Linear Programming

Introduction to Linear Programming Problems

© 2021, BYJU'S. All rights reserved

 Road Map

Illustration on Un-Bounded feasible solution

Illustration on Bounded feasible solution

Linear Programming Problem

© 2021, BYJU'S. All rights reserved

 Linear Programming Problem
Definition :
It is a process of finding the optimal value (maximum or minimum) of a linear
function (objective function) of some variables, subject to linear constraints
(equalities / inequalities).

Linear programming is a technique in which a linear function is maximized or

minimized subject to various constraints.

This technique is useful for quantitative decisions in business planning , industrial

engineering etc.

© 2021, BYJU'S. All rights reserved

 Method to solving Linear Programming Problem

Method of solving linear programming problem is referred as Corner Point

Method. The method comprises of the following steps:

1. Find the feasible region of the linear programming problem and determine its
corner points (vertices) either by inspection or by solving the two equations of
the lines intersecting at that point.

2. Evaluate the objective function 𝑍 = 𝑎𝑥 + 𝑏𝑦 at each corner point. Let 𝑀 and 𝑚 ,

respectively denote the largest and smallest values of these points.

3. When the feasible region is bounded , 𝑀 and 𝑚 are the maximum and
minimum values of 𝑍.

© 2021, BYJU'S. All rights reserved

 4. In case, the feasible region is unbounded ,

4. (𝑎) 𝑀 is the maximum value of 𝑍, if the open half plane determined by

𝑎𝑥 + 𝑏𝑦 > 𝑀 has no point in common with the feasible region.
Otherwise, 𝑍 has no maximum value

4. (𝑏) Similarly, 𝑚 is the minimum value of 𝑍, if the open half plane determined by
𝑎𝑥 + 𝑏𝑦 < 𝑀 has no point in common with the feasible region.
Otherwise, 𝑍 has no minimum value.

Let’s try to understand these points using some illustrations

© 2021, BYJU'S. All rights reserved

 Question:

Solve the linear programming problem graphically :

Maximize 𝑍 = 4𝑥 + 𝑦
Subject to constraints : 𝑥 ≥ 0 , 𝑦 ≥ 0
𝑥 + 𝑦 ≤ 50
3𝑥 + 𝑦 ≤ 90
Solution:
Maximize 𝑍 = 4𝑥 + 𝑦 ; where 𝑍 is the Objective function

Subject to constraints : 𝑥 ≥ 0 , 𝑦 ≥ 0
𝑥 + 𝑦 ≤ 50
3𝑥 + 𝑦 ≤ 90
© 2021, BYJU'S. All rights reserved
 𝑥 + 𝑦 = 50 is the line passing through the (50 , 0) and (0 , 50)
𝑌
Solution set for 𝑥 ≥ 0 , 𝑦 ≥ 0 is first quadrant 100
3𝑥 + 𝑦 = 90
including co-ordinate axis
80

(0 , 0) satisfy 𝑥 + 𝑦 ≤ 50 so solution set is the region

60
below the line 𝑥 + 𝑦 = 50 including 𝑥 + 𝑦 = 50 𝐶(0,50)
40
3𝑥 + 𝑦 = 90 is the line passing through the 𝐵(20,30)
(30 , 0) and (0, 90) 20
𝑥 + 𝑦 = 50
𝑋
(0 , 0) satisfy 3𝑥 + 𝑦 ≤ 90 so solution set is the region 𝑂 0,0 20 40 60 80 100
𝐴(30,0)
below the line 3𝑥 + 𝑦 = 90 including 3𝑥 + 𝑦 = 90

Point of intersection of 3𝑥 + 𝑦 = 90 & 𝑥 + 𝑦 = 50 is (20 , 30)

Solution set of all the constraints is shown in figure:
© 2021, BYJU'S. All rights reserved
 𝑌
100
Shaded region is known as Feasible region 3𝑥 + 𝑦 = 90
80
Points inside feasible region is Feasible solution
60
𝐶(0,50)
Points outside feasible region is Infeasible solution 40
𝐵(20,30)
20

Points in the feasible region that gives the 𝑥 + 𝑦 = 50

𝑂 0,0
𝑋
optimal value of 𝑍 : Optimal (feasible) solution 0 20 40 60 80 100
𝐴(30,0)

© 2021, BYJU'S. All rights reserved

 Here feasible region is bounded so optimal values
occur at its corner points : 𝑌
100
3𝑥 + 𝑦 = 90
From graph, corner points are : 80
𝑥 + 𝑦 = 50
0,0 , 30,0 , 20,30 , (0,50)
60
𝐶(0,50)
Vertex of feasible region Corresponding value of 𝑍 40

𝑂 0,0 0 𝐵(20,30)
20
𝐴 30,0 120 (Max)
𝑂 0,0
𝐵(20,30) 110 𝑋
0 20 40 60 80 100
𝐴(30,0)
𝐶(0,50) 50

Objective function 𝑍 is maximum at 𝐴 30,0

Hence, optimal solution is 𝑥 = 30 , 𝑦 = 0
© 2021, BYJU'S. All rights reserved
 Question:

Solve the linear programming problem graphically :

Minimize 𝑍 = 200𝑥 + 500𝑦
Subject to constraints : 𝑥 ≥ 0 , 𝑦 ≥ 0
𝑥 + 2𝑦 ≥ 10
3𝑥 + 4𝑦 ≤ 24

Solution:
Minimize 𝑍 = 200𝑥 + 500𝑦; where 𝑍 is the Objective function
Subject to constraints : 𝑥 ≥ 0 , 𝑦 ≥ 0
𝑥 + 2𝑦 ≥ 10
3𝑥 + 4𝑦 ≤ 24
© 2021, BYJU'S. All rights reserved
 𝑥 + 2𝑦 = 10 is the line passing through (10 , 0) and (0, 5)
𝑌
Solution set for 𝑥 ≥ 0 , 𝑦 ≥ 0 is first quadrant
including co-ordinate axis 8

(0 , 0) does not satisfy 𝑥 + 2𝑦 ≥ 10 so solution 6 𝐶(0,6)

set is the region above the line 𝑥 + 2𝑦 = 10 𝐴(0,5) 3𝑥 + 4𝑦 = 24

4
including 𝑥 + 2𝑦 = 10 𝐵(4,3)
2
𝑥 + 2𝑦 = 10
3𝑥 + 4𝑦 ≤ 24 is the line passing through
(8 , 0) and (0, 6) 0 2 4 6 8 10
𝑋

(0 , 0) satisfy 3𝑥 + 4𝑦 ≤ 24 so solution set is the region

below the line 3𝑥 + 4𝑦 = 24 including 3𝑥 + 4𝑦 = 24
Point of intersection of 𝑥 + 2𝑦 = 10 & 3𝑥 + 4𝑦 = 24 is (4 , 3)
Solution set of all the constraints is shown in figure:
© 2021, BYJU'S. All rights reserved
 Here feasible region is bounded so optimal
values occur at its corner points : 𝑌
8
From graph, corner points are :
0,5 , 4,3 , 0,6 6 𝐶(0,6)
𝐴(0,5) 3𝑥 + 4𝑦 = 24
4
Vertex of feasible region Corresponding value of 𝑍 𝐵(4,3)
2
𝐴 0,5 2500 𝑥 + 2𝑦 = 10
𝐵 4,3 2300 (Min) 𝑋
0 2 4 6 8 10
𝐶(0,6) 3000

Objective function 𝑍 is minimum at 𝐴 4,3 .

Hence, optimal solution is 𝑥 = 4 , 𝑦 = 3

© 2021, BYJU'S. All rights reserved

 Question:

Solve the linear programming problem graphically :

Minimize 𝑍 = 50𝑥 + 70𝑦
Subject to constraints : 𝑥 ≥ 0 , 𝑦 ≥ 0
2𝑥 + 𝑦 ≥ 8
𝑥 + 2𝑦 ≥ 10

Solution:
Minimize 𝑍 = 50𝑥 + 70𝑦 ; where 𝑍 is the Objective function
Subject to constraints : 𝑥 ≥ 0 , 𝑦 ≥ 0
2𝑥 + 𝑦 ≥ 8
𝑥 + 2𝑦 ≥ 10
© 2021, BYJU'S. All rights reserved
 2𝑥 + 𝑦 = 8 is the line passing through (4 , 0) and (0, 8)
Solution set for 𝑥 ≥ 0 , 𝑦 ≥ 0 is first quadrant
including co-ordinate axis
𝑌
(0 , 0) does not satisfy 2𝑥 + 𝑦 ≥ 8 so solution 8 𝐶(0,8)
set is the region above the line 2𝑥 + 𝑦 = 8 6 2𝑥 + 𝑦 = 8
including 2𝑥 + 𝑦 = 8
4 𝐵(2,4)
𝑥 + 2𝑦 ≥ 10 is the line passing through
(10 , 0) and (0, 5) 2 𝑥 + 2𝑦 = 10
𝐴(10,0)
(0 , 0) does not satisfy 𝑥 + 2𝑦 ≥ 10 so solution 0 2 4 6 8 10
𝑋
set is the region above the line 𝑥 + 2𝑦 = 10
including 𝑥 + 2𝑦 = 10
Point of intersection of 2𝑥 + 𝑦 = 8 & 𝑥 + 2𝑦 = 10 is (2 , 4)
Solution set of all the constraints is shown in figure:
© 2021, BYJU'S. All rights reserved
 Here feasible region is unbounded. 𝑌
8 𝐶(0,8)
From graph , corner points are : 6 2𝑥 + 𝑦 = 8
10 , 0 , 2,4 , 0,8
4 𝐵(2,4)

Corner point Corresponding value of 𝑍 2 𝑥 + 2𝑦 = 10

𝐴 10,0 500 𝐴(10,0)
𝑋
𝐵 2,4 380 0 2 4 6 8 10

𝐶(0,8) 560

Since, the region is unbounded, it’s not necessary that minimum value of 𝑍 is 380.
Now, we will consider, the region 50𝑥 + 70𝑦 < 380

© 2021, BYJU'S. All rights reserved

 0 , 0 satisfy 50𝑥 + 70𝑦 < 380 so solution
set is towards the origin excluding
50𝑥 + 70𝑦 = 380 𝑌
8 𝐶(0,8)
Since no point is common
6 2𝑥 + 𝑦 = 8
So , no such value of (𝑥 , 𝑦) exist for
4 𝐵(2,4)
which value of objective function 𝑍 is
less than 380 2 𝑥 + 2𝑦 = 10
𝐴(10,0)
𝑋
At 𝐵(2 , 4) objective function 𝑍 = 380 0 2 4 6 8 10
5𝑥 + 7𝑦 = 38

So, minimum value of 𝑍 = 380

Hence , optimal solution is 𝑥 = 2 , 𝑦 = 4

© 2021, BYJU'S. All rights reserved
 Question:

Solve the linear programming problem graphically :

Maximize 𝑍 = 3𝑥 + 9𝑦
Subject to constraints : 𝑥 ≥ 0 , 𝑦 ≥ 0
𝑥≤𝑦
𝑥 + 𝑦 ≥ 10
𝑥 + 3𝑦 ≤ 60
Solution:
Maximize 𝑍 = 3𝑥 + 9𝑦 ; where 𝑍 is the Objective function
Subject to constraints : 𝑥 ≥ 0 , 𝑦 ≥ 0
𝑥≤𝑦
𝑥 + 𝑦 ≥ 10
𝑥 + 3𝑦 ≤ 60
© 2021, BYJU'S. All rights reserved
 𝑌
𝑥 + 𝑦 = 10 is the line passing through 100

(10 , 0) and (0, 10) 𝑥=𝑦

80

𝑥 + 𝑦 = 10
60
Solution set for 𝑥 ≥ 0 , 𝑦 ≥ 0 is first
quadrant including co-ordinate axis 40

𝐶(0,20)
20
𝐷(15,15)
(1 , 0) does not satisfy 𝑥 ≤ 𝑦 so solution set is 𝐵(0,10) 𝐴(5,5)
𝑋
the region above the line 𝑥 = 𝑦 including 𝑥 = 𝑦 0 20 40 60 80 100
𝑥 + 3𝑦 = 60

(0 , 0) does not satisfy 𝑥 + 𝑦 ≥ 10 so solution

set is the region above the line 𝑥 + 𝑦 = 10
including 𝑥 + 𝑦 = 10

© 2021, BYJU'S. All rights reserved

 𝑥 + 3𝑦 = 60 is the line passing through (60 , 0) 𝑌
and (0, 20) 100

𝑥=𝑦
80

𝑥 + 𝑦 = 10
(0 , 0) satisfy 𝑥 + 3𝑦 ≤ 60 so solution 60
set is the region below the line
40
𝑥 + 3𝑦 = 60 including 𝑥 + 3𝑦 = 60
𝐶(0,20)
20
𝐷(15,15)
𝐵(0,10) 𝐴(5,5)
Point of intersection of 𝑥 + 3𝑦 = 60 0 20 40 60 80 100
𝑋
& 𝑦 = 𝑥 is (15 , 15) 𝑥 + 3𝑦 = 60

Point of intersection of 𝑥 + 𝑦 = 10 & 𝑦 = 𝑥 is (5 , 5)

© 2021, BYJU'S. All rights reserved

 Here feasible region is bounded so optimal values occur at its corner points :

From graph , corner points are : 𝑌

100
5 , 5 , 0 , 10 , 0 , 20 , (15, 15)
𝑥=𝑦
80

Corner Point Corresponding value of 𝑍

60
𝐴 5,5 60
40
𝐵 0,10 90
𝐶(0,20) 180 (Max) 20 𝐶(0,20)
𝐷(15,15)
𝐷(15,15) 180 (Max) 𝐵(0,10) 𝐴(5,5)
0 𝑋
20 40 60 80 100
Thus, For every point on entire line segment 𝐶𝐷, 𝑥 + 𝑦 = 10 𝑥 + 3𝑦 = 60
𝑍 will have maximum value.
Hence, this linear programming problem has multiple optimal solutions.
© 2021, BYJU'S. All rights reserved
 Question:
A manufacturer has three machines 𝐼, 𝐼𝐼 & 𝐼𝐼𝐼 installed in his factory. Machines
𝐼 & 𝐼𝐼 are capable of being operated for at most 12 hours whereas machine 𝐼𝐼𝐼
is being operated for at least 5 hours a day. He produces only two items 𝑀 & 𝑁
each requiring the use of all three machines. The number of hours required for
producing 1 unit of each of 𝑀 & 𝑁 on the three machines are given in the
table:
Items Number of hours required on the machines
𝐼 𝐼𝐼 𝐼𝐼𝐼
𝑀 1 2 1
𝑁 2 1 1.25
He makes a profit of Rs. 600 and Rs. 400 on items 𝑀 & 𝑁
respectively. How many of each items should he produce so
as to maximize the profit assuming that he can sell all the
items that he produced. What will be the maximum profit ?
© 2021, BYJU'S. All rights reserved
 Solution:
Let 𝑥 & 𝑦 be the number of items of 𝑀 & 𝑁 produced by manufacturer
respectively

There is a profit of Rs. 600 and Rs. 400 on each item of 𝑀 & 𝑁 respectively

Total profit on production = Rs ( 600𝑥 + 400𝑦 )

Maximize 𝑍 = 600𝑥 + 400𝑦 ; where 𝑍 is the Objective function

Items Number of hours required on the machines

𝐼 𝐼𝐼 𝐼𝐼𝐼
𝑀→ 𝑥 1 2 1
𝑁→ 𝑦 2 1 1.25

© 2021, BYJU'S. All rights reserved

 Number of items can not be negative ⇒ 𝑥 ≥ 0 , 𝑦 ≥ 0

Machines 𝐼 & 𝐼𝐼 are capable of being operated for at most 12 hours so from table ,

𝑥 + 2𝑦 ≤ 12 , 2𝑥 + 𝑦 ≤ 12

Machine 𝐼𝐼𝐼 is being operated for at least 5 hours a day so from table ,
5
𝑥+ 𝑦≥5
4
𝑥 ≥ 0 ,𝑦 ≥ 0
𝑥 + 2𝑦 ≤ 12
constraints : 2𝑥 + 𝑦 ≤ 12
5
𝑥+ 𝑦≥5
4

𝑥 + 2𝑦 = 12 is the line passing through the (12 , 0) and (0, 6)

© 2021, BYJU'S. All rights reserved
 Solution set for 𝑥 ≥ 0 , 𝑦 ≥ 0 is first quadrant
including co-ordinate axis
12
2𝑥 + 𝑦 = 12

0 , 0 satisfy 𝑥 + 2𝑦 ≤ 12 so solution set is 10

the region below the line 𝑥 + 2𝑦 = 12 8

5
including 𝑥 + 2𝑦 = 12 𝑥+ 𝑦=5
4 6
𝐵(0,6)

𝐶(4,4)
4
𝐴(0,4)
2𝑥 + 𝑦 ≤ 12 is the line passing through 𝑥 + 2𝑦 = 12
2
the (6 , 0) and (0, 12) 𝐷(6,0)
0 2 4 6 8 10 12
𝑋
𝐸(5,0)
(0 , 0) satisfy 2𝑥 + 𝑦 ≤ 12 so solution set is the
region below the line 2𝑥 + 𝑦 = 12 including
2𝑥 + 𝑦 = 12
© 2021, BYJU'S. All rights reserved
 12
5 2𝑥 + 𝑦 = 12
𝑥+ 𝑦
≥ 5 is the line passing
4 10
through the (5 , 0) and (0, 4)
8
5 𝐵(0,6)
𝑥+ 𝑦=5 6
4
5 𝐶(4,4)
0 , 0 does not satisfy 𝑥 + 𝑦 ≥ 5 so solution 4
𝐴(0,4)
4 𝑥 + 2𝑦 = 12
5
set is the region above the line 𝑥 + 𝑦 =5 2
4 𝐷(6,0)
5 𝑋
including 𝑥 + 𝑦 =5 0 2 4 6 8 10 12
4 𝐸(5,0)

Point of intersection of 𝑥 + 2𝑦 = 12 & 2𝑥 + 𝑦 = 12 is (4 , 4)

© 2021, BYJU'S. All rights reserved

 Here feasible region is bounded so optimal 𝑌
values occur at its corner points : 12
2𝑥 + 𝑦 = 12
10
From graph , corner points are :
0,4 , 0,6 , 4,4 , 6 , 0 , (5 , 0) 8
𝐵(0,6)
6
Vertex of feasible region Corresponding value of 𝑍
𝐶(4,4)
4
𝐴 0,4 1600 𝐴(0,4)
𝑥 + 2𝑦 = 12
2
𝐵 0,6 2400
𝐷(6,0)
𝐶(4,4) 4000 (Max) 0 2 4 6 8 10 12
𝑋
𝐸(5,0)
𝐷(6,0) 3600
5
𝑥+ 𝑦=5
𝐸(5,0) 3000 4

Objective function 𝑍 is maximum at 𝐴 4 , 4

Hence , manufacturer should produce 4 units of each item 𝑀 & 𝑁
© 2021, BYJU'S. All rights reserved
 Summary Sheet
❑ Linear programming is a technique in which a linear function is maximized
or minimized subject to various constraints.

❑ This technique is useful for quantitative decisions in business planning ,

industrial engineering etc.

❑ Shaded region : Feasible region

❑ Points inside feasible region : Feasible solution

❑ Points outside feasible region: Infeasible solution

❑ Points in the feasible region that gives the optimal value of 𝑍 :

© 2021, BYJU'S. All rights reserved
Optimal (feasible) solution