1

OPERATIONS RESEARCH
INTRODUCTION
Operational research is today recognized as an applied science concerned with a large
number of diverse human activities. To be precise an operation uses some valuable
resources like men, money, machines, time, effort, etc. The outcome of the operation
has also some value. An operations research work is required:
1. To minimize the input value for a specific output.
2. To maximize the output value for a specific output.
3. Maximize some functions of values. Some of the areas of
management where techniques of operations research are applied
are listed below:
1. Finance Budgeting & Investment
2. Marketing
3. Physical Distribution
4. Purchasing, Procurement & Exploration
5. Personnel
6. Production
7. Independent float

APPLICATIONS OF OPERATIONAL RESEARCH

The application of operational Research in Managerial decision-making is as follows:
1. Finance budgeting & investment: Cash flow analysis, long-range capital
requirements, dividend policies, credit policies, credit risks.
2. Marketing: Product selection, timing, competitive actions. Advertising media
concerning cost & time, number of salesmen, and effectiveness of market research.
3. Physical Distribution: Location & size of warehouses, distribution center, retail
outlets, Distribution policy, etc.
4. Purchasing, procurement & exploration: Rules of buying, determining the quantity &
timing of purchase, bidding policies & vendor analysis, equipment replacement policies

SCOPE OF OPERATIONS RESEARCH

Scope of operational research is:
1. Industry: In the field of industrial management, there is a chain of problems
starting from the purchase of raw materials to the dispatch of finished goods
with a view to optimize profit hence operation research is necessary.
2. Organization: Operational research is equally applicable to big & small
organizations. Quantitative techniques have improved the productivity of an
organization’s operations.
 2

3. Business & society: Operational research is directly applicable to business &

society. Operational research methods can also be applied in hospitals &
clinics to reduce the waiting time of outdoor patients.

TOOLS OF OPERATIONS RESEARCH

 Waiting line Model: These models have been developed to establish a trade-
off between the costs of providing & the waiting time of a customer in the
questing system.
 Competitive (Game theory) Model: These models are used to characterize
the behavior of 2 or more opponents. (Called players) who complete for
achievement of conflicting goals.
 Network Models: These are applied to the management of large-scale
projects. It improves project co-ordination & enables the efficient use of
resources.
 Sequencing Models: The Sequencing problem arises whenever there is a
problem in determining the optimal sequence in which a series of tasks can be
performed by a number of service facilities.
 Replacement Models: These models are used when one must decide the
optimal solution time to replace equipment for one reason or the other.

TECHNIQUES OF OPERATIONS RESEARCH

Techniques of operational research:
 Allocation Models: They are used to allocate resources and activities in such
a way that some measure of effectiveness is optimized.
 Inventory Models: It deals with the problem of determination of how much to
order at a point in time & when to place an order.
 Waiting line Models: These models have been developed to establish a
tradeoff b/w the costs of providing & the waiting time of a customer in the
queuing system.
 Competitive Models: These models are used to characterize the behaviour
of 2 or more opponents who compete for the achievement of conflicting goals

LIMITATIONS OF OPERATIONS RESEARCH

Limitations of operations Research:
1. Dependence or an electronic computer.
2. Non - Quantifiable factors.
3. Wrong estimation.
4. Involve time & cost
 3

5. Implementation

MODELS OF OPERATIONS RESEARCH

Static & Dynamic model:
a. Static Models: Static models present a system at some specified time & do not
account for changes over time. For example, an inventory model can be
developed & solved to determine economic order quantity for the next period.
b. Dynamic models: In this model time is considered as one of the variables &
admits the impact of changes generated by time in the selection of the optimal
courses of action.
Descriptive models & prescriptive model:
a. Descriptive models: They simply describes some aspects of a situation based on
observation survey, questionnaire results (or) other available data of a situation & do
not predict (or) recommend
Ex: Organization chart.
b. Predictive models: These models are used to predict the outcomes due to a given
set of alternatives for a problem.
Analytical & simulation models:
a. Analytical models: These models have a specific mathematical
structure & thus can be solved by known Mathematical techniques
b. Simulation models: These models also have a mathematical structure
but are not solved by applying mathematical techniques to get a
solution.
Analog & Iconic models
a. Analog models: These models represent a system by a set of properties
different from that of the original system & do not resemble physically
b. Iconic models: An iconic model is either in an idealized form or a scaled version of
the system
 4

LINEAR PROGRAMMING
Introduction:
Linear programming is a quantitative technique that is used for decision-making, especially
in the context of uncertainty. It was introduced by Russian Mathematician George B Dantzig
in 1947 and later this technique was made generally available in 1951.
Mathematical programming involves and deals with optimization i.e., maximization or
minimization of a function either profit or cost which is called an Objective Function, which
is subject to certain constraints in the form of linear equalities and/or linear inequalities. It
involves the allocation of limited resources optimally based on a given criterion of optimality.
The name linear programming is composed of two components linear and programming.
Various authors defined the concept of linear programming in different ways, which are as
follows:
“Linear programming is a mathematical technique for determining the optimal solution of
resources obtaining a particular objective, where there are alternative uses of resources i.e.,
Man, Material, Machine, and Money, etc.” Galton
“Linear Programming is defined as programming of interdependent activities in a linear
structure”. Dantzig
“Linear programming is a problem-solving approach that has been developed to help
managers to make a decision, for example, A production schedule and an inventory policy to
be developed by a production manager, that will satisfy the demand forecasted for future
periods”.
LPP helps the production manager in estimating the future demand and scheduling
production capacity and inventory policy in relation to future demand
A financial analyst using LPP constructs a portfolio that maximizes the return from a variety
of investment alternatives in an effective manner
In all linear programming, the objective is common either maximization or minimization of
profits or sales or demand or revenue or cost or distance or time, or some quantity. In
continuation to the objective, all the LPPs are subject to some set of constraints, within which
the objective can be achieved. The constraints are
a. When it comes to meeting increased future demand by increasing the supply of
products the constraint here is the production capacity of the organization which
cannot be altered in a short run period.
b. Whereas in the case of portfolio construction, the constraints faced by the financial
manager are a limited number of available investment alternatives and the maximum
amount that can be invested in each investment instrument i.e., stock or a bond, etc.,
Linear programming deals with the optimization (maximization or minimization) of a
function of variables known as objective functions. It is subject to a set of linear equalities
and/or inequalities known as constraints. Linear programming is a mathematical technique
involving allocating limited resources in an optimal manner on the basis of a given criterion
of optimality.
 5

In all linear programming problems, the objective function is either maximization or

minimization of some quantity. To apply linear programming, certain requirements have to
be met. They are:
 Objective function should be identifiable and measurable in quantitative terms
 What activities to be included should be identified distinctly and measured in
quantitative terms.
 For the attainment of goals, what resources are to be allocated, should also be
identifiable and measurable quantitatively.
 The relationship of the interrelated variables is of the form y=a+bx, where x and y
are the variables of power one and a and b are constants.
 There should be a feasible alternative course of action available to the decision
maker, which is determined by the resource constraints.
The procedure for t h e mathematical formulation of an LP problem consists of the
following steps:
Step 1: Write down the decision variables of the problem.
Step 2: Formulate the objective function to be optimized (Maximized or Minimized) as a
linear function of the decision variables
Step 3: Formulate the other conditions of the problem such as resource limitation, market
constraints, and interrelations between variables as linear in equations or equations in terms
of the decision variables.
Step 4: Add non-negativity constraints, it implies that the production or performance of the
variables will never be negative. It may be either zero or greater than zero but never less than
zero.
x1, x2, x3, x4, x5,-------, xn ≥ 0
All these steps together form a Linear Programming Problem.
Decision Variables
(Controlled)
Objective Function

Constraints (Uncontrolled
Parameters)
The objective function, the set of constraints, and the non-negative constraint together form a
linear programming problem.
General Formulation of LPP: The general formulation of the LPP can be stated as follows:
In order to find the values of n decision variables x1x2 ... xn maximize or minimize the
objective function.
Max Z = a1x1+ a2x2+ a3x3+ a4x4+a5x5+-------+ anxn
a11x1+ a12x2+ a13x3+-------+ a1nxn (≤ or = or ≥ )b1
a21x1+ a22x2+ a23x3+-------+ a2nxn (≤ or = or ≥ ) b2
 6

a31x1+ a32x2+ a33x3+-------+ a3nxn (≤ or = or ≥ )b3

|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
am1x1+ am2x2+ am3x3+-------+ amnxn (≤ or = or ≥ )bm

x1, x2, x3, x4, x5,-------, xn ≥ 0

Where constraints may be in the form of inequality ≤ or ≥ or even in the form of an

equation (=) and finally satisfy the non-negative restrictions.
x1, x2, x3, x4, x5,-------, xn ≥ 0
Matrix Form of Linear Programming Problems:
Maximize or Minimize Z = Ax
Where A = a1, a2, a3, ----- an
X= x1, x2, x3, -----xn
Subject to the constraints Ax = B

[ ]
a 11 a 12 −−−−−−−−−a 1 n
a 21 a 22 −−−−−−−−−a 2 n
¿∨¿∨¿∨¿∨¿∨¿∨¿−−−−−−−−−¿∨¿∨¿ am1 am2 −−−−−−−−−−−am 3
x1 b1
x2 = or ≥ or ≤ b2
¿∨¿∨¿ xn ¿∨¿∨¿ bm
Note:
1. The constraints are of ‘≤’ type in the maximization problems
2. The constraints in the minimization problems are of the type ‘≥’. But a given problem
may contain a mix of the constraints involving the signs of ‘= or ≥ or ≤’
3. The decision variables are non-negative, mean it takes either positive or zero values.
But in extreme cases it may take positive values, zero values or even negative values.
These variables are called as unrestricted in sign or a free variable.

Types of LPP or Nature of LPP

1. Product Mix problems – assume a company manufactures two or more products
using the available resources. Using LPP the manager estimates which products to
manufacture and in what quantities within the constraints of available resources in
order to maximize total profits or sales or revenue of organization
2. Blending problems – assume a company manufactures a product XYZ to produce it
company requires different quantities of two ingredients A and B. Using LPP a
product manager determines optimal composition of available ingredients to form a
certain quantity of product within the restrictions of production capacity or
 7

availability of raw material i.e., under strict specifications. This optimal composition
of raw materials in manufacturing leads to minimum production cost. The concept of
blending problem is same as product mix but in product mix the objective is to
maximize the profit from the sale of manufactured products whereas in blending
problems the objective is to minimize the cost of manufacturing within the given
specifications.

Assumptions underlying Linear Programming:

1. All the data provided in LPP are known with certainty and don’t change during the
study period
2. Another assumption of LPP is proportionality, it means within the constraints the
resulting value of the objective function is proportional to the value of the decision
variable
For example – manufacturing one unit of product requires 10 units of ingredient
A and 5 units of ingredient B. Then manufacturing 10 units of product requires
100 units of A ingredient and 50 units of ingredient B raw material
3. Another property of LPP posse’s additively means total utilization of each resource is
nothing but the summation of a portion of resources required for the production of
each of the various products.
4. The assumption that all the decision variables are non-negative i.e., a negative number
of products have no meaning i.e., cannot be produced
5. All the decision variables are treated as continuous i.e., it can take any fractional
values
6. The objective function is linear i.e., per unit cost or profit assumed to be unaffected by
changes in production methods or quantity produced or quantity sold.

Steps in applying LPP

1. First, identify whether the problem can be solvable using linear programming
2. If yes before solving, check whether the problem is formulated in terms of a
mathematical model or not, if not convert the unstructured problem into the form of a
mathematical model
3. After the conversion, now solve the established model using mathematical techniques
i.e., linear programming

Note
The steps to be considered in forming a mathematical model are as follows –
i. Identify the objective and express it in the form of a mathematical equation
ii. Determine the factors that influence the objective function and express these in the
form of mathematical equations

Special Cases in Linear Programming

The solution to linear programming end at different stages
 8

a. Infeasibility – this condition arises in graphical solution where we cannot find an area
or a zone in the graph that satisfies all the constraints i.e., no feasible zone whereas in
simplex method an artificial variable occurs in the final iteration table beyond which
the solution cannot be proceed further
b. Unbounded solution – in graphical solution, it arises when the obtained feasible zone
is open ended instead of closed ended i.e., the value of Z can be increased infinitely in
case of maximization problems or decreased infinitely in case of minimization
problems without violating the constraints
c. Redundancy – this situation arises in graphical solution when a feasible zone is
identified i.e., an area satisfied by all the constraints, but one of the constraint does
not influence the feasible solution at all.
d. Multiple Solutions – it means an Lpp problem with two or more alternative solutions.
In graphical method this situation roses when iso-profit line is perfectly parallel to one
of the given constraints straight line. Whereas in case of simplex method, in final
iteration table when index row shows zero values for a variable that is not one the
basis of solution.
e. Optimal Solution- the point at which objective function Z is optimized i.e.,
maximization or minimization beyond which no further improvement is possible
DUALITY
Before converting the given LPP into duality, the first step is to represent the problem in
standard form it means all the decision variables in the problem should be of non-negative
and all the constraints should belong to ‘≤’ type if the problem is of maximization type and
vice versa., for minimization problems the constraints are of the type ‘≥’.
If any of the decision variables in LPP is unrestricted in sign, then that variable is expressed
as the difference between two non-negative variables. In this case their difference might be
either positive or negative or zero. To change the directions of the inequality from ‘≤’ to ‘≥’
or from ‘≥’ to ‘≤’, by just multiplying the entire constraint i.e., both left hand side and right
hand side of the constraint by ‘-1’. If any constraint in the primal LPP is of ‘=’ type then in
this case this constraint has to be represented twice by replacing the symbol ‘=’ by one
constraint with ‘≤ or ≥’ and the other constraint with ‘≥ or ≤’ type.
Example: x1 + x2 = 6
Replace the above constraint as follows:
x1 + x2 ≤ 6
x1 + x2 ≥ 6
The original formulation of the linear programming problem is called as primal problem.
Every primal problem has a comparison problem which is called as dual problem. The
optimal solution for the dual problem is same as the primal problem.
Rules for dual formulation:
 If the primal is of maximization problem, then the dual problem belong to
minimization problem and vice versa
 Coefficients of the objective function in primal becomes the right hand side constants
for the constraints in dual
 9

 Right hand side constants of the constraints in primal becomes the coefficients of the
objective function in the dual problem
 Column elements of the constraint in primal are the row elements for the constraints
in dual
 Rows of constraint coefficients in the primal becomes columns of constraints
coefficients in the dual
 Inequality signs of the constraints are reversed from primal to dual
 If a primal variable is non negative, the associated dual constraint should be greater
than or equal to type
 If the decision variable in primal is unrestricted in sign, then the associated dual
constraint should belong to an equation
 If primal constraint is an equation, then the associated dual variable is unrestricted in
sign.
 The dual of the dual is the primal problem

Relationship between primal and a dual problems

Primal Dual
Maximization problem Minimization problem
Number of variables Number of constraints
Minimization problem Maximization problem
Number or constraints Number of decision variable
≤∨≥ type of constraint Non negative decision variable
¿ type of constraint Unrestricted decision variable
Unrestricted decision variable = type of constraints
Objective function variable coefficient Right hand side constraint value
Right hand side constraint value Objective function decision variable
coefficients
Coefficient values of variable in first Coefficient value of first decision variable in
constraint all the constraint equations
i.e. 3x1 + 4x2 + 6x3 ≥ 10 a. 3 becomes coefficient of first
therefore the coefficients are 3, 4 and 6 variable in first constraint equation
b. 4 becomes coefficient of first
variable in second constraint
c. 6 becomes coefficient of first
variable in third constraint