Linear programing

1. INTRODUCTION :

In any firm, the manufacturer is always interested in using his resources so that the cost of production is
minimised and the profit is maximised. The methods of linear programming help us in making the best
possible use of limited resources to meet the desired results. The programming is termed linear, because we
shall be considering only linear inequations in the variables of the problem under consideration. The resources
may be in the form of men, material, machines etc.

Linear programming is defined as that branch of mathematics which deals with the optimisation
(maximisation or minimisation) of a linear function of a number of variables subject to a number of conditions
on the variables, in the form of linear inequations in the variables of the optimisation funnciton.

DEFINITIONS :

(i) Objective fucntion : A function of certain variables, which is to be maximised or minimised, subject to
given conditions on the variables of the function, is called the objective function of the problem, under
consideration. In economic applications, if an objective function is a profit function then it is to be maximised
and if an objective function is a cost function,a s s es
C l then it is to be minimised.
(ii) Constraints : The conditions on the variables of an objective function are called the contraints of the
TA

problem, under consideration. The constraints x  0, y  0, are called non-negativity constraints.

GU P

(iii) Feasible region : The region which is common to all constraints of a linear programming problem
(L.P.P.) is callled the feasible region of the given L.P.P.
(iv) Feasible solution : Every point in the feasible region of linear programming problem (L.P.P.) is called a
feasible solution of the given L.P.P.
(v) Optimal feasible solution : A point in the feasible region of a linear programming problem is called an
optimal feasible solution if the objective function of the given L.P.P. is maximised or minimised at that point.

GRAPHICAL METHOD OF SOLVING LINEAR PROGRAMMING PROBLEM :

We shall use graphical method for solving linear programming problems. This method is also known as
corner point method.
Let x and y be two non-negative variables and Z = ax + by be the function of x and y which is to be optimised,
subject to finitely many constraints.
a1x + b1y  (or ) c1, a2x + by2  (or ) c2,.............and x  0, y  0.
In this method the half-plane of each linear constraint in the given L.P. problem is drawn. The intersection of
these half-planes gives the set of all feasible solutions of the L.P. problem.

STEPS FOR SOLVING LINEAR PROGRAMMING PROBLEM :

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. (i) When the feasible region is bounded, M and m are the maximum and minimum values of Z.
(ii) In case, the feasible region is unbounded, we have :
(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.
(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.
Example # 1 : Solve the problem
Maximise z = 6x + 10y
Subject to 3x + 5y  13
5x + 3y  15
and x, y  0
Ans. Max. = 26
Solution : At A (3, 0) ; z = 6 ×3 + 10(0) = 18
9 5
At B (9/4, 5/4) ; z=6× +10 = 26
4 4

 13 
At C (0, 13/ 5) ; z = 6(0) + 10   = 26
 5 

es
a ss
Cl
TA
GU P

9 5  13 
Here we find that z is maximum at B  ,  as well as at C  0,  and its maximum value is 26.
4 4  5 
Every point on the line segment BC gives this maximum value because in this case the line of
objective function lies along one boundary line of the feasible region. In such a case the problem has
infinite number of optimal solutions.

Example # 2 : Minimise Z = 3x1 + 5x2 subject to the constraints :

x1 + 3x2  3, x1 + x2  2, x1, x2  0.
Ans. 7
Solution : The objective function is Z = 3x1 + 5x2.
The constraints are : x1 + 3x2  3 ...(1) x1 + x2  2 ...(2)
x1  0 ...(3) x2  0 ....(4)
The line corresponding to (1) is x1 + 3x2 = 3
(0, 1) and (3, 0) lie on (5). The origin does not lie on (5) and it lies in the half-plane of (1)
if 0 + 3(0)  3, which is not true.
 The closed half-plane not containing the origin is the graph of (1).
The line corresponding to (2) is x 1 + x2 = 2
(0, 1) and (2, 0) lie on (6). The origin does not lie on (6) and it lies in the half-plane of (2) if 0 + 0  2,
which is not true.
 The closed half-plane not containing the origin is the graph of (2).
 x2  0, x2  0 respresent the closed half-planes on the right of x2-axis and above x1-axis respectively.
The shaded unbounded region is the feasible region of the given L.P.P. The vertices of the unbounded

feasible region are A(3, 0), P 3 , 1  and B(0, 2).

2 2

At A(3, 0), Z = 3(3) + 5(0) = 9

3 1 3  1
At P  ,  , Z = 3   + 5   = l7a s
s es
2 2 2 2 C
TA

At B(0, 2), Z = 3(0) + 5(2) = 10

GU P

The minimum of 9, 7, 10 is 7. We draw the graph of the inequality 3x1 + 5x2 < 7. The corresponding

 7 7 
equation is 3x1 + 5x2 = 7 and this passes through L  0,  and M  , 0  . The open half-plane of
 5 3  Logarit
3x1 + 5x2 < 7 is shown in the figure. This half-plane has no point in common with the feasible region.
3 1
 Minimum value of Z = 7 and occurs when x1 = and x2 = .
2 2
Different Types of Linear Programming Problems

A few linear programming problems are listed below :

Example
1. Manufacturing problems :
Solution
In these problems, we determine the number of units of different products which should be produced and sold
by a firm when each product requires a fixed manpower, machine hours, labour hour per unit product,
warehouse space per unit of the output etc. in order to make maximum profit.

Example # 3 : A producer has 30 and 17 units of labour and capital respectively which he can use to produce two
types of goods X and Y. To produce one unit of X, 2 units of labour and 3 units of capital are required.
similarly, 3 units of labour and 1 unit of capital is required to produce one unit of Y. If X and Y are
priced at Rs. 100 and Rs. 120 per unit respectively, how should the producer use his resources to
maximise the total revenue ? Solve the problem, graphically.
Solution : Let x be the no. of units of X and y be the number of units of Y produced. We can arrange the
information in the problem as :
X Y Available Units
Labour 2 3 30
Capital 3 1 17
Revenue per unit Rs. 100 Rs. 120
 We can formulate this problem as
 Maximize : z = 100 x + 120 y
subject to the constraints
2x + 3y  30
3x + y  17
x0;y0
We first plot the lines
2x + 3y = 30
3x + y = 17
These lines meet at P(3, 8). Feasible region has
been shaded. It extreme points are O. A, P and D.
we now calculate the value of the objective function :
z = 100x + 120y
at these points.

Extreme Po int Value of Z

O(0, 0 ) 0

 17  1700
A  , 0
 3  3
P(3, 8) 1260
es
a ss
D(0, 10 ) 1200 Cl
TA
GU P

 Z has maximum value at P(3, 8) and this is equal to Rs. 1260.

Example # 4 : A manufacturer produces two types of steel trunks. He has two machines A and B. The first type of
trunk requires 3 hours on machine A and 3 hours on machine B. The second type requires 3 hours
on machine A and 2 hours on machine B. Machines A and B can work at most for 18 hours and 15
hours per day respectively. He earns a profit of Rs. 30 and Rs. 25 per trunk of the first type and
second type respectively. How many trunks of each type must he make each day to make maximum
profit ?
Solution : Let x and y be the number of two types of steel trunks to be manufactured
Time(in hr.) Time(in hr.) Available Time
Machine
Trunk  I ( x ) Trunk  II ( y ) (in hr.)
A 3 3 18
B 3 2 15
Pr ofit per unit Rs. 30 Rs. 25
Then Maximize : z = 30 x + 25 y
subject to the constraints
3x + 3y  18
3x + 2y  15
x  0, y  0
Now draw the lines
3x + 3y = 18, 3x + 2y = 15
Their point of intersection is B(3, 3)
The feasible region, OABC is shaded in the adjoining diagram.
Now valueof z = 30x + 25y
at A(5, 0) is 30 × 5 + 25 × 0 = 150
at B(3, 3) is 30 × 3 + 25 × 3 = 165
at C(0, 6) is 30 × 0 + 25 × 6 = 150
 For getting maximum profit of Rs. 165,
3 trunks of each type should be manufactured.
2. Diet Problems :
In these problems, we determine the amount of different kinds of constituents/nutrients which should be
included in a diet so as to minimise the cost of the desired diet such that it contains a certain minimum
amount of each constituent/nutrients.

Example # 5 : A housewife wishes to mix up two kinds of foods X and Y in such a way that mixture contains at
least 10 units of vitamin A, 12 units of vitamin B and 8 units of vitamin C. The vitamin contents of 1
kg of food X and 1 kg of food Y are as given in the following table :

Food Vitamin A Vitamin B Vitamin C

X 1 2 3
Y 2 2 1
If one kg of food X costs Rs. 6 and one kg of food Y costs Rs. 10, find the least cost of the mixture
which will produce the desired diet.
Solution : Let x kg of food X and y kg of Food Y be mixed to get the desired diet. Then
the L.P.P. is
Minimise : z = 6x + 10y
Subject to the constraints
(Vitamin A) x + 2y  10
(Vitamin B) 2x + 2y  12
(Vitamin C) 3x + y  8
x0,y0
es
Now draw the lines
a ss
AB : x + 2y = 10 Cl
TA

CD : 2x + 2y = 12
GU P

EF : 3x + y = 8
AB and CD meet at P(2, 4)
CD and EF meet at Q(1, 5).
The shaded region APQF, is the feasible region.
The value of z = 6x + 10y
at A is z = 6 × 10 + 10 × 0 = 60
at P is z = 6 × 2 + 10 × 4 = 52
at Q is z = 6 × 1 + 10 × 5 = 56
at F is z = 6 × 0 + 10 × 8 = 80
The smallest value of Z is 52 at the point P(2, 4), but as the feasible region is unbounded therefore
we have to draw the graph of inequality 6x + 10y < 52
Now from the figure it is clear that 6x + 10y < 52 has no point common with the feasible region.
Thus the minimum value of Z is 52 i.e., when 2 kg of Food X is mixed with 4 kg of Food Y.
3. Transportation problems :
In these problems, we determine a transportation schedule in order to find the cheapest way of
transporting a product from plants/factories situated at different locations to different markets.
Example # 6 : (Transportation Problem)
There are two factories located one at place P and the other at place Q. From these locations, a
certain commodity is to be delivered to each of the three depots situted at A, B and C. The weekly
requirements of the depots are respectively 5, 5 and 4 units of the commodity while the production
capacity of the facories at P and Q are respectively 8 and 6 units. The cost of transportation per unit
is given below :

COST (in Rs)

FROM / TO
A B C
P 160 100 150
Q 100 120 100
How many units should be transported from each factory to each depot in order that the
transportation cost is minimum. What will be the minimum transportation cost?
Solution : The problem can be explained diagrammatically as follows
Let x units and y units of the commodity be transported from the factory at P to the depots at A and
B respectively. Then (8 – x – y) units will be transported to depot at C(why ?)

Factory

P
8units
Rs
x 15
0
0 y Rs 100
s 16
R
Depot A B Depot C Depot
5units 5units 4units
Rs
10 5 –y Rs 120 0 y)]
5– 0
s 10 (5–
x R )+
5 –x
Q [(
6units 6–

Hence we have x  0, y  0 and 8 – x – y  0

i.e. x  0, y  0 and x + y  8
Now, the weekly requirement of the depot at A is 5 units of the commodity. Since x units are
transported from the factory at P, the l ss es (5 – x) units need to be transported from the factory
aremaining
C
at Q. obviously,
TA

5 – x  0, i.e. x  5.
GU P

Similarly (5 – y) and 6 – (5 – x + 5 – y) = x + y – 4 units are to be transported from the factory at Q

to the depots at B and C respectively.
THus, 5 – y  0, x + y – 4  0
i.e. y  5, x + y  4
Therefore, the problem reduces to Minimise Z = 10(x – 7y + 190)
subject to constraints :
x  0, y  0 ...(i)
x+y8 ...(ii)
x5 ...(iii)
y5 ...(iv) and
x+y4 ...(v)

Y x=5

x+y=4 6
B(0,5) C(3,5)
y=5

4
A(0, 4) D(5, 3)
2
(5, 0)
X  X
O 2 4 6 8 10
F(4, 0) x+y=8

Y 
 The shaded region ABCDEF represented by the constraints (i) to (v) is the feasible region. Observe
that the feasible region is bounded. The coordinates of the corner points of the feasible region are
(0, 4), (0, 5), (3, 5), (5, 3), (5, 0) and (4, 0).
Let us evaluate Z at these points

Corner Po int Z  10( x – 7 y  190 )

(0, 4) 1620
(0, 5 ) 1550
(3, 5 ) 1580
(5, 3 ) 1740
(5, 0 ) 1950
(5, 0 ) 1940

From the table, we see that the minimum value of Z is 1550 at the point (0, 5).
Hence, the optimal transportation strategy will be to deliver 0, 5 and 3 units from the factory at P and
5, 0 and 1 units from the factory at Q to the depots at A, B and C respectively. Corresponding to this
strategy, the transportation cost would be minimum, i.e. Rs 1550.
4. Allocation Problem :

Example 7. A cooperative society of farmers has 50 hectare of land to grow two crops X and Y. The profit from
cropsX and Y per hectare are estimated as Rs 10,500 and Rs 9,000 respectively. To control weeds,
a liquid herbicide has to be used for crops X and Y at rates of 20 liters and 10 liters per hectare.
es
a ss should be used in order to protect fish and wild life using
Cl
Further, no more than 800 liters of herbicide
a pond which collects drainage from this land. How much land should be allocated to each crop so
TA

as to maximise the total profit of the society ?

GU P

Solution : Let x hectare of land be allocated to crop X and y hectare to crop Y. Obviously x  0, y  0.
Profit per hectare on crop X = Rs 10500
Profit per hectare on crop Y = Rs 9000
Therefore, total profit =Rs (10500x + 9000y)
The mathematical formulation of the problem is as follows :
Maximise Z = 10500x + 9000y
subject to the constraints :
x + y  50 (constraint related to land) ....(1)
20 x + 10 y  800 (constraint related to use of herbicide)
i.e. 2x + y  80 ....(2)
x  0, y  0 (non negative constraint) .....(3)
Let us draw the graph of the system of inequalities (1) to (3). The feasible region OABC is shown
(shaded) in the figure. Observe that the feasible region is bounded.
The coordinates of the corner points O, A, B and C are (0, 0), (40, 0), (30, 20) and (0, 50) respectively.
Let us evaluate the objective function Z = 10500x + 9000y at these vertices to find which one gives
the maximum profit.

Corner Po int Z  10500 x  9000 y

O(0,0 ) 0
A( 40,0) 420000
B(30,20 ) 495000
C(0,50 ) 450000

Hence, the society will get the maximum profit of Rs 4,95,000 by allocating 30 hectares for crop X
and 20 hectares for crop Y.