CHAPTER-2

LINEAR PROGRAMMING

Linear Programming:

 It is a mathematical technique used to optimize a linear objective function

while satisfying a set of linear constraints.
 It involves maximizing or minimizing a linear function subject to certain
limitations expressed as linear equations or inequalities.

Objective Function:

The objective function defines the goal of the optimization problem, either to
maximize or minimize a certain quantity.

Decision Variables:

Decision variables are the unknown quantities that need to be determined in a

linear programming model. These variables are typically denoted by X 1,X2……Xn

Constraints:

Constraints are limitations or restrictions on the decision variables. They can be

equality or inequality constraints.

Feasible Solution:

A feasible solution is a set of values for the decision variables that satisfies all the
constraints. It lies within the feasible region.

Feasible Region:

The feasible region is the set of all feasible solutions that satisfy all the constraints.
It is the intersection of all constraint boundaries.

Optimal Solution:

The optimal solution is the solution that maximizes or minimizes the objective
function while satisfying all constraints. It is the best possible outcome of the
optimization problem.
Assumptions of Linear programming:

1. Linearity: The relationships between variables in the objective function and

constraints are linear. It means that changes in variables result in proportional
changes in outcomes. For example. if the number of machines in a plant is
increased, the production in the plant also proportionately increases.

2. Additivity: The total contribution of each decision variable is the sum of its
individual contributions, allowing for the additive combination of variables in the
objective function and constraints. For example, the total profit earned by the sale
of three products A, B and C must be equal to the profits earned separately from A,
B and C and similarly, the amount of resources consumed by A, B, and C
individually.

3. Certainty: All parameters, coefficients, and constraints are known with

certainty and do not involve randomness or uncertainty. In other words, this
assumptions means that all the coefficients in the objectives function and
constraints are completely known with certainty and do not change during the
period being studied.

4. Divisibility: This implies that solution values of decision variables and

resources can take any non-negative values, i.e, fractional values of the decision
variables are permitted. This, however, is not always desirable. For example, it is
impossible to produce one-fourth of a bus. When it is necessary to have integer
variables, a technique known as integer programming could be used.

Applications of Linear Programming:

Some common applications of Linear Programming (LP) include:

1. Supply Chain Management: LP is used to optimize production, distribution,
and investment management to minimize costs and maximize efficiency in supply
chains.

2. Production Planning: LP helps in determining the optimal production

quantities of dif products to meet demand while considering constraints such as
resources, labor, and machine capacities.
3.Finance and Portfolio Optimization: LP is applied in portfolio management to
maximize returns while managing risks by allocating investments efficiently.

4. Transportation and Logistics: LP is used to optimize transportation routes,

vehicle scheduling, and distribution networks to minimize transportation costs and
improve delivery efficiency.

5. Marketing and Pricing Strategies: LP assists in determining optimal pricing

strategies, product mix, and marketing campaigns to maximize profits and market
share.

6. Resource Allocation: LP is utilized in allocating resources such as manpower,

funds, and materials effectively to achieve organizational objectives.

7. Energy Management: LP helps in optimizing energy production, distribution,

and consumption to minimize costs and reduce environmental impact.

8. Healthcare Management: LP is applied in healthcare to optimize resource

allocation, staff scheduling, and hospital operations to improve patient care and
reduce costs.

9. Agricultural Planning: LP assists in optimizing crop selection, land use, and

resource allocation in agriculture to maximize yields and profitability.

10. Scheduling and Timetabling: LP is used in scheduling tasks, workforce

management, and timetabling to optimize resource utilization and improve
operational efficiency.

These applications demonstrate the versatility and effectiveness of Linear

Programming in solving complex optimization problems across diverse industries
and domains.

Advantages of Linear Programming (LP)

Optimization: Linear Programming provides a systematic approach to optimizing

resources by maximizing efficiency and minimizing costs in various applications.

Flexibility: LP models can be adapted to different scenarios and constraints for the
optimization of complex systems with multiple variables.
Mathematical Precision: LP solutions are based on mathematical algorithms to
provide precise and quantifiable results that can guide decision-making processes.

Resource Allocation: LP helps in efficient allocation of resources to improve

productivity and better utilization of available resources.

Decision Support: LP models offer decision support by providing insights into the
best course of action under given constraints. It helps in strategic planning and
problem-solving

Disadvantages of Linear Programming (LP):

Linearity Assumption: LP assumes linear relationships between variables and

constraints, limiting its applicability to problems with non-linear relationships.

Sensitivity to Data: LP solutions can be sensitive to changes in input data. Hence

it is less robust in dynamic or uncertain environments.

Complexity: Solving large-scale LP problems can be computationally intensive

and time-consuming especially when dealing with a high number of variables and
constraints.

Assumption of Certainty: LP models often assume certainty in data and

parameters, which may always reflect real-world uncertainties and fluctuations.

Modeling Constraints: Formulating LP models requires careful consideration of

all constraints objectives, and incorrect modeling assumptions can lead to
suboptimal solutions.
2.8 Linear Programming Problem Formulation:

2.8.1 Process of Linear Programming Problem Formulation

A systematic approach is required to formulate a Linear Programming Problem

(LPP) effectively. The following steps outline the process of formulating an LPP

Step 1: Identify the Nature of the Problem

Determine whether the problem is a maximization or minimization problem. This

decision will define the direction in which the objective function is optimized

Step 2:Identify the Number of Variables

Determine the number of decision variables required to represent the problem.

Each variable corresponds to a quantity that needs to be determined.

Step 3: Formulate the Objective Function

Construct the objective function that needs to be optimized. The objective function
typically a linear combination of the decision variables with coefficients
representing the contribution of each variable to the objective.

Step 4: Formulate the Constraints

Define the constraints that restrict the feasible solutions of the problem. These
constraint are derived from the problem requirements and are expressed as linear
inequalities involving the decision variables.

Step 5: Develop Non-negativity Constraints

Include non-negativity constraints for each decision variable to ensure that the
variables cannot take negative values. This is a common practice in linear
programming to align with real-world scenarios.
Mathematical Formulation of a Linear Programming Problem
A Linear Programming (LP) problem can be mathematically formulated as
follows:
1. Objective Function: The objective function in an LP problem represents the
quantity to be maximized or minimized. It is typically a linear combination of
decision variables. The general form of the objective function is
(Maximize or Minimize) Z= c1x1+c2x2+c3x3+……………cnxn
Where
Z is the objective function to be optimized.

C1, C2 ,...,C n are the coefficients of the decision variables x1,x2,……xn

2. Constraints: Constraints in an LP problem represent the limitations or

restrictions on the decision variables. They are expressed as linear inequalities.
a11x1+a12x2+a13x3+a1nxn(≤,≥,=)b1
a21x1+a22x2+a23x3+a2nxn(≤,≥,=)b2
------------------------------------------
--------------------------------------------------------

am1x1+am2x2+am3x3+amnxn(≤,≥,=)bm
Where:
aij are the coefficients of the decision variables in the ith constraint.

Bi is the right-hand side constant of the ith constraint.

m is the number of constraints.

Each constraint may be either ≤ (less than or equal to), ≥ (greater than or equal
to),= (equal to)

3. Non-negativity Constraints: In addition to the objective function and

constraints, non- negativity constraints are often included in LP problems to ensure
that decision variables cannot take negative values
4. Mathematical Formulation: Combining objective function, constraints, and
non- negativity constraints, a general mathematical formulation of an LP problem
is:
(Maximize or Minimize)Z= c1x1+c2x2+c3x3+……………cnxn
Subject to
a11x1+a12x2+a13x3+a1nxn(≤,≥,=)b1
a21x1+a22x2+a23x3+a2nxn(≤,≥,=)b2
------------------------------------------
--------------------------------------------------------

am1x1+am2x2+am3x3+amnxn(≤,≥,=)bm
and x1≥0, x2≥0, xn≥0

This formulation represents the foundation of a Linear Programming problem,

where the goal is to optimize the objective function subject to linear constraints
and non-negativity constraints on the decision variables.
PROBLEM

A manufacturer produces two types of models M1 and M2. Each model of the
type M1 requires 4 hours of grinding and 2 hours of polishing: whereas each model
of the type M2 requires 2 hours of grinding and 5 hours of polishing. The
manufacturers have 2 grinders and 3 polishers. Each grinder works 40 hours a
week and each polisher works for 60 hours a week. The profit on M1 model is 3.00
and on model M2 is 4.00Whatever is produced in a week is sold in the market.
How should the manufacturer allocate his production capacity to the two types of
models, so that he may take the maximum profit in a week?
Solution:
i. Identify and define the decision variable of the problem.
Let x1 _and X2 the number of units of M1 and M2 model.

ii. Define the objective function.

Since the profits on both the models are given, the objective function is to
maximize the profit Max Z = 3X 1+ 4X2
iii. State the constraints to which the objective function should be optimized (i.e.
Maximization or Minimization).

There are two constraints one for grinding and the other for polishing.

Grinding Constraint: The total hours of grinding required for M1 and M2 should
not exceed the available grinding hours: No. of hours available on grinding
machine per week is 40 hours. There are two grinders. Hence the total grinding
hours available are 40 * 2 = 80 hours.

The grinding constraint is given by 4X1 + 2X2 ≤ 80

Polishing Constraint: The total hours of polishing required for M1 and M2 should
not exceed the available polishing hours. No. of hours available on polishing
machine per week is 60 hrs There are three grinders. Hence the total grinding hours
available are 60 * 3 = 180 hours.

The polishing constraint is given by 2X1+ 5X2 ≤ 180

Non-negativity Constraint: The number of units produced cannot be negative:

X1 , X2 ≥ 0 Finally we have,

Max Z = 3X 1+ 4X2
Subject to constraints,
4X1 + 2X2 ≤ 80
2X1+ 5X2 ≤ 180
X1, , X2 ≥ 0
Methods for the Solutions of a Linear Programming:
1. Graphical Solution:

 Suitability: The graphical method is suitable for problems with two decision
variables
 Visualization: It involves graphically representing the constraints as lines
on a coordinate plane to determine the feasible region where all constraints
are satisfied.
  Optimal Solution: The optimal solution is found at the corner point of the
feasible region that maximizes or minimizes the objective function
 Limitation: Dealing with more than two variables using the graphical
method becomes complex as it involves dealing with planes instead of
straight lines

2. Simplex Method:
Suitability: The simplex method is the most powerful method for solving
problems with more than two decision variables.
Algorithm: It is an iterative algorithm that systematically moves from one basic
feasible solution to another along the edges of the feasible region until the optimal
solution is reached
Efficiency: The simplex method is efficient for solving large-scale linear
programming problems with multiple variables.
Optimality: It guarantees finding the optimal solution in a finite number of steps
for bounded problems
Complexity: While more complex than the graphical method, the simplex method
is widely used due to its effectiveness in handling higher-dimensional problems.

Graphical Solution:
The Graphical Method is a technique used to solve linear programming problems
with two decision variables. It involves graphically representing the constraints and
objective function on a coordinate plane to determine the feasible region and find
the optimal solution.

Procedure for Solving LPP by Graphical Method

The steps involved in the graphical method are as follows:

Step 1:Formulate the Linear Programming Problem

 Objective Function: Define the objective function to be maximized or

minimized.
 Constraints: Identify the constraints (linear inequalities) that restrict the
feasible region.
Step 2: Plot the Constraints

Convert Inequalities to Equations: Convert each constraint into an equation by

replacing the inequality sign with an equality sign. For example, if the constraint is
2x + y ≤ 6 convert it into 2x + y = 6

Plot the constraints: Draw the coordinate axes (x and y axes) on a graph. Plot
each equation as a straight line by finding two points that satisfy the equation and
connecting them

Step 3: Identify Feasible Region

Intersection of Constraints: The feasible region is the area where all constraints

overlap.

Feasible Region: Shade the region below the line for constraints with ≤ sign and
above the line for constraints with ≥ sign in the first quadrant. The overlapping
shaded region is the feasible region where all constraints are satisfied
simultaneously.

Step 4: Determine Corner Points

Corner Points: Identify the corner points of the feasible region where the
constraints intersect.

Calculate Coordinates: Determine the coordinates of each corner point.

Step 5: Evaluating Objective Function

 Evaluate the objective function at each corner point to determine the optimal
solution.
 Substitute the coordinates of the corner points into the objective function to
calculate the objective function value at each point.

Selecting the Optimal Solution:

 Compare the objective function values at the corner points to identify the
optimal solution that maximizes or minimizes the objective function.
 PROBLEMS

1.Solve the following LPP by graphical method

Minimize, Z=20x1+10x2

Subject to x1+2x2 ≤ 40

3x1+x2 ≥ 30

4x1+3x2 ≥ 60

x1,x2 ≥ 0

Solution:

Replace all the inequalities of the constraints by equation,

 1.x1+2x2 = 40

If x1=0 →x2=20

If x2=0 →x1=40

x1+2x2 = 40 passes through (0,20) (40,0)

 2. 3x1+x2 = 30

If x1=0 →x2=30

If x2=0 →x1=10

3x1+x2 = 30 passes through (0,30) (10,0)

 3. 4x1+3x2 = 60

If x1=0 →x2=20

If x2=0 →x1=15

4x1+3x2 = 60 passes through (0,20) (15,0)

Plot each equation on the graph.

The feasible region is ABCD. C and D are points of intersection of lines.

x1+2x2 = 40

3x1+x2 = 30 and

4x1+3x2 = 60

On solving we get C(4,18) and D(6,12)

Value of Z=20x1+10x2

A(15,0) Z=20*15+10*0=300
B(40,0) Z=20*40+10*0=800
C(4,18) Z=20*4+10*18=260
D(6,12) Z=20*6+10*12=240(Minimum value)
Therefore the minimum value of Z occurs at D(6,12) and the optimal solution is
x1=6,x2=12
2.Find the maximum value of Z = 5x1+7x2

Subject to the Constraints,

x1+x2 ≤ 4

3x1+8x2 ≤ 24

10x1+7x2 ≤ 35

x1, x2 ≥ 0

Solution:

Replace all the inequalities of the constraints by equations

1.x1+x2 = 4

If x1=0 →x2=4

If x2=0 →x1=4

x1+x2 = 4 passes through (0,4) (4,0)

2. 3x1+8x2 = 24

If x1=0 →x2=3

If x2=0 →x1=8

3x1+8x2 = 24 passes through (0,3) (8,0)

3. 10x1+7x2 = 35

If x1=0 →x2=5

If x2=0 →x1=8

10x1+7x2 = 35 passes through (0,5) (8,0)

Plot these lines in the graph and mark the region below the line as the inequality of
the constraint is ≤ and is also lying in the first quadrant.
The feasible region is OABCD

B and C are points of instruction of lines,

On solving we get, B (1.6,2.3) C (1.6,2.4)

Corner points Value of Z=5x1+7x2

O(0,0) Z=5*0+7*0=0
A(3.5,0) Z=5*3.5+7*0=17.5
B(1.6,2.3) Z=5*1.6+7*2.3=24.1
C(1.6,2.4) Z=5*1.6+7*2.4=24.8(Maximum value)
D(0,3) Z=5*0+7*3=21

Therefore the maximum value of Z occurs at c(1.6,2.4) and the optimal solution is
x1=1.6,x2=2.4
Simplex Method

 There are many methods to solve the Linear Programming Problem, such as
Graphical Method, Trial and Error method, Vector method and Simplex
Method. We use graphical method for solution when we have two
problem variables, the other method can be used when there are more than
two decision variables in the problem.
 Among all the methods, SIMPLEX METHOD is most powerful method. It
deals with iterative process, which consists of first designing a Basic
Feasible Solution or a Programme and proceed towards the OPTIMAL
SOLUTION and testing each feasible solution for Optimality to know
whether the solution on hand is optimal or not If not an optimal solution,
redesign the programme, and test for optimality until the test confirms
OPTIMALITY

2.11.1 Comparison Between Graphical and Simplex Methods

1. The graphical method is used when we have two decision variables in the
problem. Whereas in Simplex method, the problem may have any number of
decision variables.

2. In graphical method, the inequalities are assumed to be equations, so as to

enable to draw straight lines. But in Simplex method, the inequalities are converted
into equations by:

a. Adding a SLACK VARIABLE in maximization problem and subtracting a

SURPLUS VARIABLE in case of minimization problem.

3. In graphical solution the Isoprofit line moves away from the origin to towards
the far off point in maximization problem and in minimization problem, the Isocost
line moves from far off distance towards origin to reach the nearest point to origin.

4. In graphical method, the areas outside the feasible area (area covered by all the
lines of constraints in the problem) indicates idle capacity of resource where as in
Simplex method, the presence of slack variable indicates the idle capacity of the
resources.
5. In graphical solution, if the isoprofit line coincides with more than one point of
the feasible polygon, then the problem has second alternate solution. In case of
Simplex method the net evaluation row has zero for non-basis variable the problem
has alternate solution. (If two alternative optimum solutions can be obtained, the
infinite number of optimum, solutions can be obtained).

Slack Variables:

 Slack variables represents unused resources, in the form of labor, material,

time, money, warehouse space etc.. Since unused resources variables yield
no profit, so cost coefficient of these variables is taken as zero in objective
function.
 The inequality of greater than equal to type (≥ 0) is converted into
equation(=) by subtracting non negative surplus variable.

Surplus variable:

A surplus variable represents amount by which solution value exceeds

resources. These are also called negative slack variables. The coefficient of a
surplus variable is taken zero in objective function
 Problems

Consider the following example

Maximize P= 30x+40y

Subject to 2x+y ≤ 10

x+y ≤ 7

x+2y ≤ 12

x,y ≥ 0

Solution:

The standard form of an LP is where all the constraints are equations and all
variables are non- negative.
The following transformation steps can be followed to convert all the inequalities
into equality constraints.

Step 1: If a constraint is of type (x + y ≤ c) , use a slack variable 's' (where 's' is

non-negative), such that x+y+s = c

Step 2: If a constraint is of type (x+y ≥ c) , use a surplus variable's' (where's' is

non-negative), such that x+y-s = c

Using this transformation method on our example we get the following system of
equations.
P-30x-40y = 0
2x+y+s1=10
x+y+s2=7
x+2y+s3=12
x,y ≥ 0

Step 3: Set up the initial simplex tableau

We need to create a tableau from the system of equations we just obtained. It looks
something like this.
x y s1 s2 s3 P C
2 1 1 0 0 0 10
1 1 0 1 0 0 7
1 2 0 0 1 0 12

-30 -40 0 0 0 1 0

Step 4 : Identify the pivot

In the simplex tableau, the pivot is identified using the below two conditions

1. The pivot column is chosen by identifying the most negative entry in the bottom
row of the tableau

2. The pivot row is identified by dividing the constants (C column in the tableau)
by the coefficients of the pivot column. The row containing the smallest quotient
would be the pivot row.

In our example, the ‘y’ column is the pivot column since it contains - 40 and the
pivot row is R3 (row 3), since 12/2 = 6 is the smallest coefficient in the pivot
column. Hence, our pivot element is 2.

x y s1 s2 s3 P C
2 1 1 0 0 0 10
1 1 0 1 0 0 7
1 2 0 0 1 0 12 12/2=6
-30 -40 0 0 0 1 0

After we identified the pivot element, perform row transformations on R1,R2,R3,

R4 ) . We will transform the pivot element to a 1 and all of the of other elements
in the pivot column as 0.In this case, we will use the following row transforms in
order to get our result

 R3→R3/2
 R1→R1-R3
 R2→R2-R3
 R4→R4+40R3
  R3→R3/2

Old R3

1 2 0 0 1 0 12
2 2 2 2 2 2 2

New R3 [1/2, 1 ,0 , 0 ,1/2 , 0 , 6]

 R1→R1-R3
Old R1 New R3

[2 1 1 0 0 0 10] – [1/2 1 0 0 1/2 0 6]

New R1 [3/2, 0, 1, -1/2, 0 , 4]

 R2→R2-R3
Old R2 New R3

[1 1 0 1 0 0 7] - [1/2 1 0 0 1/2 0 6]

New R2 [1/2, 0, 0, -1/2, 0, 1]

 R4→R4+40R3
R4 40R3

[-30 -40 0 0 0 1 0] + 40 [1/2 1 0 0 1/2 0 6]

[-30 -40 0 0 0 1 0] +[20 40 0 0 20 1 240]

New R4 [-10 0 0 0 20 1 240]

Output tableau will be

x y s1 s2 s3 P C
3/2 0 1 0 -1/2 0 4
1/2 0 0 1 -1/2 0 1
1/2 1 0 0 1/2 0 6

-10 0 0 0 20 1 240

Iteration 2

Using the procedure in step 3 we identify our new pivot table as ½ in R2

x y s1 s2 s3 P C
3/2 0 1 0 -1/2 0 4
1/2 0 0 1 -1/2 0 1
1/2 1 0 0 1/2 0 6

-10 0 0 0 20 1 240

We will use the row transformations to make all entries 0 in the pivot 1

 R2→2 R2
 R1→ R1 -3/2 R2
 R3→ R3-1/2 R2
 R4→R4+10R2

x y s1 s2 s3 P C
0 0 1 -3 1 0 1
1 0 0 2 -1 0 2
0 1 0 -1 1 0 5

0 0 0 20 10 1 260

Setting the non- basic variables to 0, we get x=2, y=5, s1=1 and P=260

Hence our optimal values for x and y are 2,5

Artificial Variables :

Artificial variables, are introduced in linear programming to convert inequality

constraints into equality constraints.

In linear programming, the objective is to maximize or minimize a linear function

subject to a set of linear constraints, which constraints can be in the form of
equalities or inequalities. However the simplex method which is commonly used to
solve linear programming problems, requires all constraints to be in the form of
equalities.
To convert inequality constraints into equality constraints, artificial variables are
introduced. These variables are added to the problem as if they were decision
variables, but they do not represent any physical quantity. They are used only to
Convert inequality constraints into equality constraints.

The Big M Method

 The Big M method is another method of removing artificial variables from

the basis. In this we assign coefficients to artificial variables, undesirable
from the objective function.
 If objective function Z is to be minimized, then a very large positive
price(called penality) is assigned to each artificial variable. Similarly, if Z is
to be maximized, then a very large negative price (also calleg penalty) is
assigned to each of these variables. The penalty will designated by - M for a
maximization problem and +M for a minimization problem, where M > 0

BigM Method Steps (Rules)

Step-1:

Formulate the Problem

a .Formulate the mathematical model of the given linear programming problem.

b. All the XBi > 0. So if any XBi < 0 then multiply the corresponding constraint by
-1 to make XBi> 0. So sign ≤ changed to ≥ and vice varsa.
c. Convert LP problem into standard form by adding slack, surplus and artificial
variables for each constraint.

d. In the objective function, assign 0 coefficient to slack and surplus variables.

Then add-M (for Maximization case) or +M (for Minimization case) to artificial
variables. (M is very large positive value)
Step-2:
Find out the Initial basic solution
Find the initial basic feasible solution by setting zero value to the decision
variables.
Step-3
Test for optimality:
a. Calculate the values of cj-zj in the last row of simplex table.
b. For Maximization case
1.if all cj-zj ≤ 0, the current basic feasible solution is the optimal solution.
2.In cj-zj > 0 then select the variable that has largest cj-zj and enter this variable
into the table. This column is called key column(pivot column).
c. For Minimization case
1.if all cj-zj ≥ 0, the current basic feasible solution is the optimal solution.
2.In cj-zj < 0 then select the variable that has smallest cj-zj and enter this variable
into the table. This column is called key column(pivot column).
Step-4
Test for Feasibility(variables to leave the basis)
a. Find the ratio by dividing the values of XB column by the positive values of
column (say aij > 0)
b. Find the minimum ratio and this row is called key row(Pivot row) and
corresponding variable will leave the solution.
c. The intersection element of key row and key column is called key element(pivot
element).
Step-5
Determine the new solution:
a. The new values of key row can be obtained by dividing the row elements by the
pivot element.
b. The number in the remaining rows can be computed by utilizing the following
formula
Row(new)=Row (old)-(values of key column and row(old))x keyRow(new)
Step-6
Repeat the procedure
Go to step 3and repeat the procedure until all the values of cj-zj ≤ 0 (For
maximization case) or cj-zj ≥ 0 (for minimization case)

PROBLEMS

1. Find solution using Simplex method (BigM method)

Min z = 2000x1+ 1500 x2
Subject to 6x1+ 2x2 ≥ 8
2x1+4 x2 ≥ 12
4x1 +12 x2 ≥ 24
and x1, x2 ≥ 0

Solution
Min z = 2000x1+ 1500 x2
Subject to
6x1+ 2x2 ≥ 8
2x1+4 x2 ≥ 12
4x1 +12 x2 ≥ 24
and x1, x2 ≥ 0
The problem is converted into canonical form by adding slack, surplus and
artificial variables as appropriate
1.As the constraint -1 is of type ≥ we should subtract surplus variable S1 and add
artificial variable A1
2. As the constraint -1 is of type ≥ we should subtract surplus variable S1 and add
artificial variable A2
3. As the constraint -1 is of type ≥ we should subtract surplus variable S1 and add
artificial variable A3
After introducing surplus, artificial variables
Min Z=2000 x1 +1500 x2+0s1+0s2+0s3+MA1+MA2+MA3
Subject to
6x1+ 2x2 -S1 +A1 =8
2x1+4 x2 -S2 +A2 =12
 4x1 +12 x2 –S3 + A3 = 24
and x1, x2, S1, S2, S3, A1, A2, A3 ≥ 0
Iteration-1 Cj 2000 1500 0 0 0 M M M
B C XB x1 x2 S1 S2 S3 A1 A2 A3 Min Ratio
B XB/X2
A1 M 8 6 2 -1 0 0 1 0 0 8/2=4
A2 M 12 2 4 0 -1 0 0 1 0 12/4=3
A3 M 24 4 (12) 0 0 -1 0 0 1 24/12=2
Z = 44M Zj 12M 18M -M -M -M M M M
Zj-Cj 12M-2000 18M-1500 -M -M -M 0 0 0

Positive maximum Zj-Cj is 18M-1500 and its column index is 2.So,the entering
variable is x2
Minimum ratio is 2 and its row index is 3.So,the leaving variable is A3
⸫ The pivot element is 12.
Entering = x2, Departing = A3, Key element = 12
R3(new) = R3(old) ÷12

R3(old) 24 4 12 0 0 -1 0 0
R3(new) = R3(old) ÷ 12 2 0.3333 1 0 0 -0.0833 0 0

R1(new) = R1(old) -2 R3(new)

R1(old) 8 6 2 -1 0 0 1 0
R3(new) 2 0.3333 1 0 0 -0.0833 0 0
2x R3(new) 4 0.6667 2 0 0 -0.1667 0 0
R1(new) = R1(old) -2 R3(new) 4 5.3333 0 -1 0 0.1667 1 0

R2(new) = R2(old) -4 R3(new)

 R2(old) 12 2 4 0 -1 0 0 1
R3(new) 2 0.3333 1 0 0 -0.0833 0 0
4x R3(new) 8 1.3333 4 0 0 -0.3333 0 0
R2(new) = R2(old) -4 R3(new) 4 0.6667 0 0 -1 0.3333 0 1

Iteration-2 Cj 2000 1500 0 0 0 M M

B CB XB x1 x2 S1 S2 S3 A1 A2 Min
Ratio
XB/X1
A1 M 4 (5.3333) 0 -1 0 0.1667 1 0 4/5.3
333
=0.75
A2 M 4 0.6667 0 0 -1 0.3333 0 1 4/0.6
667=
6
x2 1500 2 0.3333 1 0 0 -0.0833 0 0 2/0.3
333=
6
Z Zj 6M+500 1500 -M -M M2- M M
=8M+3000 125
Zj-Cj 6M-1500 0 -M -M M2-125 0 0

Positive maximum Zj-Cj is 6M-1500 and its column index is 1.So,the entering
variable is x1
Minimum ratio is 0.75 and its row index is 1.So,the leaving variable is A1
⸫ The pivot element is 5.3333
Entering = x1, Departing = A1, Key element = 5.3333

R1(new) = R1(old) -5.3333

 R1(old) 4 5.3333 0 -1 0 0.1667 0

R1(new) = 0.75 1 0 -0.1875 0 0.0312 0

R1(old) -5.3333

R2(new) = R2(old) -0.6667 R1(new)

R2(old) 4 0.6667 0 0 -1 0.3333 1

R1(new) 0.75 1 0 -0.1875 0 0.0312 0
0.6667 XR1(new) 0.5 0.6667 0 -0.125 0 0.0208 0
R2(new) = R2(old) - 3.5 0 0 0.125 -1 0.3125 1
0.6667 R1(new)

R3(new) = R3(old) -0.3333R1(new)

R3(old) 2 0.3333 1 0 0 -0.0833 0

R1(new) 0.75 1 0 -0.1875 0 0.0312 0
0.3333R1(new) 0.25 0.3333 0 -0.0625 0 0.0104 0
R3(new) = 1.75 0 1 0.0625 0 -0.0938 0
R3(old) -
0.3333R1(new)

Iteration-3 Cj 2000 1500 0 0 0 M

B CB XB x1 x2 S1 S2 S3 A2 Min
 Ratio
XB/S3
X1 2000 0.75 1 0 -0.1875 0 0.0321 0 0.75/0.0
312=24

A2 M 3.5 0 0 0.125 -1 0.3125 1 3.5/0.31

25=11.2
x2 1500 1.75 0 1 0.0625 0 -0.0938 0 --------
Z Zj 2000 1500 0.125M -M 0.3125 M
=3.5M+4125 -281.25 M-
78.125
Zj-Cj 0 0 0.125M -M 0.3125 0
-281.25 M-
78.125

Positive maximum Zj-Cj is 5M16-6258 and its column index is 5.So,the entering
variable is S3
Minimum ratio is 11.2 and its row index is 2.So,the leaving variable is A2
⸫ The pivot element is 0.3125
Entering = S3, Departing = A2, Key element = 0.3125

R2(new) = R2(old) ÷0.3125

R2(old) 3.5 0 0 0.125 -1 0.3125
R2(new) = R2(old) 11.2 0 0 0.4 -3.2 1
÷0.3125

R1(new) = R1(old) -0.0312 R2(new)

R1(old) 0.75 1 0 -0.1875 0 0.3125
R2(new) 11.2 0 0 0.4 -3.2 1
0.0312 R2(new) 0.35 0 0 0.0125 -0.1 0.0312

R3(new) = R3(old) +0.0938R2(new)

R3(old) 1.75 0 1 0.0625 0 -0.0938
R2(new) 11.2 0 0 0.4 -3.2 1
 0.0938R2(new) 1.05 0 0 0.0375 -0.3 0.0938
R3(new) = 2.8 0 1 0.1 -0.3 0
R3(old)
+0.0938R2(new)

Iteration-4 Cj 2000 1500 0 0 0

B CB XB x1 x2 S1 S2 S3 Min Ratio
XB/S3
X1 2000 0.4 1 0 -0.2 0.1 0 ------------
S3 0 11.2 0 0 0.4 -3.2 1 11.2/0.4
=28
x2 1500 2.8 0 1 0.1 -0.3 0 2.8/0.1 =28
Z =5000 Zj 2000 1500 -250 -250 0

Zj-Cj 0 0 -250 -250 0

Since all Zj-Cj ≤ 0

Hence, optimal solution is arrived with value of variables as :

x1 =0.4 , x2 = 2.8

Min z=5000

Degeneracy:

During solving LP problem , a solution may arise in which there is a tie between 2
or more basic variables for leaving the basis.(minimum ratios are same).It is called
degeneracy

To resolve this we can select any of them arbitrarily. but if artificial variable is
present then it must be removed first.

PROBLEM
1.Find the solution using Simplex (Big M) method

Max Z=3x1+9x2

Subject to

x1+4 x2 ≤ 8

x1+2 x2 ≤ 4

and x1, x2 ≥ 0

Solution:

Max Z=3x1+9x2

Subject to

x1+4 x2 ≤ 8

x1+2 x2 ≤ 4

and x1, x2 ≥ 0

The problem is converted into canonical form by adding slack, surplus and
artificial variables as appropriate
1.As the constraint -1 is of type ≤ we should add slack variable S1
2. As the constraint -1 is of type ≤ we should add slack variable S2

After introducing Slack Variables

Max Z= 3x1+9x2+0s1+0s2
Subject to
x1+4x2+S1=8
x1+2x2+S2=4
and x1, x2, S1, S2 ≥ 0

Iteration 1 Cj 3 9 0 0
 B CB xB x1 x2 S1 S2 Min ratio
xB/ x2

S1 0 8 1 4 1 0 8/4=2
S2 0 4 1 2 0 1 4/2=2
Z=0 Zj 0 0 0 0
Cj-zj 3 9 0 0

Positive maximum Zj-Cj is 9 and its column index is 2.So,the entering variable is
x2
Minimum ratio is 2 and its row index is 2.So,the leaving variable is S2
⸫ The pivot element is 2
Entering = x2, Departing = S2, Key element = 2
R2(new) = R2(old) ÷2

R2(old) 4 1 2 0 1
R2(new) = R2(old) ÷2 2 1/2 1 0 1/2

R1(new) = R1(old) -4 R2(new)

R1(old) 8 1 4 1 0
R1(new) = R1(old)-4R2(new) 0 -1 0 1 -2

Iteration 2 Cj 3 9 0 0
B CB xB x1 x2 S1 S2 Min ratio
xB/ x2

S1 0 0 -1 0 1 -2
x2 9 2 ½ 1 0 ½
Z=18 Zj 9/2 9 0 9/2
Cj-zj -3/2 0 0 -9/2

Since all Zj- Cj ≤ 0

Hence optimal solution is arrived with value of variables as :

x1 = 0, x2 = 2

Max z =18

Two-Phase Method

Find solution using Two-Phase method

MIN Z=5x1+2x2+10x3

Subject to x1 – x3 < = 10

x2 + x3 > =10

and x1, x2, x3 > = 0

Solution:

MIN Z=5x1+2x2+10x3

Subject to x1 – x3 < = 10

x2 + x3 > =10

and x1, x2, x3 > = 0

Phase – 1

The problem is converted into canonical form by adding slack, surplus and
artificial variables as appropriate
1.As the constraint -1 is of type ≤ we should add slack variable S1
2. As the constraint -2 is of type ≥ we should subtract surplus variable S2 and add
artificial variable A1

After introducing Slack, surplus, artificial variables

Min Z = A1
 Subject to
x1 – x3 + S1 = 10

x2 + x3 – S2 + A1=10

and x1,x2,x3

Iteration-1 Cj 0 0 0 0 0 1
B CB XB x1 x2 x3 S1 S2 A1 Min Ratio
XB/ x2
S1 0 10 1 0 -1 1 0 0 -----------
A1 1 10 0 (1) 1 0 -1 1 10/1=10
Z=10 Zj 0 1 1 0 -1 1
Cj-zj 0 1 1 0 -1 0

Positive maximum Zj-Cj is 1 and its column index is 2.So,the entering variable is x2
Minimum ratio is 10 and its row index is 2.So,the leaving variable is A1
⸫ The pivot element is 1
Entering = x2, Departing = A1, Key element = 1
R2(new) = R2(old)

R2(old) 10 0 1 1 0 -1

R2(new) =R2(old) 10 0 1 1 0 -1

R1(new) = R1(old)

R1(old) 10 1 0 -1 1 0

R1(new) =R1(old) 10 1 0 -1 1 0

Iteration-2 Cj 0 0 0 0 0
B CB XB x1 x2 x3 S1 S2 Min Ratio
XB/ x2
S1 0 10 1 0 -1 1 0
x2 0 10 0 1 1 0 -1
Z=0 Zj 0 0 0 0 0
Cj-zj 0 0 0 0 0

Since all Zj- Cj ≤ 0

Hence optimal solution is arrived with value of variables as :

x1 = 0, x2 = 10, x3=0

Max Z =0

Phase-2

We eliminate the artificial variables and change the objective function for the
original,

Min Z = 5x1+2x2+10x3+0S1+0S2

Iteration-1 Cj 5 2 10 0 0
B CB XB x1 x2 x3 S1 S2 Min Ratio
XB/ x2
S1 0 10 1 0 -1 1 0
x2 2 10 0 1 1 0 -1
Z=20 Zj 0 2 2 0 -2
Cj-zj -5 0 -8 0 -2

Since all Zj- Cj ≤ 0

Hence, optimal solution is arrived with value of variables as :

x1 = 0, x2 = 10, x3=0

Min Z =20
 Chapter-3
Transportation Problem
Definition:
 The transportation problem is a classic optimization problem in
operations research and logistics.
 It deals with determining the most cost-effective way to transport
goods from multiple suppliers to multiple destination satisfying
 supply and demand constraints while minimizing total
transportation costs.
Rim Condition:
The total supply available from all suppliers must equal the total demand
at all destinations. In other words, the total quantity of goods shipped
from all suppliers must equal the total quantity demanded by all
destinations. This is called the rim condition.
 The goal of the transportation problem is to determine how much
to ship from each supplier to each destination in order to minimize
total transportation costs while satisfying supply and demand
constraints.
Balanced Transportation Problem:
* The problem is considered to be a balanced transportation problem
when both supplies and demands are equal.
Unbalanced Transportation Problem:
Unbalanced transportation problem is defined as a situation in which
supply and demand are not equal. A dummy row or a dummy column is
added to this type of problem, depending o the necessity, to make it a
balanced problem. The problem can then be addressed in the same way
as the balanced problem.
Feasible Solution:

A set of non-negative values xij i = 1 ,2....m;j=1,2,....n. that satisfies the constraints

(rim condition and also the non-negativity restrictions) is called a feasible solution
to the transportation problems

A balanced transportation problem will always have a feasible solution.

Basic Feasible Solution (BFS)

A feasible solution to a ( m* n) transportation problem that contains no more that
m + n - 1 non negative allocation is called basic feasible solution (BFS) to the
transportation problem.

Non-Degenerate Basic Feasible Solution

A basic feasible solution to a ( m * n) transportation problem is said to be a non-

degenerate basic feasible solution if it contains exactly m + n - 1 non-negative
allocations in independent positions.

Degenerate Basic Feasible Solution

A basic feasible solution that contains less than m + n - 1 non-negative allocation is

said to be degenerate basic feasible solution.

Optimal Solution

A feasible solution (not necessarily basic) is said to be an optimal solution if it

minimizes the total transportation cost.

 Initial Basic Feasible Solution

There are several methods available to obtain an initial basic feasible solution.
However, we shall discuss here the following three methods:

 North-West Corner Method.

 Least-Cost Method.
 Vogel's Approximation Method

North-West Corner Method:

It is a simple and efficient method to obtain an initial basic feasible solution.
Various steps of the method are:

North-West Corner Method (NWCM) Steps (Rule)

sep

Seled this valuer cell of the transportation matrix and allocate min(sd) a.
Subtract this value from supply and demand of respective row and column.

Step-2:

b. If the supply is 0, then cross (strike) that row and move down to the next
cell.

c. If the demand is 0, then cross (strike) that column and move right to the
next cell.

d. If supply and demand both are 0, then cross (strike) both row & column
and move diagonally to the next cell.

step-3:

Repeat this steps until all supply and demand values are 0.
Illustration 1

Find the solution using North-West Corner Method

D D D D D. Supply

2 11 10 37

721

3948 12

9
Demand

33

456

Solution

Total number of supply constraints:3

Total number of demand constraints: 5

Problem Table is

Supply

2 11 10 3

14
72

39

812

Demand 3

3456

The rim values for S_{1} = 4 and D_{1} = 3 are compared.

The smaller of the two Le min(4,3) = 3 is assigned to (S_{1}*D_{1})

This meets the complete demand of D, and leaves 4-3-1 units with S_{1}

Table-1:

DDDD D_{2}

Supply

2(3)

11

10 3

4
72

8 12

Demand

3456