XII\Maths\Notes\LPP Amit Gupta

LINEAR PROGRAMMING
Consider the following problem.
A furniture dealer deals in only two items – tables and chairs. He has ` 11000 to invest
and space to store at most 40 pieces. A table costs him ` 500 and chair costs him ` 200.
He can sell a table at a profit of ` 50 and a chair at a profit of ` 25. Assume that he can sell
all the items that he buys, determine the number of tables and chairs he must purchase
so as to maximise his profit.
Several scenarios that can come up are tabulated below:
Scenario # Tables Chairs Investment (`) Space Profit (`) Valid
1 22 0 11000/11000 22/40 1100 No
2 0 40 8000/11000 40/40 1000 No
3 10 30 11000/11000 40/40 1250 Yes
How do we choose the best scenario among the several ones that might come up? The
answer to this is Linear Programming.

Linear Programming (LP)

Linear Programming is a methodology to solve a given problem where a given quantity
needs to be maximised or minimised, under the effect of several conditions, using a set of
linear equations or inequations.

Terms Associated with Linear Programming

➢ Objective Function
The function that represents the quantity to be optimised (maximised or minimised) is
called the Objective function. It is denoted by Z. Some examples of objective functions
can be profit function, cost function, time function, interest function, etc.
In the above example, the profit function is the objective function and needs to be
maximised. Thus, Z = 50 x + 25y

➢ Decision Variables
The variables that form the objective function are called decision variables as they
decide the optimal value of the objective function. These are usually represented by x
and y.
In the above example, the number of tables and chairs are the decision variables and
can be denoted by x and y.
➢ Constraints
The conditions that govern and control the values of the decision variables are called
constraints. These are usually in form of equations and inequations of  or  type.

1 9811442618, 9811442628
XII\Maths\Notes\LPP Amit Gupta
In our current context, we will only deal with constraints involving linear
inequalities.
In the above example, the conditions of investment and space for items will form the
constraints of the problem. Thus, we have
500 x + 200y  11000 (investment constraint)
x + y  40 (space constraint)

➢ Non-negativity Constraints
In real life, there is no concept of negative values. Thus, in questions related to linear
programming, no variable can have a negative value. These constraints are called
non-negativity constraints and are usually denoted by x  0, y  0 .

➢ Linear Programming Problem Model (LPP Model)

A mathematical model of the given linear programming problem, clearly stating the
objective function and the related constraints is called the Linear Programming
Problem (LPP) Model.
For the above example, the LPP model is
Maximise Z = 50 x + 25y
subject to constraints
500 x + 200y  11000
x + y  40
x  0, y  0

Solution of a Linear Programming Problem

A linear programming problem can be solved algebraically or graphically. We will study
the graphical solution of a linear programming problem. Some terms related to the
graphical solution of a linear programming problem are:
➢ Feasible Region
The region that is the common solution set of the inequations representing all the
constraints in the given linear programming problem is called the Feasible region.
Every point inside the feasible region satisfies all the constraints and can be a solution
of the given problem.
The feasible region of a linear programming problem may be a bounded region or an
unbounded one.
➢ Corner Points
A feasible region, whether bounded or unbounded, always has one or more corner
points that represent the vertices of the polygonal region representing the feasible
region.

2 9811442618, 9811442628
XII\Maths\Notes\LPP Amit Gupta
➢ Feasible Solution
The set of values of the decision variables that satisfy all the constraints of a given
linear programming problem are called its feasible solutions.
➢ Optimal Solution
The set of values of the decision variables that optimises (maximises or minimises) the
objective function is called its Optimal solution.
➢ According to the Corner Point method of graphical solution of a LPP, the optimal
solution of a linear programming problem lies at one of the corner points representing
its feasible region.

Steps to Solve a LPP Graphically

1. Draw the graph of the equation representing every constraint in the LPP model,
including the non-negativity constraints.
2. Obtain the common solution set of all the constraints.
3. Determine the coordinates of the corner points of the polygonal feasible region so
obtained.
4. Evaluate the value of the objective function Z at all the corner points and obtain the
optimal solution of the problem.
Example 1: Solve the following linear programming problem graphically:
Maximise Z = 4 x + y
subject to the constraints:
x + y  50
3 x + y  90
x  0, y  0

Solution: The various equations corresponding to the constraints are

x + y = 50 line through (50, 0), (0, 50)
3x + y = 90 line through (30, 0), (0, 90)
x=0 y-axis
y=0 x-axis

The feasible region for the above LPP model is given in the graph.

3 9811442618, 9811442628
XII\Maths\Notes\LPP Amit Gupta

The corner points of the feasible region are O(0, 0), A(30, 0), B(20, 30) and C (0, 50)
Vertex Value of Z = 4 x + y Result
O(0, 0) Z = 4(0) + 0 0
A(30, 0) Z = 4(30) + 0 120 (maximum)
B(20, 30) Z = 4(20) + 30 110
C (0, 50) Z = 4(0) + 50 50
Thus, the maximum value of Z is 120 at x = 30, y = 0

Example 2: Solve the following linear programming problem graphically:

Minimise Z = 200 x + 500y
subject to the constraints:
x + 2y  10
3 x + 4 y  24
x  0, y  0

Solution: The various equations corresponding to the constraints are

x + 2y = 10 line through (10, 0), (0, 5)
3x + 4y = 24 line through (8, 0), (0, 6)
x=0 y-axis
y=0 x-axis

The feasible region for the above LPP model is given in the graph.
4 9811442618, 9811442628
XII\Maths\Notes\LPP Amit Gupta

The corner points of the feasible region are A(0, 5), B(4, 3) and C (0, 6)
Vertex Value of Z = 200 x + 500y Result
A(0, 5) Z = 200(0) + 500(5) 2500
B(4, 3) Z = 200(4) + 500(3) 2300 (minimum)
C (0, 6) Z = 200(0) + 500(6) 3000
Thus, the minimum value of Z is 2300 at x = 4, y = 3

Example 3: Solve the following problem graphically:

Minimise and Maximise Z = 3x + 9y
subject to the constraints:
x + 3y  60
x + y  10
x  0, y  0

Solution: The various equations corresponding to the constraints are

x + 3y = 60 (60, 0), (0, 20)
x + y = 10 (10, 0), (0, 10)
x=0 y-axis
y=0 x-axis

The feasible region for the above LPP model is given in the graph.

5 9811442618, 9811442628
XII\Maths\Notes\LPP Amit Gupta

The corner points of the feasible region are A(10, 0), B(60, 0), C (0, 20) and D(0, 10)
Vertex Value of Z = 3x + 9y Result
A(10, 0) Z = 3(10) + 9(0) 30 (minimum)
B(60, 0) Z = 3(60) + 9(0) 180 (maximum)
C (0, 20) Z = 3(0) + 9(20) 180 (maximum)
D(0, 10) Z = 3(0) + 9(10) 90
Thus, the maximum value of Z is 180 at x = 60, y = 0 and x = 0, y = 20 , while the minimum
value of Z is 30 at x = 10, y = 0

Example 4: Minimise Z = 3x + 2y
subject to the constraints:
x+y8
3 x + 5y  15
x  0, y  0
Solution: The various equations corresponding to the constraints are
x+y=8 line through (8, 0), (0, 8)
3 x + 5y = 15 line through (5, 0), (0, 3)
x=0 y-axis
y=0 x-axis

The feasible region for the above LPP model is given in the graph.

6 9811442618, 9811442628
XII\Maths\Notes\LPP Amit Gupta

∵ there is no feasible region (common solution set) for the LPP. Thus, the given LPP has
no optimal solution.
Example 5: Determine graphically the minimum value of the objective function
Z = − 50 x + 20y
subject to the constraints:
2x − y  − 5
3x + y  3
2 x − 3y  12
x  0, y  0

Solution: The various equations corresponding to the constraints are

−5
2x − y = − 5 ( , 0), (0, 5)
2
3x + y = 3 (1, 0), (0, 3)
2 x − 3y = 12 (6, 0), (0, − 4)
x=0 y-axis
y=0 x-axis

The feasible region for the above LPP model is given in the graph.

7 9811442618, 9811442628
XII\Maths\Notes\LPP Amit Gupta

The corner points of the feasible region are A(0, 5), B(0, 3), C (1, 0) and D(6, 0)
Vertex Value of Z = − 50 x + 20y Result
A(0, 5) Z = − 50(0) + 20(5) 100
B(0, 3) Z = − 50(0) + 20(3) 60
C (1, 0) Z = − 50(1) + 20(0) –50
D(6, 0) Z = − 50(6) + 20(0) –300 (minimum)
Thus, the minimum value of Z is –300 at x = 6, y = 0 .
But, there are other points in the feasible region, such as (10, 4), where the value of
Z = − 50(10) + 20(4) = − 420 is lower than the minimum value obtained at one of the
corner points of the feasible region.
Thus, the given LPP has no minimum value and has an unbounded solution.
Example 6: Maximize Z = 8 x + 2y (Solve yourself)
subject to the constraints:
4 x + 5y  35; x + 5y  20; x  0; y  0

8 9811442618, 9811442628
XII\Maths\Notes\LPP Amit Gupta
Example 7: Find the linear inequalities for which the shaded region in the given figure
is the solution set.
Solution: If we consider the equation
−7x + 4y = 14 , we find that (0, 0) is a solution
of the corresponding inequation. Since, on
putting (0, 0) in the LHS of the equation, we
get 0 which is less than 14, we get the
following inequation as the constraint for the
equation
−7x + 4y  14

If we consider the equation 2x + 3y = 3 , we

find that (0, 0) is not a solution of the
corresponding inequation. Since, on putting
(0, 0) in the LHS of the equation, we get 0
which is less than 3, we get the following
inequation as the constraint for the equation
2x + 3y  3

Using similar logic, the other two constraints are

3x + 4y  18 and x − 6y  3

Also, as the entire feasible region lies in the first quadrant only, we have
x  0, y  0
Thus, the various linear inequalities representing the given feasible region are
−7x + 4y  14 , 2x + 3y  3 , 3x + 4y  18 , x − 6y  3 , x  0, y  0

Example 8: The corner points of the feasible region determined by the following system
of linear inequalities:
2x + y  10, x + 3y  15, x , y  0
are (0, 0), (5, 0), (3, 4) and (0, 5). Let Z = px + qy , where p, q > 0. Write the relation
between p and q so that the maximum of Z occurs at both (3, 4) and (0, 5).
Solution: The value of Z at (3, 4) is Z = 3 p + 4q
The value of Z at (0, 5) is Z = 5q

∵ the value of Z is same at both points

 3p + 4q = 5q  q = 3p
which is the required relation.

9 9811442618, 9811442628