self Assessment Questions


State Yes or No
6. The key column is determined by Zj - Cj row.
7. Pivotal element lies on the crossing of key column and key row.
8. The negative and infinite ratios are considered for determining key row.

3.5 Penalty Cost Method or Big-M Method


Consider a LPP when at least one of the constraints is of type  or =. While
expressing in the standard form, add a non negative variable to each of
such constraints. These variables are called artificial variables.
Addition of artificial variables causes violation of the corresponding
constraints as they are added to only one side of an equation. The new
system is equivalent to the old system of constraints only if the artificial
variables are valued at zero. To guarantee such assignments in the optimal
solution, you can incorporate artificial variables into the objective function
with large positive or large negative coefficients in a minimisation and
maximisation programs respectively. You denote these coefficients by  M.
Whenever artificial variables are part of the initial solution X0, the last row of
simplex table will contain the penalty cost M. You can make the following
modifications in the simplex method to minimise the error of incorporating
the penalty cost in the objective function. This method is called Big M-
method or Penalty cost method.

Sikkim Manipal University Page No. 64


Operations Research Unit 3

Figure 3.2: Big M-method

1) The last row of the simplex table is decomposed into two rows, the first
of which involves those terms not containing M, while the second
involves those containing M.
2) The step 1 of the simplex method is applied to the last row created in the
above modification and followed by steps 2, 3 and 4 until this row
contains no negative elements. Then step 1 of simplex algorithm is
applied to those elements next to the last row that are positioned over
zero in the last row.
3) Whenever an artificial variable ceases to be basic, it is removed from the
first column of the table as a result of step 4; it is also deleted from the
top row of the table as is the entire column under it.
4) The last row is removed from the table whenever it contains all zeroes.
5) If non-zero artificial variables are present in the final basic set, then the
program has no solution. In contrast, zero valued artificial variables in
the final solution may exist when one or more of the original constraint
equations are redundant.

Solved Problem 5
Use Penalty Cost method to
Maximise z = 2x1 + 3x2
Subject to x1 + 2x2  2
6x1 + 4x2  24,
x1, x2  0
Solution: Rewriting in the standard form, we have
Maximise z = 2x1 + 3x2 + 0S1 + 0S2 – M A1
Subject to x1 + 2x2 + S1 = 2
6x1 + 4x2 – S2 + A1 = 24,
x1, x2, S1, S2, A1  0

Sikkim Manipal University Page No. 65


Operations Research Unit 3

The initial simplex table is


Table 3.13: The initial simplex table

x1 x2 S1 S2 A1
2 3 0 0 –M Ratio
S1 0 1* 2 1 0 0 2 2
2
1
A1 – M 6 4 0 –1 1 24 24
4
6
–2 –3 0 0 0 0
– 6M – 4M 0 M –M –24M

Work column
The first iteration gives the following table:
Table 3.14: The first iteration table
x1 x2 S1 S2 A1
2 3 0 0 –M
x1 2 1 2 1 0 0 2
Ai – M 0 –8 –3 –1 1 12
0 1 2 0 0 4
0 8M 6M 1M 0 – 12M
The process is complete as all elements of the last two rows are non
negative. But existence of non-zero artificial variables in the basic
solution shows that the problem has no solution.

Self Assessment Questions


State Yes or No
9. The value of artificial value is “M”.
10. Artificial variables enter as basic variables.

3.6 Two Phase Method

Sikkim Manipal University Page No. 66


Operations Research Unit 3

The drawback of the penalty cost method is the possible computational error
resulting from assigning a very large value to the constant M. To overcome
this difficulty, a new method is considered, where the use of M is eliminated
by solving the problem in two phases.
Phase I: Formulate the new problem. Start by eliminating the original
objective function by the sum of the artificial variables for a minimisation
problem and the negative of the sum of the artificial variables for a
maximisation problem. The Simplex method optimises the ensuing objective
with the constraints of the original problem. If a feasible solution is arrived,
the optimal value of the new objective function is zero (suggestive of all
artificial variables being zero). Subsequently proceed to phase II. If the
optimal value of the new objective function is non-zero, it means there is no
solution to the problem and the method terminates.
Phase II: Start phase II using the optimum solution of phase I as the base.
Then take the objective function without the artificial variables and solve the
problem using the Simplex method.

Solved Problem 6
Use the two phase method to
Maximise z = 3x1 – x2
Subject to 2x1 + x2  2
x1 + 3x2  2
x2  4,
x1, x2  0
Rewriting in the standard form,
Maximise z = 3x1 – x2 + 0S1 – MA1 + 0.S2 + 0.S3
Subject to 2x1 + x2 – S1 + A1 = 2
x1 + 3x2 + S2 = 2
x2 + S3 = 4,
x1, x2, S1, S2, S3, A1  0
Phase 1: Consider the new objective,
Maximise Z* = – A1
Subject to 2x1 + x2 – S1 + A1 = 2
x1 + 3x2 + S2 = 2
x2 + S3 = 4,

Sikkim Manipal University Page No. 67


Operations Research Unit 3

x1, x2, S1, S2, S3, A1  0


The initial simplex table arrived through the simplex method,
Table 3.15: The initial simplex table

x1 x2 S1 A1 S2 S3

0 0 0 –1 0 0 Ratio

A1 –1 2* 1 –1 1 0 0 2 2
=1
2

S2 0 1 3 0 0 1 0 2 2
=2
1

S3 0 0 1 0 0 0 1 4

–2 –1 1 0 0 0 –2
 Work column * pivot element
x1 enters the basic set replacing A1.
The first iteration gives the following table:
Table 3.16: The first iteration table

x1 x2 x1 A1 S2 S3

0 0 0 –1 0 0

X1 0 1 1 1 1 0 0 1

2 2 2

S2 0 0 5 1 1 1 0 1
2 2 –2

S3 0 0 1 0 0 0 1 4

0 0 0 1 0 0 0

Phase I is complete, since there are no negative elements in the last


row.

Sikkim Manipal University Page No. 68


Operations Research Unit 3

The optimal solution of the new objective is Z* = 0.

Phase II:
Consider the original objective function,
Maximise z = 3x1 – x2 + 0S1 + 0S2 + 0S3
x S
Subject to x1 + 2 – 1 =1
2 2
5 S
x2 + 1 + S2 =1
2 2
x2 + S3 = 4
x1, x2, S1, S2, S3  0
With the initial solution x1 = 1, S2 = 1, S3 = 4, the corresponding simplex
table is
Table 3.17: The corresponding simplex table
x1 x2 S1 S2 S3
3 –1 0 0 0 Ratio
x1 3 1 1 1 0 0 1

2 2
S2 0 0 5 1 1 0 1 1
= 2
2 2 * 1
2
S3 0 0 1 0 0 1 4
0 5 2 0 0 3

2 3
 Work column * pivot element
Proceeding to the next iteration, we get the following table:
Table 3.18: The Phase II iteration table
x1 x2 S1 S2 S3
3 –1 0 0 0
x1 3 1 3 0 1 0 2
S1 0 0 5 1 2 0 2
S3 0 0 1 0 0 1 4
0 10 0 3 0 6
Since all elements of the last row are non negative, the current solution is

Sikkim Manipal University Page No. 69


Operations Research Unit 3

optimal.
The maximum value of the objective function
Z = 6 which is attained for x1 = 2, x2 = 0.
Solved Problem 7
Maximise z = 3x1 + 2x2,
Subject to 2x1 + x2  2,
3x1 + 4x2 – S2 + A1  12,
x1, x2  0
Rewriting in the standard form,
Maximise z = 3x1 + 2x2 + 0S1 + 0.S2 – MA1
Subject to 2x1 + x2 + S1 = 2
3x1 + 4x2 – S2 + A1 = 2
x1, x2, S1, S2, A1  0
Solving the above using the two phase method.
Phase I:
Consider the new objective function
Maximise z* = – A1
Subject to 2x1 + x2 + S1 =2
3x1 + 4x2 – S2 + A1 = 12,
x1, x2, S1, S2, A1  0
The initial simplex table
Table 3.19: The initial simplex table
x1 x2 S1 S2 A1 Ratio
0 0 0 0 –1
S1 0 2 1* 1 0 0 2 2
=2
1
A1 – 1 3 4 0 –1 1 12 12
=3
4
–3 –4 0 1 0 –12

 Work column

Sikkim Manipal University Page No. 70


Operations Research Unit 3

The first iteration gives the following table:


Table 3.20: The first iteration table

x1 x2 S1 S2 A1
0 0 0 0 –1
X2 0 2 1 1 0 0 2
A1 – 1 –5 0 –4 –1 1 4
5 0 4 1 0 –4
Since all elements of the last row are non negative, the procedure is
complete. But the existence of non-zero artificial variables in the basic
set indicates that the problem has no solution.

3.7 Solved Problems on Minimisation

Solved Problem 8
Minimise = Z = 3x1 + 8x2
Subject to
x1 + x2 = 200
x1  80
x2  60
x1, x2  0
Solution:
In a standard form
Minimise Z = 3x1 + 8x2 + MA1 + OS1 + MA2 + OS2
Subject to
x1 + x2 +A1 = 200
x1 – S1 + A2 = 80
x2 + S2 = 60
S1, S2, A1, A2  0

Sikkim Manipal University Page No. 71


Operations Research Unit 3

Table 3.21: Simplex Table 1


Cj -3 -8 0 0 -M -M
C.B B.V x1 x2 S1 S2 A1 A3 Qty Ratio
-M A1 1 1 0 0 1 0 200 200
-M A2 1 0 -1 0 0 0 80 80  K R
P.E
0 S2 0 1 0 1 0 1 60 
-2M +3
Zj - Cj -M + 8 M 0 0 0
KC

Table 3.22: Simplex Table 2


-
Cj -3 -8 0 0
M
C.B B.V x1 x2 S1 S2 A1 Qty Ratio Transformation
1 1
-M A1 0 1 1 0 1 120 120 R1 = R1-R2
1
-3 x1 1 0 -1 0 0 80  R2 = R 2
0 S2 0 1 0 1 0 60 60  1
KR R3 = R 3
P.E
-M +
-M
Zj - Cj 0 8 0 0
+3
KC

Table 3.23: Simplex Table 3


Cj -3 -8 0 0

C.B B.V x1 x2 S1 S2 A1 Qty Ratio Transformation

-M A1 0 0 1 -1 1 60 60
1
R1 = R 1
P.E
1
-3 x1 1 0 -1 0 0 80 -ve R2 = R 2
1 1
-8 x2 0 1 0 1 0 60  R3 = R 3 – R 1

-M+
M-
Zj - Cj 0 0 3 0 0
8
KC

Sikkim Manipal University Page No. 72


Operations Research Unit 3

Table 3.24: Simplex Table 4


Cj -3 -8 0 0

Transfor-
C.B B.V x1 x2 S1 S2 Qty Ratio
mation
1
0 S1 0 0 1 -1 60 - ve R1 = R 1
1 1
-3 x1 1 0 0 -1 140 -ve R2 = R 2 + R 1

-8 x2 0 1 0 1 60 80  K 1 1
R R3 = R 1
P.E

Zj - Cj 0 0 0 -5

Table 3.25: Simplex Table 5

Cj -3 -8 0 0

C.B B.V x1 x2 S1 S2 Qty Ratio Transformation

0 S1 0 1 1 0 120 R11 = R1 + R31

-3 x1 1 1 0 0 200 R21 = R2 + R31

0 S2 0 1 0 1 60 R31 = R3

Zj - Cj 0 4 0 0

Since all Zj - Cj  0, the optimum solution is x1 = 200 x2 = 0


Min Z = 60

3.8 Summary
This unit explains how to solve LPP using the simplex method. The unit also
explains the constraints for which you need to introduce the slack, surplus
and artificial variables. Examples are used to illustrate the method of solving
LPP using the simplex method.

Sikkim Manipal University Page No. 73


Operations Research Unit 3

3.9 Terminal Questions


1. Maximise z = 3x1 – x2
Subject to the constraints
2x1 + x2  2
x1 + 3x2  3
x2  4,
x1, x2  0.

2. Minimise Z = 6x1 + 7x2


Subject to the constraints
x1 + 3x2  12
3x1 + x2  12
x1 + x2  8
x1 + x2  0

3.10 Answers to SAQs and TQs


Answers to Self Assessment Questions

1. False
2. True
3. True
4. True
5. Yes
6. Yes
7. No
8. Yes
9. Yes
10. Yes
Answers to Terminal Questions
1. Z = 9 X1 = 3 X2 = 0
2. Z = 5.8 X1 = 8/3 X2 = 6

3.11 References

Sikkim Manipal University Page No. 74


Operations Research Unit 3

No external sources of content have been referred for unit 3

Sikkim Manipal University Page No. 75

Você também pode gostar