Madan Mohan Malaviya Univ.

of Technology, Gorakhpur

SUBJECT NAME: OPERATION RESEARCH

SUBJECT CODE: BOE03

Unit 1
Linear Programming Problem

Department of Mathematics and Scientific Computing

Madan Mohan Malaviya University of Technology, Gorkhpur
16-10-2020 Side 1
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Formulation of Linear
Programming Problem

16-10-2020 Side 2
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
FORMULATION OF LINEAR PROGRAMMING PROBLEM

INTRODUCTION TO LINEAR PROGRAMMING

• Linear Programming is a problem solving approach that has been developed to help managers to make decisions.
• Linear Programming is a mathematical technique for determining the optimum allocation of resources and
obtaining a particular objective when there are alternative uses of the resources, money, manpower, material,
machine and other facilities.

FORMULATION OF THE LINEAR PROGRAMMING MODEL

The Decision Variables

Objective Function

The Constraints

16-10-2020 Side 3
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

TERMINOLOGY USED IN LINEAR PROGRAMMING PROBLEM

1. Components of LP Problem:
Every LPP is composed of a. Decision Variable, b. Objective Function, c. Constraints.
2. Optimization: Linear Programming attempts to either maximise or minimize the values of the objective
function.
Max 𝑧 or min 𝑧 = 𝑐1 𝑥1 + 𝑐2 𝑥2 + … + 𝑐𝑛 𝑥𝑛
3. Profit of Cost Coefficient: The coefficient of the variable in the objective function express the rate at which
the value of the objective function increases or decreases by including in the solution by chnaging of one unit of
each of the decision variable.
4. Constraints: The maximization (or minimization) is performed subject to a set of constraints. Therefore LP
can be defined as a constrained optimisation problem. They reflect the limitations of the resources.
5. Input-Output coefficients: The coefficient of constraint variables are called the Input-Output Coefficients.
They indicate the rate at which a given resource is unitized or depleted. They appear on the left side of the
constraints.
6. Capacities: The capacities or availability of the various resources are given on the right hand side of the
constraints.

16-10-2020 Side 4
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
BASIC DEFINITIONS
1. Decision Variables
These are the variables, whose quantitative values are to be found from the solution of the model so as to
maximize or minimize the objective function. The decision variables are usually denoted by 𝑥1 , 𝑥2 , 𝑥3 , … 𝑥𝑛 . It
may be controllable or uncontrollable.
Controllable variables are those, whose values are under control of the decision makers. Uncontrollable variables
are those, whose values are not under control.
2. Objective Function
It is the determinants of quantity either to be maximized or to be minimized. An objective function must include
all the possibilities with profit or cost coefficient per unit of output. It is denoted by 𝑧. The objective function can
be stated as Max 𝑧 or min 𝑧 = 𝑐1 𝑥1 + 𝑐2 𝑥2 + … + 𝑐𝑛 𝑥𝑛
3. Constraints (Inequalities)
These are the restrictions imposed on decision variables. It may be in terms of availability of raw materials,
machine hours, man-hours, etc.
𝑎11 𝑥1 + 𝑎12 𝑥2 + … … … … … . + 𝑎1𝑛 𝑥𝑛 (≤, =, ≥) 𝑏1
𝑎21 𝑥1 + 𝑎22 𝑥2 + … … … … … . + 𝑎2𝑛 𝑥𝑛 (≤, =, ≥)𝑏2
……………………………………………….
……………………………………………….
𝑎𝑚1 𝑥1 + 𝑎𝑚2 𝑥2 + … … … … … . + 𝑎𝑚𝑛 𝑥𝑛 ≤, =, ≥ 𝑏𝑚
𝑥1 , 𝑥2 , 𝑥3 … 𝑥𝑛 ≥ 0.
16-10-2020 Side 5
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
THE MATHEMATICAL EXPRESSION OF THE LP MODEL
The general LP Model can be expressed in mathematical terms as shown below:
Let
𝑎𝑖𝑗 Input-Output Coefficient
𝑐𝑗 Cost (Profit) Coefficient
𝑏𝑖 Capacities (Right Hand Side)
𝑥𝑗 Decision Variables
Objective Function:
Find a vector (𝑥1 , 𝑥2 , 𝑥3 , … , 𝑥𝑛 ) that minimize or maximize a linear objective function z where
z = 𝑐1 𝑥1 + 𝑐2 𝑥2 + … + 𝑐𝑛 𝑥𝑛
subject to linear constraints
𝑎11 𝑥1 + 𝑎12 𝑥2 + … … … … … . + 𝑎1𝑛 𝑥𝑛 (≤, =, ≥) 𝑏1
𝑎21 𝑥1 + 𝑎22 𝑥2 + … … … … … . + 𝑎2𝑛 𝑥𝑛 (≤, =, ≥)𝑏2
……………………………………………….
……………………………………………….
𝑎𝑚1 𝑥1 + 𝑎𝑚2 𝑥2 + … … … … … . + 𝑎𝑚𝑛 𝑥𝑛 (≤, =, ≥)𝑏𝑚
and non-negativity constraints
𝑥1 ≥ 0, 𝑥2 ≥ 0, … … … … … … , 𝑥𝑛 ≥ 0.
Here 𝑥1 , 𝑥2 , … . , 𝑥𝑛 are the decision variables.
16-10-2020 Side 6
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
THE MATHEMATICAL EXPRESSION OF THE LP MODEL IN MATRIX FORM

The LPP is usually expressed in matrix form and then it becomes:

Maximize or Minimize 𝑧 = 𝑐 𝑇 𝑥
Subjected to 𝐴𝑥(≤, =, ≥)𝑏, 𝑥 ≥ 0. Here

𝑐1 𝑥1
𝑎11 𝑎12 … 𝑎1𝑛
𝑐2 𝑥2
𝑎21 𝑎22 … 𝑎2𝑛 … and 𝑥 = … are the matrix of order 𝑛 × 1,
𝐴= …… …
… … … is a matrix of order m × 𝑛, 𝑐 =
… … … …
𝑎𝑚1 𝑎𝑚2 … 𝑎𝑚𝑛 𝑐𝑛 𝑥𝑛
𝑏1
𝑏2
and 𝑏 = … is the matrix of order 𝑚 × 1.
…
𝑏𝑚

16-10-2020 Side 7
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

MATHEMATICAL FORMULATION
Introduction
In fact, the most difficult problem in the application of management
science is the formulation of a model. Therefore, it is important to
consider model formulation before launching into the details of linear
programming solution. Model formulation is the process of transforming a
real word decision problem into an operations research model.
The procedure for mathematical formulation of linear programming
problem consists of the following major steps:
1. Write down the decision variables of the problem.
2. Formulate the objective function to be optimized (maximized or
minimized) as a linear function of the decision variables.
16-10-2020 Side 8
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

MATHEMATICAL FORMULATION

3. Formulate the other conditions of the problem such as resource

limitations, market constraints, inter-relation between variables etc. as
linear equations or in equations in terms of the decision variables.
4. Add the ‘Non-negativity’ constraint from the consideration that
negative values of the decision variables do not have any valid
physical interpretation
Now the objective function, the set of constraints and the non-negative
constraint together form a Linear Programming Problem.

16-10-2020 Side 9
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
MATHEMATICAL FORMULATION
Example:
A farmer has an area of 𝐴 square kilometres to be planted with a combination of wheat and barley. A
limited amount F of fertilizer and P of insecticide can be used, each of which is required in different
amounts per unit area for wheat (F1, P1) and barley (F2, P2). Let 𝑆1 be the selling price of wheat per
square Kilometre, and 𝑆2 the price of barley, and denote the area planted with wheat and barley
as 𝑥1 and 𝑥2 respectively. Formulate this for the optimal number of square kilometres to plant with
wheat vs. barley can be expressed as a linear programming problem.
Solution:
Here We need to maximise something, but what? 𝑧 = 𝑆1 𝑥1 + 𝑆2 𝑥2
What stops us making z as big as we want?
constraints on 𝑥1 an 𝑥2
▪ Must fit in the field 𝑥1 + 𝑥2 ≤ 𝐴
▪ Need enough fertilizer 𝐹1 𝑥1 + 𝐹2 𝑥2 ≤ 𝐹
▪ Need enough pesticide 𝑃1 𝑥1 + 𝑃2 𝑥2 ≤ 𝑃
▪ cannot plant a negative area 𝑥1 ≥ 0, 𝑥2 ≥ 0.
16-10-2020 Side 10
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
CONTD…

Hence Mathematical formulation of given problem as LPP is given by

Maximize 𝑧 = 𝑆1 𝑥1 + 𝑆2 𝑥2
subjected to constraints
𝑥1 + 𝑥2 ≤ 𝐴
𝐹1 𝑥1 + 𝐹2 𝑥2 ≤ 𝐹
𝑃1 𝑥1 + 𝑃2 𝑥2 ≤ 𝑃
𝑥1 ≥ 0, 𝑥2 > 0.

Example 2: A firm produces three products. These products are processed on three different machines. The time
required to manufacture one unit of each of the three products and the daily capacity of the three machines are
given in the table below:
Machine Time per unit (Minutes) Machine Capacity
Product 1 Product 2 Product 3 (Minutes per Day)

𝑀1 2 3 2 440
𝑀2 4 - 3 470
𝑀3 2 5 - 430
16-10-2020 Side 11
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
CONTD…
It is required to determine the daily number of units to be manufactured for each product. The profit
per unit for product 1, 2 and 3 is Rs. 4, Rs.3 and Rs.6 respectively. It is assumed that all the amounts
produced are consumed in the market. Formulate the mathematical (L.P.) model that will maximise
the daily profit.
Formulation of Linear Programming Model
Step 1
From the study of the situation find the key-decision to be made. In the given situation key decision
is to decide the extent of products 1, 2 and 3, as the extents are permitted to vary.
Step 2
Assume symbols for variable quantities noticed in step 1. Let the extents (amounts) of products 1, 2
and 3 manufactured daily be 𝑥1 , 𝑥2 and 𝑥3 units respectively.
Step 3
Express the feasible alternatives mathematically in terms of variable. Feasible alternatives are those
which are physically, economically and financially possible. In the given situation feasible
alternatives are sets of values of 𝑥1 , 𝑥2 and 𝑥3 units respectively.
Since negative production has no meaning and is not feasible. So 𝑥1 , 𝑥2 and 𝑥3 ≥ 0.

16-10-2020 Side 12
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
CONTD…
Step 4
Mention the objective function quantitatively and express it as a linear function of variables. In the
present situation, objective is to maximize the profit. i.e. Maximize z = 4𝑥1 + 3𝑥2 + 6𝑥3
Step 5
Put into words the influencing factors or constraints. These occur generally because of constraints on
availability (resources) or requirements (demands). Express these constraints also as linear
equations/inequalities in terms of variables. Here, constraints are on the machine capacities and can be
mathematically expressed as
2𝑥1 + 3𝑥2 + 2𝑥3 ≤ 440, 4𝑥1 + 0𝑥2 + 3𝑥3 ≤ 470, 2𝑥1 + 5𝑥2 + 0𝑥3 ≤ 430.
Hence L.P.P. formulation of given problem is
Maximize z = 4𝑥1 + 3𝑥2 + 6𝑥3
Subjected to constraint
2𝑥1 + 3𝑥2 + 2𝑥3 ≤ 440,
4𝑥1 + 3𝑥3 ≤ 470,
2𝑥1 + 5𝑥2 ≤ 430,
𝑥1 , 𝑥2 and 𝑥3 ≥ 0.
16-10-2020 Side 13
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

CONTD…
Example 3: An advertising agency is planning to launch an ad campaign. Media under consideration are T.V.,
Radio & Newspaper. Each medium has different reach potential and different cost. Minimum 10, 000, 000
households are to be reached through T.V. Expenditure on newspapers should not be more than Rs. 10, 00, 000.
Total advertising budget is Rs. 20 million. Following data is available:
Medium Cost per Unit (Rs.) Reach per unit
(No. of households)
Television 2, 00, 000 20, 00, 000
Radio 80, 000 10, 00, 000
Newspaper 40, 000 2, 00, 000

Formulate the LPP for best media selection that maximizes the total number of household reach.
Solution:
Decision Variables:
𝑥1 = Number of units of T.V. ads,
𝑥2 = Number of units of Radio ads,
𝑥3 = Number of units of Newspaper ads.
Objective function: (Maximize reach)
Max. z = 20, 00, 000 𝑥1 + 10, 00, 000 𝑥2 + 2, 00, 000𝑥3
16-10-2020 Side 14
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
CONTD..
Subject to constraints:
20, 00, 000 𝑥1 ≥ 10, 000, 000 ie 𝑥1 ≥ 5........ (for T.V.)
40, 000 𝑥3 ≤ 10, 00, 000 𝑖𝑒 𝑥3 ≤ 25 ........... (for Newspaper)
2, 00, 000𝑥1 + 80, 000𝑥2 + 40, 000𝑥3 ≤ 20, 000,000 ie 5𝑥1 + 2𝑥2 + 𝑥3 ≤ 500 .......... (Ad. budget)
𝑥1 , 𝑥2 , 𝑥3 ≥ 0.
Hence mathematical formulation of given L.P.P is
Max. z = 20, 00, 000 𝑥1 + 10, 00, 000 𝑥2 + 2, 00, 000𝑥3
Subjected to constraint
𝑥1 ≥ 5,
𝑥3 ≤ 25,
5𝑥1 + 2𝑥2 + 𝑥3 ≤ 500,
𝑥1 , 𝑥2 , 𝑥3 ≥ 0.

16-10-2020 Side 15
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
CONTD…
Example 4:
A manufacturer of a line of patent medicines is preparing a production plan on
medicines A and B. There are sufficient ingredients available to make 20,000
bottles of A and 40, 000 bottles of B but there are only 45, 000 bottles into
which either of the medicines can be put. Further more, it takes 3 hours to
prepare enough material to fill 1000 bottles of A, it takes one hours to prepare
enough material to fill 1000 bottles of B and there are 66 hours available for
this operation. The profit is Rs. 8 per bottle for A and Rs. 7 per bottle for B.
Formulate this problem as a linear programming problem.

16-10-2020 Side 16
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
CONTD…
Solution:
Let production of each of 𝐴 and 𝐵 are 𝑥1 unit bottels and 𝑥2 unit
bottoles respectively. Since profits on each bottle of 𝐴 and 𝐵 are Rs. 8 and
Rs. 7 per bottle respectively. So profit on 𝑥1 bottles of 𝐴 and 𝑥2 bottles
of B are 8𝑥1 and 7𝑥2 respectively.
Let 𝑧 be total profit on bottles so, 𝑧 = 8𝑥1 + 7𝑥2 .
Since, it takes 3 hours and 1 hour to prepare enough material to
fill 1000 bottles of Type A and Type B respectively. so x bottles
3𝑥1 𝑥2
of A and y bottles of B are preparing is hours and hours
1000 1000
respectively, but only 66 hours are available, so
3𝑥1 𝑥2
+ ≤ 66 ……..(Time available constraint)
1000 1000
or, 3𝑥1 + 𝑥2 ≤ 66000.

16-10-2020 Side 17
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
CONTD…
Since raw material available to make 20000 bottles of A and 40000 bottles of B but there
are 45000 bottles in which either of these medicines can be put so,
𝑥1 ≤ 20000, ……….(Bottle A constraint)
𝑥2 ≤ 40000, ……….(Bottle B constraint)
𝑥1 + 𝑥2 ≤ 45000……(combined production constraint)
with 𝑥1 , 𝑥2 ≥ 0.
So mathematical formulation of the given LPP is
Max z= 8𝑥1 + 7𝑥2
Subject to constraints
3𝑥1 + 𝑥2 ≤ 66000,
𝑥1 ≤ 2000,
𝑥2 ≤ 40000,
𝑥1 + 𝑥2 ≤ 45000
𝑥1 , 𝑥2 ≥ 0.

16-10-2020 Side 18
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
CONTD…
Example 5:
A firm can produce three types of cloth 𝐴, 𝐵 and 𝐶. Three kinds of wool
is required for it, say red, green and blue wools. One unit length of type
𝐴 cloth needs 2 yards of red wool, 5 yards of blue wools, one unit length
of type 𝐵 cloth needs 3 yards of red wool, 4 yards of green wool, and 2
yards of blue wool, and one unit length of type 𝐶 cloth needs 6 yards of
green and 5 yards of blue wools. The firm has only a stock of 10 yards
of red wool, 12 yards of green wool, and 17 yards of blue wool. It is
assumed that the income obtained from one unit length of type 𝐴, 𝐵 and
𝐶 are Rs 4.00, 5.00 and 6.00 respectively. Determine how the firm
should use the available material, so as to maximize the income from
the finished cloths.

16-10-2020 Side 19
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

CONTD…
Solution:
The above problem can be tabulated as:

Kinds of wool Types of cloth Stock of

A B C wool (yards)
Red 2 3 0 10
Green 0 4 6 12
Blue 5 2 5 17
Income from one unit of 4.00 5.00 6.00
clothes (Rs)

16-10-2020 Side 20
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
CONTD…
Let the firm produce 𝑥1 , 𝑥2 and 𝑥3 yards of three types of cloth
𝐴, 𝐵 and 𝐶 respectively. Therefore, 𝑥1 , 𝑥2 and 𝑥3 can be treated as
decision variables.
Since the profit per unit length of type 𝐴, 𝐵 and 𝐶 are given and we have
to maximize the profit, therefore, we have
Max 𝑍 = 4𝑥1 + 5𝑥2 + 6𝑥3
As per the statement of given problem, we have
2𝑥1 + 3𝑥2 + 0𝑥3 ≤ 10
0𝑥1 + 4𝑥2 + 6𝑥2 ≤ 12
5 𝑥1 + 2𝑥2 + 5𝑥3 ≤ 17,
𝑥1 , 𝑥2 , 𝑥3 ≥ 0.
16-10-2020 Side 21
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

CONTD…
So mathematical formulation of the given LPP is
𝑀𝑎𝑥 𝑧 = 4𝑥1 + 5𝑥2 + 6𝑥3
Subjected to constraints
2𝑥1 + 3𝑥2 ≤ 10
4𝑥2 + 6𝑥3 ≤ 12,
5𝑥1 + 2𝑥2 + 5𝑥3 ≤ 17
𝑥1 , 𝑥2 , 𝑥3 ≥ 0.

16-10-2020 Side 22
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

CONTD…

Example 6 A research laboratory has two melts 𝐴 and 𝐵 of copper (Cu), Nickel (Ni) and Zinc (Zn) alloy to make
up a new alloy. The composition of metals are as follows.
Melt Composition (Parts)

Cu Ni Zn

𝐴 3 2 1
𝐵 2 2 1
To make up a new alloy at least 15 kg of copper, 10 kg of nickel, and 6 kg of zinc are needed. Melt A cost Rs 45
per kg and melt B cost Rs 50 per kg. Formulate the L.P.P. for the quantities of each melt to be used to minimized
cost.

16-10-2020 Side 23
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
Solution:
The above data can be tabulated as follows.
Melt Requirement of
Composition elements(Rs.)
𝐴 𝐵
𝐶𝑢 3 2 15
𝑁𝑖 2 2 10
𝑍𝑛 1 1 6
Cost per kg of 45 50
melt (Rs)
Decision Variables: Let 𝑥1 and 𝑥2 be the quantity of melt A and B respectively. Therefore, 𝑥1 and 𝑥2 can be treated
as decision variables.
Objective Function: Since cost per kg melt of product A and B are given and we have to minimize the cost.
Therefore,
𝑀𝑖𝑛 𝑧 = 45𝑥1 + 50𝑥2

16-10-2020 Side 24
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Contd

Constraints: As per the statement of problem, we have

3 2
𝑥1 + 𝑥2 ≥ 15 𝑜𝑟 5𝑥1 + 4 𝑥2 ≥ 150,
6 5
2 2
𝑥
+ 𝑥2 ≥ 10 𝑜𝑟 5𝑥1 + 6 𝑥2 ≥ 150,
6 15
1 1
𝑥 + 𝑥 ≥6 𝑜𝑟 5𝑥1 + 6 𝑥2 ≥ 180,
6 1 5 2
Non-negative Restrictions: 𝑥1 , 𝑥2 ≥ 0.
So mathematical formulation of the given LPP is
𝑀𝑖𝑛 𝑍 = 45𝑥1 + 50𝑥2
subjected to constraints
5𝑥1 + 4 𝑥2 ≥ 150
5𝑥1 + 6 𝑥2 ≥ 150
5𝑥1 + 6 𝑥2 ≥ 180
𝑥1, 𝑥2 ≥ 0
16-10-2020 Side 25
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

SOLUTION OF LINEAR
PROGRAMMING PROBLEM-
GRAPHICAL METHOD

16-10-2020 Side 1
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
SOLUTION OF LINEAR PROGRAMMING PROBLEM

Solution of L.P.P.
The set of values of decision variables that satisfy the constraints of an LPP
is said to be solution of L.P.P.
Feasible Solution L.P.P.
The solution of L.P.P. which satisfy nonnegative restriction is called
feasible solution.
Infeasible solution of L.P.P.
The set of values of decision variables 𝑥𝑗 (𝑗 = 1 2 … … 𝑛) which do not
satisfy all the constraints and non-negativity conditions of an LP problem
simultaneously is said to constitute the infeasible solution to that linear
programming problem.
16-10-2020 Side 2
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

METHOD
Unbounded solution of L.P.P. :
A solution which increases (in case of maximization) or decreases (in case of
minimization) the value of objective function of the LP problem infinitely is called
unbounded solution.
Optimal Solution of L.P.P.
The feasible solution which optimize objective function is called optimal solution.
Method for Solving L.P.P
▪ Graphical Method
▪ Simplex Method
▪ Dual Simplex Method
▪ Two-phase Simplex Method
▪ Big M Method
16-10-2020 Side 3
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
GRAPHICAL METHOD OF SOLVING LINEAR PROGRAMMING PROBLEMS
Graphical method is applicable to find the simple linear programming problem
with two decision variables. Various steps for solving the problems are given
below:
1. Draw the nonnegative restrictions as axes.
2. Consider each inequality constraint as equation and draw each equations on
the graph such that each will geometrically represent a straight line.
3. Take region of the constraint as an inequality form. If the inequality
constraint corresponding to that line is ≤, then the region below the line in
the first quadrant is to be shaded. For the inequality constraint ≥, then the
region above the line in the first quadrant is shaded. The points lying in
common region will satisfy all the constraints simultaneously. This
common region is called feasible region.
16-10-2020 Side 4
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

CONTD….

4. Locate the corner points (Vertices) of the feasible region. These corner
points are also called extreme point.
5. Test the objective function at each corner point of the feasible region
and choose the point, where objective function obtains optimal value.

16-10-2020 Side 5
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

GRAPHICAL METHOD

EXAMPLE 1:
Solve the following L.P.P. by graphical method
Min z = 20𝑥1 + 10𝑥2
Subject to constraint
𝑥1 + 2 𝑥2 ≤ 40,
3𝑥1 + 𝑥2 ≥ 30,
4𝑥1 + 3 𝑥2 ≥ 60,
𝑥1 , 𝑥2 ≥ 0.
16-10-2020 Side 6
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

GRAPHICAL METHOD

Solution
Convert all the inequalities of the constraints into equations, we
have
𝑥1 + 2 𝑥2 = 40
3𝑥1 + 𝑥2 = 30
4𝑥1 + 3 𝑥2 = 60
𝑥1 + 2 𝑥2 = 40 passes through 0, 20 & (40, 0)
3𝑥1 + 𝑥2 = 30 passes through 0, 30 & (10, 0)
4𝑥1 + 3 𝑥2 = 60 passes through 0, 20 & 15, 0
16-10-2020 Side 7
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

CONTD…..
Plot above equations on graph, we have

Here feasible region is 𝐴𝐵𝐶𝐷.

The coordinates of 𝐴𝐵𝐶𝐷 are
𝐴 15, 0 , 𝐵 40, 0 , 𝐶 4, 18 ,
and 𝐷(6, 12)

𝑥1 + 2 𝑥2 = 40
3𝑥1 + 𝑥2 = 30
4𝑥1 + 3 𝑥2 = 60

16-10-2020 Side 8
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
CONTD…
Here feasible region is ABCD.
The coordinates of extreme points of region 𝐴𝐵𝐶𝐷 are A(15, 0) 𝐵(40, 0), 𝐶(4, 18)
𝐷(6, 12)

Corner Points Coordinate Value of objective function 𝑧

𝐴 (15, 0) 300
𝐵 (40, 0) 800
𝐶 (4, 18) 260
𝐷 (6, 12) 240
Therefore, minimum value of 𝑍 occurs at 𝐷(6, 12). Hence, optimal solution is 𝑥1 =
6, 𝑥2 = 12.

16-10-2020 Side 9
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

CONTD…

Example 2:
Solve the following L.P.P. using graphical methods
Max z = 6 𝑥1 + 8 𝑥2
Subject to
5𝑥1 + 10𝑥2 ≤ 60,
4𝑥1 + 4 𝑥2 ≤ 40,
𝑥1 , 𝑥2 ≥ 0.

16-10-2020 Side 10
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
CONTD…
Solution: Convert all the equalities of the constraint into equations, we have

16-10-2020 Side 11
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

CONTD…
Here feasible region is O𝐴𝐵𝐶.
The coordinates of O𝐴𝐵𝐶 are O(0, 0) 𝐴(10, 0), 𝐵(8, 2), 𝐶(0, 6)
Now
Corner Points Coordinate Value of 𝑧
𝑜 (0, 0) 0
𝐴 (10, 0), 60
𝐵 (8, 2) 64
𝐶 (0, 6) 48
Therefore, Maximum value of 𝑍 occurs at B(8, 2). Hence, optimal solution is 𝑥1 =
8, 𝑥2 = 2.

16-10-2020 Side 12
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Special Cases in Linear Programming Problem

Unbounded Solution

When values of decision variable may be increased infinitely without violating any of
the constraints, the solution space (feasible region) is said to be unbounded. If value of
objective function increases (for maximization case) or decreases (for minimization
case) infinitely, solution space and objective function both are said to be unbounded.

16-10-2020 Side 13
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

CONTD…
Example 3:
Solve the following L.P.P. by graphical method
Min z = 3𝑥1 + 2𝑥2
Subject to constraint
𝑥1 + 4 𝑥2 ≥ 12,
𝑥1 + 𝑥2 ≥ 6,
5𝑥1 + 𝑥2 ≥ 10,
𝑥1 , 𝑥2 ≥ 0.

16-10-2020 Side 14
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

CONTD…
Convert all the equalities of the constraint into equations, we have
𝑥1 + 4 𝑥2 = 12, 𝑥1 + 𝑥2 = 6, 5𝑥1 + 𝑥2 = 10,
Plot the above equations on graph, we have

16-10-2020 Side 15
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

CONTD...

16-10-2020 Side 16
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

CONTD…

Example 4

Solve the following L.P.P. by graphical method

Max z = 3𝑥1 + 2𝑥2
Subject to constraint
𝑥1 − 𝑥2 ≤ 1,
𝑥1 + 𝑥2 ≥ 3,
𝑥1 , 𝑥2 ≥ 0.

16-10-2020 Side 17
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
CONTD…
Convert the inequality constraints into equations. We have
𝑥1 − 𝑥2 = 1, 𝑥1 + 𝑥2 = 3,
𝑥1 – 𝑥2 = 1 passes through (0, – 1) and (1, 0) , 𝑥1 + 𝑥2 = 3 passes through (0, 3) and (3, 0)
Plot above equations on graph, we have

16-10-2020 Side 18
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

CONTD…

• Here the solution space is unbounded. The value of objective

function at the vertices A and B are Z(A) = 6, Z(B) = 8. But
there exists points in the feasible region for which the value
of the objective function is more than 8. In fact, the
maximum value of z occurs at infinity. Hence, the problem
has an unbounded solution.

16-10-2020 Side 19
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Special Cases in Linear Programming Problem

Infeasible Solution
Infeasible solution arises when there is no solution to linear programming problem
that satisfies all the constraint simultaneously. This means there would be no
unique feasible region. Such a problem arises when wrong model is formulated
that has conflicting constraints. Every possible solution violates at least one
constraint.
Example 5: Use Graphical method to solve the following LP problem
𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑧 = 6𝑥1 − 4𝑥2
subjected to constraints
2𝑥1 + 4𝑥2 ≤ 4;
4𝑥1 + 8𝑥2 ≥ 16;
𝑥1 , 𝑥2 ≥ 0

16-10-2020 Side 20
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
2𝑥1 + 4𝑥2 ≤ 4;
4𝑥1 + 8𝑥2 ≥ 16;
𝑥1 , 𝑥2 ≥ 0

16-10-2020 Side 21
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Solution:

𝒙𝟐
1

1 2 3 4
𝒙𝟏
The constraints are plotted on graph shown in the figure. We can see that there is no unique
feasible solution space allows us to obtain unique set of values of 𝑥1 and 𝑥2 that satisfy all
the constraints simultaneously. Hence there is no feasible solution to the given problem.
16-10-2020 Side 22
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

An Infeasible Solution
Example 6 Plot given inequalities on graph, we have
Maximize Z = 5x1 + 3x2
subject to: 4x1 + 2x2  8
x1  4
x2  6
x1, x2  0.
Solution:
There is no common feasible region generated by
two constraints together i.e. we cannot identify
even a single point satisfying all the constraints
simultaneously. Hence there is no feasible
solution to the given problem.

16-10-2020 Side 23
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
4x1 + 2x2  8
x1  4
x2  6
x1, x2  0.

16-10-2020 Side 24
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Special Cases in Linear Programming Problem

Alternative or Multiple Optimal Solutions
There are two conditions that should be satisfied for the alternative solution
1. The given objective function should be parallel to the constraint that
forms the boundary of the feasible region.
2. The constraints should form a boundary of the feasible region in the
direction of the optimal movement of the objective function.
Note: If we get an optimal solution of the objective function at two extreme
points 𝑥 = 𝑥1 , 𝑥2 and 𝑦 = 𝑦1 , 𝑦2 , then L.P. problem has infinite
numbers of solution at the point 𝜆𝑥 + 1 − 𝜆 𝑦 = (𝜆𝑥1 + 1 − 𝜆 𝑦1 , 𝜆𝑥2 +
1 − 𝜆 𝑦2 ) 0 ≤ 𝜆 ≤ 1.

16-10-2020 Side 25
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Alternative or Multiple Optimal Solutions

Example 7:
Maximize 𝑧 = 𝑥1 + 2𝑥2
subject to
𝑥1 ≤ 80,
𝑥2 ≤ 60,
60 − 20𝜆, 10𝜆 + 50
5𝑥1 + 6𝑥2 ≤ 600,
𝑥1 + 2𝑥2 ≤ 160,
𝑥1, 𝑥2 ≥ 0.

16-10-2020 Side 26
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

CONTD…
Here feasible region is OTRQPU. The coordinates of extreme points region OTRQPU are
O 0, 0 , 𝑇(80, 0), 𝑅(80,100/3), 𝑄(60, 50), 𝑃(40,60), 𝑈(0,60).
Since at the two different points P and Q, the Corner point or Coordinate Value of the objective
optimum value of objective function is same. Extreme point function
Hence all points from P to Q lying on line PQ 𝑂 (0,0) 0
represent optimal solutions and all these will 𝑇 (80,0) 80
give the same optimal value (maximum profit) 𝑅 (80, 100/3), 440/3
of 160. This is indicated by the fact that both 𝑄 (60, 50), 160
the points P with co-ordinates (40, 60) and Q 𝑃 (40,60), 160
with co-ordinates (60, 50) are on the line 𝑥1 + 𝑄 (0,60). 120
2𝑥2 = 160. Thus, every point on the
line PQ maximizes the value of the objective function and the problem has infinite number
of optimal solutions. Hence given LPP has infinitely many solution at the point 𝑥ҧ = 𝜆𝑥 +
1 − 𝜆 𝑦 = 𝜆 40, 60 + 1 − 𝜆 60,50 = 60 − 20𝜆, 10𝜆 + 50 0 ≤ 𝜆 ≤ 1.

16-10-2020 Side 27
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Alternative or Multiple Optimal Solutions

Example 8:
𝑀𝑎𝑥 𝑧 = 10𝑥1 + 6𝑥2
subjected to
constraint
5𝑥1 + 3𝑥2 ≤ 30; 10
𝑥1 + 2𝑥2 ≤ 18; 𝑥2 𝐵
𝐶
𝑥1 , 𝑥2 ≥ 0. 8

2
𝐴
𝑂 2 4 6 8 10 12 14 16 18
𝑥1

16-10-2020 Side 28
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

CONTD…
Here feasible region is 𝑂𝐴𝐵𝐶. The coordinates of extreme points region 𝑂𝐴𝐵𝐶 are O 0, 0 ,
A 6, 0 , B(6Τ7, 60Τ7), 𝐶(0, 9).
Since at the two different points A and B, the Corner point or Coordinate Value of the
optimum value of objective function is same. Extreme point objective function
Hence all points from A to B lying on line AB 𝑂 (0,0) 0
represent optimal solutions and all these will 𝐴 (6,0) 60
give the same optimal value (maximum value 𝐵 (6Τ7 , 60/7), 60
of 𝑧) of 60. This is indicated by the fact that 𝐶 (0, 9), 54

both the points A with co-ordinates (6, 0) and B with co-ordinates (6Τ7 , 60Τ7) are on the
line 5𝑥1 + 3𝑥2 = 30. Thus, every point on the line AB maximizes the value of the
objective function and the problem has infinite number of optimal solutions. Hence given
LPP has infinitely many solution at the point 𝑥ҧ = 𝜆𝑥 + 1 − 𝜆 𝑦 = 𝜆 6, 0 + (1 −
36 6 60
𝜆)(6Τ7,60/7)= 𝜆+ , 1−𝜆 0 ≤ 𝜆 ≤ 1.
7 7 7

16-10-2020 Side 29
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
Limitations or Assumptions of Linear Programming Problem
Linearity of relations: A primary requirement of linear programming is that the objective function and every
constraint must be linear. However, in real life situations, several business and industrial problems are
nonlinear in nature.
Single objective: Linear programming takes into account a single objective only, i.e., profit maximization or
cost minimization. However, in today's dynamic business environment, there is no single universal objective
for all organizations.
Certainty or Deterministic: Linear Programming assumes that the values of co-efficient of decision variables
are known with certainty. Due to this restrictive assumption, linear programming cannot be applied to a wide
variety of problems where values of the coefficients are probabilistic.
Constant parameters: Parameters appearing in LP are assumed to be constant, but in practical situations it is
not so.
Divisibility: In linear programming, the decision variables are allowed to take non-negative integer as well as
fractional values. However, we quite often face situations where the planning models contain integer valued
variables. For instance, trucks in a fleet, generators in a powerhouse, pieces of equipment, investment
alternatives and there are a myriad of other examples. Rounding off the solution to the nearest integer will not
yield an optimal solution. In such cases, linear programming techniques cannot be used.
16-10-2020 Side 30
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
CONTD…
Example 4:
A manufacturer of a line of patent medicines is preparing a production plan on
medicines A and B. There are sufficient ingredients available to make 20,000
bottles of A and 40, 000 bottles of B but there are only 45, 000 bottles into
which either of the medicines can be put. Further more, it takes 3 hours to
prepare enough material to fill 1000 bottles of A, it takes one hours to prepare
enough material to fill 1000 bottles of B and there are 66 hours available for
this operation. The profit is Rs. 8 per bottle for A and Rs. 7 per bottle for B.
Solve the problem to maximize the profit.

16-10-2020 Side 31
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
CONTD…
Solution:
Let production of each of 𝐴 and 𝐵 are 𝑥1 unit bottels and 𝑥2 unit
bottoles respectively. Since profits on each bottle of 𝐴 and 𝐵 are Rs. 8 and
Rs. 7 per bottle respectively. So profit on 𝑥1 bottles of 𝐴 and 𝑥2 bottles
of B are 8𝑥1 and 7𝑥2 respectively.
Let 𝑧 be total profit on bottles so, 𝑧 = 8𝑥1 + 7𝑥2 .
Since, it takes 3 hours and 1 hour to prepare enough material to
fill 1000 bottles of Type A and Type B respectively. so x bottles
3𝑥1 𝑥2
of A and y bottles of B are preparing is hours and hours
1000 1000
respectively, but only 66 hours are available, so
3𝑥1 𝑥2
+ ≤ 66 ……..(Time available constraint)
1000 1000
or, 3𝑥1 + 𝑥2 ≤ 66000.

16-10-2020 Side 32
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

CONTD…
Since raw material available to
make 20000 bottles of A and 40000 bottles
of B but there are 45000 bottles in which either
of these medicines can be put so,
𝑥1 ≤ 20000, ……….(Bottle A constraint)
60000
𝑥2 ≤ 40000, ……….(Bottle B constraint)
𝑥1 + 𝑥2 ≤ 45000……(combined production 𝑥2
𝐸 𝐷
constraint) 40000
with 𝑥1 , 𝑥2 ≥ 0.
So mathematical formulation of the given LPP 𝐶
20000
is Max z= 8𝑥1 + 7𝑥2
Subject to constraints 𝐵
3𝑥1 + 𝑥2 ≤ 66000,
𝑥1 ≤ 2000, 𝑥2 ≤ 40000, 𝑂 𝐴 20000 40000 60000
𝑥1 + 𝑥2 ≤ 45000 𝑥1
𝑥1 , 𝑥2 ≥ 0.
16-10-2020 Side 33
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

CONTD…
Here feasible region is O𝐴𝐵𝐶𝐷𝐸. The coordinates of O𝐴𝐵𝐶 are O 0, 0 , 𝐴 20000, 0 ,
𝐵 20000,6000 , 𝐶 10500, 34500 , 𝐷 5000,40000 , 𝐸(0,40000).

Corner Points Coordinate Value of 𝑧

𝑜 (0, 0) 0
𝐴 (20000, 0), 160000
𝐵 (20000, 6000) 202000
𝐶 (10500, 34500) 325500
𝐷 (5000, 40000) 320000
𝐸 (0,40000) 280000
Therefore, Maximum value of 𝑧 occurs at C(10500, 34500). Hence, optimal solution
is 𝑥1 = 10500, 𝑥2 = 10500 and max profit =Rs. 325500
16-10-2020 Side 34
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Simplex Method for Solving

Linear Programming
Problem

16-10-2020 Side 1
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
Canonical form of L.P.P.
1. Max 𝑧 = 𝑐1 𝑥1 + 𝑐2 𝑥2 + … + 𝑐𝑛 𝑥𝑛
Subject to constraints
𝑎11 𝑥1 + 𝑎12 𝑥2 + … … … … … . + 𝑎1𝑛 𝑥𝑛 ≤ 𝑏1
𝑎21 𝑥1 + 𝑎22 𝑥2 + … … … … … . + 𝑎2𝑛 𝑥𝑛 ≤ 𝑏2
……………………………………………….
……………………………………………….
𝑎𝑚1 𝑥1 + 𝑎𝑚2 𝑥2 + … … … … … . + 𝑎𝑚𝑛 𝑥𝑛 ≤ 𝑏𝑚
𝑥1 , 𝑥2 , 𝑥3 … 𝑥𝑛 ≥ 0.
2. Min 𝑧 = 𝑐1 𝑥1 + 𝑐2 𝑥2 + … + 𝑐𝑛 𝑥𝑛
Subject to constraints
𝑎11 𝑥1 + 𝑎12 𝑥2 + … … … … … . + 𝑎1𝑛 𝑥𝑛 ≥ 𝑏1
𝑎21 𝑥1 + 𝑎22 𝑥2 + … … … … … . + 𝑎2𝑛 𝑥𝑛 ≥ 𝑏2
……………………………………………….
……………………………………………….
𝑎𝑚1 𝑥1 + 𝑎𝑚2 𝑥2 + … … … … … . + 𝑎𝑚𝑛 𝑥𝑛 ≥ 𝑏𝑚
𝑥1 , 𝑥2 , 𝑥3 … 𝑥𝑛 ≥ 0.
16-10-2020 Side 2
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Standard form of L.P.P.

Max 𝑧 = 𝑐1 𝑥1 + 𝑐2 𝑥2 + … + 𝑐𝑛 𝑥𝑛
Subject to constraints
𝑎11 𝑥1 + 𝑎12 𝑥2 + … … … … … . + 𝑎1𝑛 𝑥𝑛 = 𝑏1
𝑎21 𝑥1 + 𝑎22 𝑥2 + … … … … … . + 𝑎2𝑛 𝑥𝑛 = 𝑏2
……………………………………………….
……………………………………………….
𝑎𝑚1 𝑥1 + 𝑎𝑚2 𝑥2 + … … … … … . + 𝑎𝑚𝑛 𝑥𝑛 = 𝑏𝑚
𝑥1 , 𝑥2 , 𝑥3 … 𝑥𝑛 ≥ 0.
The standard form of L.P.P. Should have following characteristics
1. Objective function should be of maximization type.
2. All the constraint should be expressed in equation form.
3. The right hand side of each constraints should be non-negative.

16-10-2020 Side 3
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Procedure for changing into Standard form

1. The minimization of function z is equivalent to the maximization of
negative expression of this function ie – 𝑧. Hence objective function
Min 𝑧 = 𝑐1 𝑥1 + 𝑐2 𝑥2 + … + 𝑐𝑛 𝑥𝑛 is equivalent to Max
𝑧 ∗= −𝑧 = −(𝑐1 𝑥1 + 𝑐2 𝑥2 + … + 𝑐𝑛 𝑥𝑛 )
2. The inequality constraints are changed into equality constraints by
adding or subtracting a nonnegative variable from left side of the
constraint . These new variables are known as slack variable. These
variable are added if the constraints are ≤ and subtracted if the
constraint are ≥. Since in case of ≥ constraints, the subtracted
variable represent the surplus of left hand side over right hand side,
it is commonalty known as surplus variable in fact negative slack.
3. If R.H.S. of any constraints is negative, then it should be made non-
negative by multiplying −1 on both side of the constraints.
4. Replace each unrestricted decision variable with difference of two
non-negative variable.

16-10-2020 Side 4
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Standard Form
Example: Express the following LPP in standard form
Min 𝑧 = 2𝑥1 + 𝑥2 + 4𝑥3
Subjected to constraints
−2𝑥1 + 4𝑥2 ≤ 4
𝑥1 + 2𝑥2 + 𝑥3 ≥ 5,
2𝑥1 + 3𝑥2 ≤ 2,
𝑥1 , 𝑥2 ≥ 0 𝑎𝑛𝑑 𝑥3 is unrestricted.
Solution: Since 𝑥3 is unrestricted. So suppose 𝑥3 = 𝑥3′ − 𝑥3′′
Hence Standard form Is given by
Max 𝑧 ∗ = −𝑧 = −(2𝑥1 + 𝑥2 + 4𝑥3′ − 4𝑥3′′ )
Subjected to constraints
−2𝑥1 + 4𝑥2 + 𝑠1 = 4
𝑥1 + 2𝑥2 + 𝑥3′ − 𝑥3′′ − 𝑠2 = 5,
2𝑥1 + 3𝑥2 + 𝑠3 = 2,
𝑥1 , 𝑥2 , 𝑥3′ , 𝑥3′′ , 𝑠1 , 𝑠2 , 𝑠3 ≥ 0.
Here 𝑠1 and 𝑠3 are the slacks variable and 𝑠2 is surplus variable.
16-10-2020 Side 5
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
Standard Form
Example: Express the following LPP in standard form
M𝑎𝑥 𝑧 = 3𝑥1 + 2𝑥2 + 5𝑥3
Subjected to constraints
2𝑥1 + 3𝑥2 − 2𝑥3 ≤ 40
4𝑥1 − 2𝑥2 + 𝑥3 ≤ 24,
𝑥1 − 5𝑥2 − 6𝑥3 ≥ 2,
𝑥1 ≥ 0.
Sol. Since x2 and 𝑥3 are unrestricted. So suppose 𝑥2 = 𝑥2′ − 𝑥2′′ and 𝑥3 =
𝑥3′ − 𝑥3′′ .
Hence Standard form Is given by
M𝑎𝑥 𝑧 = 3𝑥1 + 2(𝑥2′ − 𝑥2′′ ) + 5(𝑥3′ − 𝑥3′′ )
Subjected to constraints
2𝑥1 + 3 𝑥2′ − 𝑥2′′ − 2 𝑥3′ − 𝑥3′′ + s1 = 40
4𝑥1 − 2 𝑥2′ − 𝑥2′′ + 𝑥3′ − 𝑥3′′ + 𝑠2 = 24,
𝑥1 − 5 𝑥2′ − 𝑥2′′ − 6 𝑥3′ − 𝑥3′′ − 𝑠3 = 2,
𝑥1 , 𝑥2′ , 𝑥2′′ , 𝑥3′ , 𝑥3′′ , 𝑠1 , 𝑠2 , 𝑠3 ≥ 0.
Here 𝑠1 and 𝑠2 are the slack variable and 𝑠3 is surplus variable.
16-10-2020 Side 6
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
BASIC AND BASIC FEASIBLE SOLUTIONS
Matrix form of L.P.P. (Standard Form)
Maximize 𝑧 = 𝑐 𝑇 𝑥
subjected to constraints
𝐴𝑥 = 𝑏, 𝑥 ≥ 0.
Here

𝑐 𝑥1
𝑎11 𝑎12 … 𝑎 1𝑛 1
𝑐 𝑥2
𝑎21 𝑎22 … 𝑎 2𝑛 2
… and 𝑥 = …
𝐴= … … …
… … … …
… is a matrix of order m × 𝑛, 𝑐 =
… …
𝑎𝑚1 𝑎𝑚2 … 𝑎𝑚𝑛 𝑐𝑛 𝑥𝑛
𝑏1
𝑏2
are the matrix of order 𝑛 × 1, and 𝑏 = … is the matrix of order 𝑚 × 1.
…
𝑏𝑚
16-10-2020 Side 7
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
BASIC AND BASIC FEASIBLE SOLUTIONS
Here given system of equations 𝐴𝑥 = 𝑏 has 𝑛 unknowns and 𝑚 equations
(𝑚 < 𝑛).
Suppose 𝐴 be 𝑚 × 𝑛 matrix of rank 𝑚. Let 𝐵 be any 𝑚 × 𝑚 non-singular
submatrix after deleting (𝑛 − 𝑚) columns of 𝐴. Now equate zero to (𝑛 −
𝑚) varriables which are not associated with columns of 𝐵.
A basic solution is one in which (𝑛 − 𝑚) variables not associated with
columns of 𝐵 equal to zero and solution obtained for the remaining 𝑚
number variables.
The m variable which may be different from zero, are called basic variables
and remaining variables are called non-basic variables.
𝐵𝑥𝐵 = 𝑏,
𝑥𝐵 = 𝐵−1 𝑏
Basic feasible Solution:
This is the basic solution that satisfies the non-negativity conditions.

16-10-2020 Side 8
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
NONDEGENERATE BASIC FEASIBLE SOLUTION AND OPTIMAL
SOLUTION
Nondegenerate basic feasible solution:
This is a basic feasible solution that has got exactly 𝑚 positive decision
variables 𝑥𝑖 .
Optimal solution:
A feasible solution that optimized the objective function is called an optimal
solution

16-10-2020 Side 9
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
Example
Example: Find all basic solution of the following system of equation
𝑥1 + 2𝑥2 + 𝑥3 = 4
2𝑥1 + 𝑥2 + 5𝑥3 = 5
Solution:
𝑥1
1 2 1 𝑥 4
2 =
2 1 5 𝑥 5
3
1 2 1 4
Here 𝐴 = ,b=
2 1 5 5
𝑚 = 2, 𝑛 = 3
(i) 𝑥1 = 0
2 1 𝑥2 4
=
1 5 𝑥3 5

𝑥2 −1 5/3
2 1 4 1 5 −1 4
𝑥3 = = =
1 5 5 9 −1 2 5 2/3

16-10-2020 Side 10
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
Basic Solutions
(ii) 𝑥2 = 0
1 1 𝑥1 4
=
2 5 𝑥3 5
𝑥1 −1
1 1 4 1 5 −1 4 5
𝑥3 = = =
2 5 5 3 −2 1 5 −1
(iii) 𝑥3 = 0
1 2 𝑥1 4
=
2 1 𝑥2 5
𝑥1 −1
1 2 4 1 1 −2 4 2
𝑥2 = = =
2 1 5 −3 −2 1 5 1
Hence
5 2
• All basic solutions are 0, , , 5, 0, −1 and 2,1,0 ,
3 3
5 2
• All basic feasible solution are 0, , , and 2,1,0 .
3 3
• All basic feasible solution are nondegenerate basic feasible solution.

16-10-2020 Side 11
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
Example:
Example: Find the basic solutions to the following problem
𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑧 = 𝑥1 + 3𝑥2 + 3𝑥3
Subjected to constraint
𝑥1 + 2𝑥2 + 3𝑥3 = 4
2𝑥1 + 3𝑥2 + 5𝑥3 = 7.
Also find which of the basic solution are
(a) Basic feasible solution
(b) Non-degenerate basic feasible solution
(c) Optimal basic feasible solution
Solution:
𝑥1
1 2 3 𝑥 4
2 =
2 3 5 𝑥 7
3
1 2 3 4
Here 𝐴 = ,𝐵=
2 3 5 7
𝑚 = 2, 𝑛 = 3

16-10-2020 Side 12
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
Basic Solutions
(i) 𝑥1 = 0
2 3 𝑥2 4
=
3 5 𝑥3 7
𝑥2 −1
2 3 4 1 5 −3 4 −1
𝑥3 = = =
3 5 7 1 −3 2 7 2
(ii) 𝑥2 = 0
1 1 𝑥1 4
=
2 5 𝑥3 7
𝑥1 −1
1 3 4 1 5 −3 4 1
𝑥3 = = =
2 5 7 −1 −2 1 7 1
(iii) 𝑥3 = 0
1 2 𝑥1 4
=
2 3 𝑥2 7
𝑥1 −1
1 2 4 1 3 −2 4 2
𝑥3 = = =
2 3 7 −1 −2 1 7 1

16-10-2020 Side 13
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Basic Solutions
Hence basic solutions are 0, −1, 2 , 1,0,1 , [2, 1, 0]
(a) Basic feasible solutions are 1,0,1 , [2, 1, 0]
(b) Nondegenerate basic feasible solutions are 1,0,1 , 2, 1, 0
(c) Values of objective function at basic feasible solutions are
(i) 𝑥1 = 1, 𝑥2 = 0, 𝑥3 = 1, 𝑧=4
(ii) 𝑥1 = 2, 𝑥2 = 1, 𝑥3 = 0, 𝑧=5
Hence optimal basic feasible solution is
𝑥1 = 2, 𝑥2 = 1, 𝑥3 = 0, 𝑧=5

16-10-2020 Side 14
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

How many numbers of basic solutions?

How many numbers of basic solutions?
𝑛!
𝑛𝐶𝑚 =
𝑛−𝑚!𝑚!
The problem we have just solved has 3 basic solutions.
Now consider a problem of 10 variables and 8 equations, we will
have 45 different basic solutions.
If a problem of 15 variables and 10 equations, we will have
3003 different basic solutions.
As such, it is not possible to find solutions for all the basic
solutions. Moreover, many basic solutions, we may get
infeasible solutions. As such we need some set of rules to switch
from one feasible solution another feasible solution.
This rule may be known as Simplex method.
16-10-2020 Side 15
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
Simplex Method For solving LPP
We use this method for solving LPP where we use only slack variable in its
standard form.
Example: Solve the following LP problem by simplex method
𝑀𝑖𝑛 𝑧 = 𝑥1 − 3𝑥2 + 2𝑥3
Subjected to constraints
3𝑥1 − 𝑥2 + 3𝑥3 ≤ 7, −2𝑥1 + 4𝑥2 ≤ 12,
−4𝑥1 + 3𝑥2 + 8𝑥3 ≤ 10, 𝑥1 , 𝑥2 , 𝑥3 ≥ 0.
Solution:
Step 1: Write the given LPP in to standard form.
𝑀𝑎𝑥 𝑧 ∗ = −𝑧 = −𝑥1 + 3𝑥2 − 2𝑥3
Subjected to constraints
3𝑥1 − 𝑥2 + 3𝑥3 + 𝑠1 = 7,
−2𝑥1 + 4𝑥2 + 𝑠2 = 12,
−4𝑥1 + 3𝑥2 + 8𝑥3 + 𝑠3 = 10,
𝑥1 , 𝑥2 , 𝑥3 , 𝑠1 , 𝑠2 , 𝑠3 ≥ 0.
Here 𝑠1 , 𝑠2 𝑎𝑛𝑑 𝑠3 are the slack variable.
16-10-2020 Side 16
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
Simplex Method
Step 2: Write down the coefficient of all the variables in the standard form of
L.P.P. in tabular form, as shown in the table below. In order to find the initial
basic feasible solution 𝑥𝐵 = 𝐵−1 𝑏.
Here 𝐵 is the identity matrix. Since we have taken 𝑥1 = 0, 𝑥2 = 0, 𝑥3 = 0 as
non basic variable and 𝑠1 , 𝑠2 and 𝑠3 are the basic variable.
Hence 𝑠1 = 7, 𝑠2 = 12 and 𝑠3 = 10 𝑥𝐵 = {𝑠1 , 𝑠2 , 𝑠3 }

16-10-2020 Side 17
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Write the given LPP in to standard form.

𝑀𝑎𝑥 𝑧 ∗ = −𝑧 = −𝑥1 + 3𝑥2 − 2𝑥3
Subjected to constraints
3𝑥1 − 𝑥2 + 3𝑥3 + 𝑠1 = 7,
−2𝑥1 + 4𝑥2 + 𝑠2 = 12,
−4𝑥1 + 3𝑥2 + 8𝑥3 + 𝑠3 = 10,
𝑥1 , 𝑥2 , 𝑥3 , 𝑠1 , 𝑠2 , 𝑠3 ≥ 0.
Here 𝑠1 , 𝑠2 𝑎𝑛𝑑 𝑠3 are the slack variable.

16-10-2020 Side 18
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Simplex Method

𝑐𝑗 −1 3 −2 0 0 0

Basic variable 𝐶𝐵 𝑋𝐵 𝑥1 𝑥2 𝑥3 𝑠1 𝑠2 𝑠3 Min

in the basis Ratio
𝑠1 0 7 3 −1 3 1 0 0

𝑠2 0 12 −2 4 0 0 1 0 3∗ Key
Row
𝑠3 0 10 −4 3 8 0 0 1 10/3

𝑧𝑗 − 𝑐𝑗 1 −3∗ 2 0 0 0

Key Column
• 𝑧𝑗 − 𝑐𝑗 = σ 𝐶𝐵 𝑥𝑖 − 𝑐𝑗 . Here 𝐶𝐵 denotes coefficient of basic variable in the objective
function.
16-10-2020 Side 19
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Simplex Method

Step 3: (Condition for optimality) If all the 𝑧𝑗 − 𝑐𝑗 ≥ 0, then

this basic feasible will be optimal. We shall always get an
identity matrix corresponding to basic variable and 𝑧𝑗 − 𝑐𝑗 will
be zero corresponding to basic variable.
Step 4: (Select the variable to enter the basis) If 𝑧𝑗 − 𝑐𝑗 ≤ 0
for any non basic variable, this indicate an improvement in the
value of the objective is possible. Now non basic variable for
which 𝑧𝑗 − 𝑐𝑗 is most negative will enter in to new solution and
column corresponding to this variable is called key column or
pivot column.

16-10-2020 Side 20
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Simplex Method
Step 5: Test the feasibility (Variable leave the basis)
After identifying variable to become basic variable, one variable to be removed
from existing set of basic variable is to be determined.
For this each number in 𝑥𝐵 column is divided by the corresponding (positive)
number in key column and a row is selected for which this ratio is nonnegative
and minimum.
𝑥𝐵𝑟 𝑥𝐵𝑖
= min , 𝑎𝑖𝑗 > 0
𝑎𝑟𝑗 𝑎𝑖𝑗
It may be noted that division by zero or negative number is not permitted.
Note: If all element of key column is zero or negative solution will be unbounded.
The row selected in this manner is called key row or pivot row and it will
represent the variable which will leave the solution.
The element that lies at the intersection key row and key column is called key
element or pivot element.
16-10-2020 Side 21
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
Simplex Method
Step 6: (Find new solution)
(i) If the key element is 1, then row remain same in the new simplex table.
(ii) If key element is other than 1, then divide each element in the key row
(including the element in 𝑥𝐵 −column) by the key element to new values in
that row.
(iii) The new values of the elements in remaining rows of new simplex table can be
obtained by performing elementary rows operations on all rows so that all
element except key element in key column are zero.
The new entries in 𝐶𝐵 (coefficient of basic variable) and 𝑥𝐵 (value of basic variable)
columns are updated in new simplex table of current solution.
Step 7: (Repeat the procedure) Go to step 3 and repeat the procedure until all
entries in 𝑧𝑗 − 𝑐𝑗 row is either positive or zero.

16-10-2020 Side 22
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Simplex Method
𝑐𝑗 −1 3 −2 0 0 0

Basic variable 𝐶𝐵 𝑋𝐵 𝑥1 𝑥2 𝑥3 𝑠1 𝑠2 𝑠3 Min

in the basis Ratio
𝑠1 0 7 3 −1 3 1 0 0

𝑠2 0 12 −2 𝟒 0 0 1 0 3∗ Key
Row
𝑠3 0 10 −4 3 8 0 0 1 10/3

𝑧𝑗 − 𝑐𝑗 1 −3∗ 2 0 0 0

Key ColumnKey element

𝑹𝟐 𝟏 𝟑
𝑹𝟐 → , 𝑹𝟏 → 𝑹𝟏 + 𝑹𝟐 , 𝑹𝟑 → 𝑹𝟑 − 𝑹𝟐
𝟒 𝟒 𝟒

16-10-2020 Side 23
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Simplex Method

𝑐𝑗 −1 3 −2 0 0 0

Basic variable 𝐶𝐵 𝑋𝐵 𝑥1 𝑥2 𝑥3 𝑠1 𝑠2 𝑠3 Min

in the basis Ratio
𝑠1 0 10 𝟓/𝟐 0 3 1 1/4 0 20/5
Key
=4
Row
𝑥2 3 3 −1/2 𝟏 0 0 1/4 0 −

𝑠3 0 1 −5/2 0 8 0 −3/4 1 −

𝑧𝑗 − 𝑐𝑗 −1/2 0 2 0 3 0
4
Key Column Key element
𝟐 𝟏
𝑹𝟏 → 𝑹𝟏 , 𝑹𝟐 → 𝑹𝟐 + 𝑹𝟏 , 𝑹𝟑 → 𝑹𝟑 + 𝑹𝟏
𝟓 𝟓
16-10-2020 Side 24
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

FINAL TABLE OF SIMPLEX METHOD

𝑐𝑗 −1 3 −2 0 0 0

Basic variable 𝐶𝐵 𝑋𝐵 𝑥1 𝑥2 𝑥3 𝑠1 𝑠2 𝑠3 Min

in the basis Ratio
𝑥1 −1 4 1 0 6Τ5 2/5 1/10 0 _

𝑥2 3 5 0 1 3/5 1/5 3/10 0 _

𝑠3 0 11 0 0 11 1 −1/2 1 _

𝑧𝑗 − 𝑐𝑗 0 0 13/5 1/5 4/5 0

Since 𝑧𝑗 − 𝑐𝑗 ≥ for all varriabes.

Therefore optimal solution reached with
𝑥1 = 4, 𝑥2 = 5 , 𝑚𝑎𝑥 𝑧 = −4 + 15 = 11
Hence Min 𝑧 = −𝑧 ∗ = −11
16-10-2020 Side 25
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
Simplex Method For solving LPP

Example: Solve the following LP problem by simplex method

𝑀𝑎𝑥 𝑧 = 4𝑥1 + 5𝑥2
Subjected to constraints
2𝑥1 + 3𝑥2 ≤ 12,
2𝑥1 + 𝑥2 ≤ 8,
𝑥1 , 𝑥2 ≥ 0.
Solution: Write the given LPP in to standard form.
𝑀𝑎𝑥 𝑧 = 4𝑥1 + 5𝑥2 + 0. 𝑠1 + 0. 𝑠2
Subjected to constraints
2𝑥1 + 3𝑥2 + 𝑠1 = 12,
2𝑥1 + 𝑥2 + 𝑠2 = 8,
𝑥1 , 𝑥2 , 𝑠1 , 𝑠2 ≥ 0.
Here 𝑠1 and 𝑠2 are the slack variable.

16-10-2020 Side 26
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Simplex Method
𝑐𝑗 4 5 0 0

Basic variable in 𝐶𝐵 𝑋𝐵 𝑥1 𝑥2 𝑠1 𝑠2 Min

the basis Ratio
𝑠1 0 12 2 𝟑 1 0 12/3
Key
=4
Row
𝑠2 0 8 2 1 0 1 8/1
=8
𝑧𝑗 − 𝑐𝑗 −4 −5∗ 0 0

Key Column Key element

𝑹𝟏 𝟏
𝑹𝟏 → , 𝑹𝟐 → 𝑹𝟐 − 𝑹𝟏
𝟑 𝟑

16-10-2020 Side 27
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Simplex Method
𝑐𝑗 4 5 0 0

Basic variable in 𝐶𝐵 𝑋𝐵 𝑥1 𝑥2 𝑠1 𝑠2 Min

the basis Ratio
𝑥2 5 4 2/3 1 1/3 0 12/2
=6
𝑠2 0 4 𝟒/𝟑 0 −1/3 1 12/4 Key
=3 Row
𝑧𝑗 − 𝑐𝑗 −2/3 0 5/3 0

Key Column Key element

𝟑 𝟏
𝑹𝟐 → 𝑹𝟐 , 𝑹𝟏 → 𝑹𝟏 − 𝑹𝟐
𝟒 𝟐

16-10-2020 Side 28
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Simplex Method
𝑐𝑗 4 5 0 0

Basic variable in 𝐶𝐵 𝑋𝐵 𝑥1 𝑥2 𝑠1 𝑠2 Min

the basis Ratio
𝑥2 5 2 0 1 1/2 −1/2 −

𝑥1 4 3 1 0 −1/4 3/4 −

𝑧𝑗 − 𝑐𝑗 0 0 3/2 1/2

Since 𝑧𝑗 − 𝑐𝑗 ≥ for all varriabes.

Therefore optimal solution reached with
𝑥1 = 2, 𝑥2 = 3
Hence Max 𝑧 = 5 × 2 + 4 × 3 = 22.
16-10-2020 Side 29
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Dual Simplex Method for

Solving Linear
Programming Problem

16-10-2020 Side 1
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
Dual Simplex Method for Solving LPP
We use this method for solving LPP where initial solution is optimal but
infeasible ie this method is used when value of basic variables are negative and
optimality condition is satisfied. Hence
𝑧𝑗 − 𝑐𝑗 ≥ 0 (for optimal) and value of basic variable ≤ 0, for using this method.
Computational Procedure of Dual Simplex Algorithm :
Step 1: (1) Write all the constraint in the form of inequality of ≤ sign
(2) Now write the given LPP in to standard form.
To write the given LPP in to standard form,
(1) If problem is minimization type, convert it in to maximization type.
(2) Introduce slack variable in the constraint to reduce them into equality form.
Step 2: Find initial basic solution
To find the initial basic solution suppose all the given variable equal to zero and
calculate the slack variable which may not be feasible.

16-10-2020 Side 2
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Dual Simplex Method

Step 3: Construct the simplex table as usual in the simplex method.
Step 4: To test the initial solution optimality compute 𝑧𝑗 − 𝑐𝑗 = 𝐶𝐵 𝑥𝑖 − 𝑐𝑗 for
each 𝑥𝑖 column.
(1) If all the 𝑧𝑗 − 𝑐𝑗 ≥ 0 and 𝑥𝐵 𝑖 are non-negative , the solution is optimal
basic feasible solution.
(2) If all the 𝑧𝑗 − 𝑐𝑗 ≥ 0 and at least one 𝑥𝐵 𝑖 is negative, then proceed for the
next step.
(3) If any 𝑧𝑗 − 𝑐𝑗 < 0, the method is not possible.
Step 5: (1) Determine the outgoing vector: Row corresponding to most
negative basic variable is known as key row.

16-10-2020 Side 3
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
Dual Simplex Method

(2) Determine the incoming vector:

𝑧𝑗 − 𝑐𝑗
minimum ratio = −
−𝑣𝑒 value of key row
If all the element of key row is +𝑣𝑒 or zero then we shall get infeasible
solution. Column corresponding to minimum ratio is key column and
corresponding variable will be incoming variable.
Mark the element which is intersection of key row and key column. This
element is known as key element.
(3) Make key element unity, carry out the row operation as Regular simplex
method.
Step 6: If all the basic variable all positive then this solution is optimal basic
feasible solution but at least one basic variable is – 𝑣𝑒 then solution is not
optimal F.S. In this case repeat the step (4) and (5) till an optimal solution
achieved.

16-10-2020 Side 4
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Dual Simplex Method

Example: Solve the following LP problem by Dual Simplex method
𝑀𝑖𝑛 𝑧 = 3𝑥1 + 𝑥2
Subjected to constraints
𝑥1 + 𝑥2 ≥ 1,
2𝑥1 + 3𝑥2 ≥ 2, 𝑥1 , 𝑥2 ≥ 0.
Solution:
Step 1: Write the given LPP in to standard form.
𝑀𝑎𝑥 𝑧 ∗ = −𝑧 = −3𝑥1 − 𝑥2
Subjected to constraints
−𝑥1 − 𝑥2 ≤ −1,
−2𝑥1 − 3𝑥2 ≤ −2,
𝑥1 , 𝑥2 , ≥ 0.
Introduce slack variable 𝑠1 , 𝑠2 to reduce the constraint in to equality form
−𝑥1 − 𝑥2 + 𝑠1 = −1,
−2𝑥1 − 3𝑥2 + 𝑠2 = −2,
16-10-2020 Side 5
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur
Dual Simplex Method
Take 𝑥1 = 𝑥2 = 0, we get 𝑠1 = −1, 𝑠2 = −2, which is the initial basic
infeasible solution. Now simplex table is given by

𝑀𝑎𝑥 𝑧 ∗ = −𝑧 = −3𝑥1 − 𝑥2
−𝑥1 − 𝑥2 + 𝑠1 = −1,
−2𝑥1 − 3𝑥2 + 𝑠2 = −2,

16-10-2020 Side 6
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Dual Simplex Method

𝑐𝑗 −3 −1 0 0

Basic 𝐶𝐵 𝑋𝐵 𝑥1 𝑥2 𝑠1 𝑠2
variable in
the basis
𝑠1 0 −1 −1 −1 1 0
Key
𝑠2 0 −2 −2 −3 0 1
Row
𝑧𝑗 − 𝑐𝑗 3 1 0 0

𝑀𝑖𝑛𝑖𝑛𝑚𝑢𝑚 𝑅𝑎𝑡𝑖𝑜 3 3 1 1 0 0
− = − =
−2 2 −3 3

Key Column Key element

Here 𝑧𝑗 − 𝑐𝑗 ≥ 0 and XB ≤ 0. Hence initial basic feasible is infeasible optimal.
16-10-2020 Side 7
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Dual Simplex Method

Now since 𝑠2 = −2 is most – 𝑣𝑒 value. So it will be ongoing variable.
𝑧𝑗 −𝑐𝑗 3 1 3 1 1
Now 𝑚𝑖𝑛𝑖𝑚𝑢𝑚 𝑟𝑎𝑡𝑖𝑜 = − = − ,− = , = ,
−𝑣𝑒 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑘𝑒𝑦 𝑟𝑜𝑤 −2 −3 2 3 3
which is correspond to 𝑥2 column.
Hence now 𝑥2 will work as basic variable.
𝟏 𝟏
𝑹𝟐 → − 𝑹𝟐 , 𝑹𝟏 → 𝑹𝟏 − 𝑹𝟑
𝟑 𝟑

16-10-2020 Side 8
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Dual Simplex Method

𝑐𝑗 −3 −1 0 0

Basic variable in 𝐶𝐵 𝑋𝐵 𝑥1 𝑥2 𝑠1 𝑠2
the basis
𝑠1∗ 0 1 1 0 1 1 Key
− − −
3 3 3 Row
𝑥2 −1 2 2 1 0 1
−
3 3 3
𝑧𝑗 − 𝑐𝑗 7 0 0 1
3 3
Min ratio 7 − 0 1
Key element Key Column
𝑹𝟏 → −𝟑𝑹𝟏 , 𝑹𝟐 → 𝑹𝟐 − 𝑹𝟏
16-10-2020 Side 9
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Dual Simplex Method

𝑐𝑗 −3 −1 0 0

Basic variable in 𝑪𝑩 𝑿𝑩 𝒙𝟏 𝒙𝟐 𝒔𝟏 𝒔𝟐
the basis
𝑠2 0 1 1 0 −3 1

𝑥2 −1 1 1 1 −1 0

𝑧𝑗 − 𝑐𝑗 2 0 1 0

Since 𝑧𝑗 − 𝑐𝑗 ≥ 0 and 𝑋𝐵 ≥ 0. Hence this solution is optimal

solution. Hence 𝑥1 = 0, 𝑥2 = 1, Max 𝑧 = −3 × 0 − 1 × 1 =
− 1. So 𝑀𝑖𝑛 𝑧 = − −1 = 1

16-10-2020 Side 10
 Madan Mohan Malaviya Univ. of Technology, Gorakhpur

Assignment Work
Use dual simplex method to solve following problem:
1. Min 𝑧 = 𝑥1 + 2𝑥2 subject to 2𝑥1 + 𝑥2 ≥ 4, 𝑥1 + 7𝑥2 ≥ 0, 𝑥1 , 𝑥2 ≥ 0.
2. Min 𝑧 = 2𝑥1 + 𝑥2 subjected to 2𝑥1 + 𝑥2 ≥ 3, 4𝑥1 + 3𝑥2 ≥ 3, 𝑥1 , 𝑥2 ≥ 0.
Use simplex method to solve following problem.

16-10-2020 Side 11