QUANTITATIVE TECHNIQUES IN

MANAGEMENT-
LINEAR PROGRAMMING
Course: Industrial Management and Safety
Course Code: 5001
Module: III

Arun Kumar S L
Lecturer in Mechanical Engineering
Operations Research (OR)
• OR is a problem solving and • Various techniques used in OR to
decision making science. solve optimization problems are:
• It is a kit of scientific and 1. Linear Programming
i. Graphical Method
programmable rules providing
ii. Transportation Method
the management a ‘quantitative a. Vogel’s Approximation Method
basis’ for decisions regarding the b. North West Corner Method
operations under its control. iii. Simplex Method
2. Waiting Line or Queuing Theory
3. Game Theory
4. Dynamic Programming

Arun Kumar S L
Lecturer in Mechanical Engineering
Linear Programming (LP) and its formulation
• Linear programming deals with optimization • Formulation of LP Problem:
(maximization or minimization) of a function • The standard manner of expressing a Linear
of variables known as objective function, Programming:
subject to a set of linear equalities and/or
inequalities known as constraints.
• Maximize (or Minimize),
• LP is a mathematical tool/technique for z = C1x1 + C2x2 + …+ Cnxn  Objective
determining the best uses of an organization’s Function
resources.
Subjected to conditions,

• a11x1+a12x2+a13x3+…+a1nxn ≤ b1
• a21x1+a22x2+a23x3+…+a2nxn ≤ b2

• am1x1+am2x2+am3x3+…+amnxn≤bm Constraint
bi ≥ 0, i = 1,2,3, …..m
Equations
xj ≥ 0, i = 1,2,3, …..n

Arun Kumar S L
Lecturer in Mechanical Engineering
Conditions to be satisfied for using Linear
Programming for optimization problems
• LP can be used for optimization n 3. There must alternative courses of
problems if following conditions action. For example, a given
are satisfied: product may be expressed by two
different machines and problem
1. There must be a well defined may be as to how much of the
objective function which is to be product to allocate to each
either maximized or minimized and machine.
which can be expressed as a linear 4. The decision variables should be
function of decision variables. inter-related and non negative.
2. There must be constraints on the 5. The resources must be in limited
extend of attainment of the supply and the total production
objective and these constraints capacity is limited.
must be capable of being expressed
as linear equalities or inequalities
in terms of variables.

Arun Kumar S L
Lecturer in Mechanical Engineering
Formulation of LP Problem
• A firm produces three products. These products
are processed on three different machines. The Solution:
profit per unit for product 1, 2 and 3 is Rs. 4, Rs 3 Let x1,x2,x3 = Number of product1, product2 and
and Rs 6 respectively. The time required for each of product3 manufactured daily respectively
the three products and the daily capacity of the Let z = profit = 4x1 + 3x2 + 6x3
three machines are given below. Formulate a LPP
to determine the daily number of units to be
Formulation of LPP:
manufactured for each product.
Time per unit (minutes) Machine Maximize z = 4x1 + 3x2 + 6x3  Objective Function
Capacity
Machine Product1 Product2 Product3
(minutes 2x1 + 3x2 + 2x3 ≤ 440
/day) 4x1 + 0x2 + 3x3 ≤ 470 Constraint
M1 2 3 2 440 2x1 + 5x2 + 0x3 ≤ 430 Equations
x 1 , x2 , x 3 ≥ 0
M2 4 - 3 470
M3 2 5 - 430
Arun Kumar S L
Lecturer in Mechanical Engineering
 Total
Formulation of LP Problem Raw
Units of raw material
making one unit of
units of
raw
Materials
- Example product material
available
A B
• A company manufactures and X 2 4 24
sells two products A and B. They Y 3 2 16
require the use of 3 different Z 4 2 22
raw materials X, Y, Z and are
available in limited quantities. Solution:
Let x, y = Number of products of A and B respectively.
The profit per unit of product A Let z = profit = 4x + 3y
and B is 4 and 3 units of money Formulation of LPP:
respectively. Formulate a linear
Maximize z = 4x + 3y  Objective Function
programming model to
maximize the profit. The other 2x + 4y ≤ 24 Constraint Equation
3x + 2y ≤ 16
details are given 4x + 2y ≤ 22
Arun Kumar S L x, y ≥ 0
Lecturer in Mechanical Engineering
 Yield per unit
LPP Formulation – Food Type
Proteins Fats Carb.
Cost per
unit

Example 1
2
3
4
2
2
6
4
45
40
• A person wants to decide the 3 8 7 7 85
constituents of a diet which will 4 6 5 4 65

fulfill his daily requirements of Min. 800 200 700

requirement
proteins, fats and carbohydrates
at minimum cost. The choice is Solution:
to be made from four different Let x1, x2, x3 = Number of units of food types 1, 2 and 3
consumed respectively
types of foods. The yields per Let z = cost of food = 45x1 + 40x2 + 85x3 + 65x4
unit of these food are given in Formulation of LPP:
table: Minimize z = 45x1 + 40x2 + 85x3 + 65x4  Objective Function

3x1 + 4x2 + 8x3 + 6x4 ≥ 800

2x1 + 2x2 + 7x3 + 5x4 ≥ 200 Constraint
6x1 + 4x2 + 7x3 + 4x4 ≥ 700 Equations
Arun Kumar S L x1 , x 2 , x 3 , x 4 ≥ 0
Lecturer in Mechanical Engineering
LPP – Graphical Method – Example(1)
• A furniture manufacturer makes two products, chairs and tables. Formulate
and solve the LP Problem.
Solution:
Let number of chairs produced = x; Let number of tables produced = y
Time per Unit (hours) Let z = profit = 20x + 40y
Processing Section
Step 1: Formulation of Linear Programming Model:
Sections Availability (hours)
Chair Table • Objective Function
• Maximize z = 20x + 40y …………(1)
S1 1 4 24 • Constraint Equations
• 1x + 4y ≤ 24 ………………..(2)
S2 3 1 21 • 3x + 1y ≤ 21 ………………..(3)
• 1x + 1y ≤ 8 ………………..(4)
S3 1 1 8 • x,y ≥ 0 ………………..(5)
Step 2: Convert the constraint inequalities temporarily into equations:
Profit per unit Rs 20 Rs 40 • 1x + 4y = 24 ………………..(6)
• 3x + 1y = 21 ………………..(7)
• 1x + 1y = 8 ………………..(8)
Step 3: Axes are marked on the graph paper and are labelled with x and y

Arun Kumar S L
Lecturer in Mechanical Engineering
 LPP – Graphical Method – Example(1)-
Continuation
Step 4: Draw straight lines on the graph using
constraint equations.
• To find the points to draw the line 1x + 4y = 24,
1) when x = 0, y = 6
2) When y = 0, x = 24
Therefore the points are (0,6) and (24,0).
Draw a straight line between these points.
• Similarly points to draw line 3x + 1y = 21 are
(0,21) and (7,0). Draw a straight line between
these points.
• Similarly points to draw line 1x + 1y = 8 are (0,8)
and (8,0). Draw a straight line between these
points.

Arun Kumar S L
Lecturer in Mechanical Engineering
LPP – Graphical Method – Example(1)-
Continuation
• Step 5: Substitute the corner points of the Feasible Solution Region in the
objective function and identify the corner point that maximizes the
objective function, which is the required solution of the LP problem.

Corner Point Coordinate(x,y) z = 20x+40y

O (0,0) z=0
Therefore, the required solution
A (0,6) z = 240
is x = 2.67 and y = 5.33 (Ans.)
G (2.67,5.33) z= 266.6

H (6.5,1.5) z= 190

D (7,0) z= 140

Arun Kumar S L
Lecturer in Mechanical Engineering
LPP – Graphical Method – Example(2)
• Solve the following LPP graphically: • Solution:
• Minimize z = 2x + 3y • Converting constraint inequalities
• Subjected to: temporarily into equations.
• x+y≥6 • x + y = 6 …………(1)
• 2x + y ≥ 7 • 2x + y = 7…………(2)
• x + 4y ≥ 8 • x + 4y = 8…………(3)
• x, y ≥ 0 • From equation (1)
• When x = 0, y = 6 A(0,6)
• When y = 0, x = 6 B(6,0)
• From equation (2)
• When x = 0, y = 7 C(0,7)
• When y = 0, x = 3.5 D(3.5,0)
• From equation (3)
• When x = 0, y = 2 E(0,2)
• When y = 0, x = 8 F(8,0)

Arun Kumar S L
Lecturer in Mechanical Engineering
 LPP – Graphical Method – Example(2) -
Continuation
y
Corner Point Coordinate(x,y) z = 2x + 3y
C (0,7) 21
G (1,5) 17
H (5.33,0.67) 12.67
F (8,0) 16

The minimum value of z is obtained at point H.

Therefore the required solution is x = 5.33 and
y = 0.67
x

Arun Kumar S L
Lecturer in Mechanical Engineering
LPP – Graphical Method – Example(3)

• Solve the following LPP by • From eq.(C1):

graphical method: • When x1 = 0, x2 = 300
• Maximize z = 50x1 + 60x2 • When x2 = 0, x1 = 150
• Subjected to: • Therefore(0,300) and (150,0) are the
• 2x1 + x2 ≤ 300 points required to sketch the
• 3x1 + 4x2 ≤ 509 equation of the line represented by
• 4x1 + 7x2 ≤ 812 C1
• x1, x2 ≥ 0 • Similarly the required points for
• Solution: sketching C2 are (0,127.25) and
• The mathematical model is (169.67,0) and the points for
• Maximize z = 50x1 + 60x2 sketching C3 are (0,116 ) and
• Subjected to: (203,0)
• 2x1 + x2 = 300 …………………(C1)
• 3x1 + 4x2 = 509………………..(C2)
• 4x1 + 7x2 = 812…………………(C3)
Arun Kumar S L
Lecturer in Mechanical Engineering
LPP – Graphical Method – Example(3) -
Continuation
• Feasible Solution Region: OABCD

Corner
Coordinates (x1, x2) z = 50x1 + 60x2
Points
O (0,0) 0
A (0,116) 6960
B (63,80) 7950
C (138,24) 8340
D (150,0) 7500

z is maximized at point C(138,24) and

hence the required solution is x1 = 138
and x2 = 24
Arun Kumar S L
Lecturer in Mechanical Engineering