OPERATIONS

RESEARCH

Visit Our Website www.iaritoppers.com for Agri.Study Material

COURSE OUTLINE
Lesson Name Page No
Module 1: Introduction to Operations Research
Lesson 1. Elementary concepts and objectives of Operations Research 1-4
Lesson 2. Applications of Operations Research in decision making 5-10

Module 2: Linear Programming Problem

Lesson 3. Mathematical formulation of the linear programming problem
and its graphical solution 11-22
Lesson 4. Simplex Method 23-32

Module 3: Transportation Problem

Lesson 5. Definition and mathematical formulation 33-36
Lesson 6. Initial basic feasible solution 37-44
Lesson 7. Optimal solution 45-54

Module 4: Assignment Problem

Lesson 8. Introduction and mathematical formulation 55-57
Lesson 9. Solution of Assignment problem 58-63

Module 5: Inventory Control

Lesson 10. Introduction and general notations 64-67
Lesson 11. Economic lot size models with known demand 68-74

Module 6: Replacement Theory

Lesson 12. Introduction and elementary concepts 75-76
Lesson 13. Replacement of items deteriorating with time 77-83

Module 7: Sequencing Problem

Lesson 14. Introduction and general notations 84-85
Lesson 15. Solution of a Sequencing problem 86-92

Module 8: Queuing Theory

Lesson 16. Introduction and classification of queues 93-96
Lesson 17. Solution of Queuing models 97-109

Module 9: Project Planning and Network Analysis

Lesson 18. Introduction and basic definitions in Network Analysis 110-113
Lesson 19. Rules for drawing Network Analysis 114-117
Lesson 20. Critical Path Method (CPM) 118-126
Visit Our Website www.iaritoppers.com for Agri.Study Material
Lesson 21. Project Evaluation and Review Technique (PERT) 127-135

Visit Our Website www.iaritoppers.com for Agri.Study Material

 Operations Research

Module 1. Introduction to Operations Research

Lesson 1
ELEMENTARY CONCEPTS AND OBJECTIVES OF OPERATIONS RESEARCH

Introduction

Operations Research (OR) is relatively a new discipline. The first formal activities of OR were initiated in
England during the Second World War, when a team of British scientists set out to make decisions regarding
the best utilization of war material. OR begins when some mathematical and quantitative technique is used to
verify the decision being taken. OR provides a quantitative technique or a scientific approach to the
executives for making better decisions for operations under their control.

Historical Background of Operations Research

The term Operations Research was first coined in 1940 by McClosky and Trefthen in a smal town Bowdsey
of U.K. This new science came into existence in military context. As the name implies, ‘Operations Research’
was apparently invented by the team dealing with research on (military) operations. The work of the team of
science was named as Operations Research in U.K. During the Second World War, military management
cal ed upon scientists from various disciplines, and organized them into teams to assist in solving strategic and
tactical problems i.e. to discuss, evolve and suggest ways and means to improve the execution of various
military projects. By their joint ef orts, experience and deliberations, they suggested certain approaches that
showed remarkable progress. This new approach to systematic and scientific study of the operations of the
system is cal ed the ‘Operations Research’ or ‘Operational Research’.

The encouraging results obtained by the British OR team quickly motivated the military management of the
United States to start on similar activities. The successful applications of the U.S. teams included the invention
of new flight patterns, planning sea mining and ef ective utilization of electronic equipment. This work of OR
team was given various names in the United States: Operational Analysis, Operations Evaluation, Operations
Research, Systems Analysis, Systems Evaluation, System Research and Management Science.

Fol owing the end of the war, the success of military teams attracted the attention of industrial managers who
were seeking solutions to their complex managerial type problems. The most common problem was to seek
methods so as to minimize the total cost and maximize the total profit. The first mathematical technique in the
field, cal ed the Simplex Method of linear programming, was developed in 1947 by an American
Mathematician George B. Dantzig. Since then, new techniques and applications have been developed through
the ef orts and cooperation of interested individuals in both academic institutions and industry.

In India, Operations Research came into existence in 1949 with the opening of an OR unit at the Regional
Research Laboratory at Hyderabad. At the same time, another group was set up in the Defence Science
Laboratory which devoted itself to the problems of stores, purchase and planning. In 1953, OR unit was
established in Indian Statistical Institute, Calcutta for the application of OR methods in national planning and
survey. OR Society of India was formed in 1955. In India, Prof. P. C. Mahalanobis made the first important
application of OR in formulating the Second Five Year Plan in order to forecast the trends of demand,

Visit Our Website www.iaritoppers.com for Agri.Study Material

 Operations Research

availability of resources and for scheduling the complex schemes necessary for developing the economy of the

Visit Our Website www.iaritoppers.com for Agri.Study Material

 Operations Research

country. In the industrial sector, in spite of the fact that opportunities of OR work at present are very much
limited, organized industries in India are gradual y becoming conscious of the role of Operations Research and
a good number of them have wel trained OR teams. Most popular practical application of OR in India is
linear programming.

Definitions of Operations Research

Literal y the word ‘operation’ may be defined as some action that we apply to some problems or hypotheses
and the word ‘Research’ means an organized process of seeking out facts about the same. In fact, it is
dif icult to precisely define OR mainly because of the fact that its boundaries are not clearly marked. OR has
been variously described as ‘Science of Use’, ‘Quantitative Common Sense’, ‘Scientific approach to
decision making problems’, etc. Some of the commonly used and widely accepted definitions of OR are
given as under:

1. “OR is a scientific method of providing executive departments with a quantitative basis for decisions
under their control”.

- P.M. Morse and G.E. Kimbal

2. “OR is the application of scientific methods, techniques and tools to problems involving the operations
of a system so as to provide those in control of the system with optimum solutions to the problems”.

- Churchman, Ackof and Arnof

3. “OR is applied decision theory. It uses any scientific, mathematical or logical means to attempt to
cope with the problems that confront the executive, when he tries to achieve a thorough going
rationality in dealing with his decision problems”.

- D.W. Mil er and M.K. Starr

4. “OR is the attack of modern science on problems of likelihood that arise in the management and
control of men and machines, materials and money in their natural environment, its special technique is
to invent a strategy of control by measuring, comparing and predicting probable behavior through a
scientific model of a situation”.

- Beerr

5. “OR is a scientific method of providing the executive with an analytical and objective basis for
decisions”.

- P.M.S. Blackett

6. The term OR connotes various attempts to study operations of war by scientific methods. From a
more general point of view OR can be considered to be an attempt to study these operations of
modern society which involve organizations of men or of men and machines.

- P.M. Morse

Visit Our Website www.iaritoppers.com for Agri.Study Material

Operations Research
7. OR is a management activity pursued in two complementary ways – one half by the free and bold

Visit Our Website www.iaritoppers.com for Agri.Study Material

 Operations Research

exercise of commonsense untrammeled by any routine and other half by the application of a
repertoire of wel established pre-created methods and techniques.

- Jagjit Singh

8. OR is the attack of modern science on complex problems arising in the direction and management of
large systems of men, machines, materials and money in industry, business and defence. The
distinctive approach is to develop a scientific model of the system, incorporating measurements of
factors such as chance and risk with which to predict and compare the outcomes of alternative
decisions, strategies or controls. The purpose is to help management to determine its policies and
actions scientifical y.

- Operational Research Quarterly

9. “OR is the art of giving bad answers to problems which otherwise have worse answers”.

- T.L. Saaty

From al the above opinions, it may be concluded that whatever else OR may be, it is certainly concerned
with optimization theory.

Scope of Operations Research

OR is mainly concerned with the techniques of applying scientific knowledge, besides the development of
science. It provides an understanding which gives the expert/manager new insights and capabilities to
determine better solutions in his decision making problems, with great speed, competence and confidence.
OR has been found to be used in the fol owing five major areas of research:

1. OR is useful to the Directing Authority in deciding optimum al ocation of various limited resources
such as men, machines, material, time, money, etc., for achieving the optimum goal.

2. OR is useful to Production Specialist in

i. Designing, selecting and locating sites.

i. Determining the number and size.

ii. . Scheduling and sequencing the production runs by proper al ocation of machines; and

iv. Calculating the optimum product mix.

3. OR is useful to the Marketing Manager (executive) in determining:

i. How to buy, how often to buy, when to buy and what to buy at the minimum possible cost.

i. Distribution points to sel the products and the choice of the customers.

ii. . Minimum per unit sale price.

iv. The customer’s preference relating to the size, colour, packaging etc., for various products
and the size of the stock to meet the future demand; and

Visit Our Website www.iaritoppers.com for Agri.Study Material

Operations Research
v. The choice of dif erent media of advertising.

Visit Our Website www.iaritoppers.com for Agri.Study Material

Operations Research

4. OR is useful to the Personnel Administrator in finding out:

i. Skil ed persons at a minimum cost.

i. . The number of persons to be maintained on ful time basis in a variable work load like
freight handling etc.; and

ii. . The optimum manner of sequencing personnel to a variety of jobs.

5. OR is useful to the Financial Control er to

i. Find out a profit plan for the company.

i. . Determine the optimum replacement policies.

ii. . Find out the long-range capital requirements as wel as the ways and means to
generate these requirements.

Visit Our Website www.iaritoppers.com for Agri.Study Material

 Operations Research

Module 1. Introduction to Operations Research

Lesson 2
APPLICATIONS OF OPERATIONS RESEARCH IN DECISION MAKING

Introduction

The Operations Research may be regarded as a tool which is utilized to increase the ef ectiveness of
management decisions. Scientific method of OR is used to understand and describe the phenomena of
operating system. Mathematical and logical means of Operations Research provides the executive with
quantitative basis for decision making and enhance ability to make long range plans and to solve everyday
problems of industry with greater ef iciency and competence.

Features (Characteristics) of OR

OR is a tool employed to increase the ef ectiveness of managerial decisions as an objective supplement to the
subjective feeling of the decision makers. There are five salient features of OR.

Decision making

Primarily OR is addressed to managerial decision making or problem solving.

Scientific approach

OR employs scientific methods for the purpose of solving problems. It is a formalized process of reasoning.

Objective

OR attempts to locate the best or optimal solution to the problem under consideration. For this purpose, it is
necessary that a measures of ef ectiveness is defined which is based on the goals of the organization. This
measure is then used as the basis to compare the alternative courses of action. The examples are profit, net
returns, cost of production etc.

Inter-disciplinary team approach

OR is inter-disciplinary in nature and requires a team approach to a solution of the problem. Managerial
problems have economic, physical, psychological, biological, sociological and engineering aspects. This
requires a blend of people with expertise in the areas of Mathematics, Statistics, Engineering, Economics,
Management, Computer Science and so on.

Digital computer

Use of a digital computer has become an integral part of the OR approach to decision making. The computer
may be required due to the complexity of the model, volume of data required and the computations to be
made.

Modeling in Operations Research

A model as used in OR is defined as idealized representation of a real-life system. It shows the relationships
(direct or indirect) and inter-relationships of action and reaction in terms of cause and ef ect. A model in OR
is a simplified representation of an operation or a process in which only the basic aspects or the most
Visit Our Website www.iaritoppers.com for Agri.Study Material
 Operations Research

important features of a typical problem under investigation are considered. The construction of a model helps
in putting the complexities and possible uncertainties in a decision making problem, into a logical framework
amenable to comprehensive analysis. With such a model we can have several decision alternatives, their
anticipated ef ects indicating the need for relevant data for analyzing the alternatives that leads to informative
conclusions.Models can be broadly classified according to fol owing characteristics:

Classification by structure

Iconic models

These models are pictorial representation of real systems and have the appearance of the real thing. They
represent the system as it is by scaling it up or down (i.e., by enlarging or reducing the size). In other words it
is an image. Examples of such models are child’s toy, photograph, a schedule of operations, histogram,
pictogram, cartogram, a physical model such as smal scale model of a dairy plant, an engine etc. These kinds
of models are cal ed ‘Iconic’ because they are look alike items to understand and interpret the real things. An
iconic model is said to be ‘scaled down’ or ‘scaled up’ according to the dimensions of the model and are
smal er or greater than those of the real item. Iconic models are easy to observe, build and describe, but are
dif icult to manipulate and not very useful for the purpose of prediction. Commonly these models represent a
static event. These models can be constructed up to three dimensions (e.g., atom, globe, smal aeroplane,
cube, etc.) while it is not possible to construct them physical y for higher dimensions. Further these models do
not include those aspects of the real system that are irrelevant for the analysis.

Analogue models

Analogue models

These models are more abstract than the iconic ones as there is no ‘look alike’ correspondence between
these models and real items. These models are the one in which one set of properties is used to represent
another set of properties. After the problem is solved, the solution is re-interpreted in terms of the original
system. For example, graphs and maps in various colours are analogue models in which dif erent colours
correspond to dif erent characteristics e.g., blue representing water, brown representing land, yel ow
representing production etc. Further, graphs are very simple analogues because distance is used to represent
such properties as time, number, per cent, age, weight, and many other properties. Contour lines on the map
represent the rise and fal of the heights. Demand curves, flow charts in production control and frequency
curves in statistics are analogue models of the behaviour of events. Graphs of time series, bar diagrams, stock
market changes are other examples of analogue models. Analogue models are less specific, less concrete but
easier to manipulate and can represent dynamic situations. These are, general y, more useful than the iconic
ones because of their vast capacity to represent the characteristics of the real system under investigation.

Mathematical models (Symbolic Model)

The Symbolic or mathematical model is one which employs a set of mathematical symbols (i.e. letters,
numbers etc.) to represent the decision variables of the system. These variables are related together by means
of a mathematical equation or set of equations to describe the behavior (or properties) of the system. These
models are most general and precise. The solution to these models is then obtained by applying wel
Visit Our Website www.iaritoppers.com for Agri.Study Material
 Operations Research
developed mathematical techniques to the model. The symbolic model is usual y the easiest to manipulate

Visit Our Website www.iaritoppers.com for Agri.Study Material

 Operations Research

experimental y and the most general and abstract. Its function is more often explanatory rather than
descriptive. Explanatory model is that which contains control ed variables, while the descriptive model does
not contain controled variables.

Classification by purpose

Descriptive model

It describes the some aspects of a situation based on observation, survey, questionnaire results or other
available data.

Predictive models

Such models can answer ‘what if’ type of questions i.e. they can make predictions regarding certain events.

Prescriptive models

When a predictive model has been repeatedly successful, then it can be used to prescribe a source of action.
For example, linear programming is a prescriptive model because it prescribes what the managers ought to
do.

Classification by nature of environment

Deterministic model

Such models assume conditions of complete certainty and perfect knowledge for example; linear
programming, transportation and assignment models are deterministic models.

Probabilistic (or Stochastic) model

These models handle those situations in which consequences or payof of managerial actions cannot be
predicted with certainty.

Classification by Behavior

Static models

These models do not consider the impact of changes that take place during the planning horizon i.e. they are
independent of time.

Dynamic models

These models consider time as one of the important variables and depict the impact of changes generated
over time. In this instead of one decision, a series of independent decisions are required during the planning
horizons.

Characteristics of a Good Model

 A good model should be capable of taking into account new formulations without having any
significant change in its frame.

 Assumptions made in the model should be as few as possible.

Visit Our Website www.iaritoppers.com for Agri.Study Material

Operations Research
 It should be simple and coherent having less number of variables.

Visit Our Website www.iaritoppers.com for Agri.Study Material

 Operations Research

 It should be open to parametric type treatments.

Advantages of a Model

 Through the model, the problem under consideration becomes control able.

 It provides some logical and systematic approach to the problem.

 It indicates the limitations and scope of an activity

 Models help in incorporating useful tools that eliminate duplication of methods applied to solve any
specific problem.

 Models help in finding avenues for new research and improvements in a system.

Methodology of Operations
Research The major phases of an O.R.
study are as fol ows:

Phase I: Formulation of the Problem

In this phase, the problem is formulated in an appropriate form. This phase should give a statement of the
problem’s elements that include the control able (decision) variables, the uncontrol able parameters, the
restrictions or constraints on the variables and the objectives for defining a good or improved solution.

Phase II: Constructing a mathematical model

The second phase of the investigation is concerned with the choice of proper data inputs and the design of the
appropriate information output. In this phase, both static and dynamic structural elements and the
representation of inter-relationship among the elements in terms of mathematical formulae need to be
specified. A mathematical model should include mainly the fol owing three basic sets of elements:
(i) Decision Variables and Parameters

(i ) Constraints or Restrictions

(ii) ) Objective Function

Phase III: Deriving the solution from the model

This phase of the study deals with the mathematical calculations for obtaining the solution to the model. A
solution of the model means those values of the decision variables that optimize one of the objectives and give
permissible levels of performance on any other of the objectives.

Phase IV: Testing the model and its solution

This phase of the study involves checking the validity of the model used. A model may be said to be valid if it
can give a reliable prediction of the system’s performance.

Phase V: Controlling the solution

Visit Our Website www.iaritoppers.com for Agri.Study Material

 Operations Research
This phase of the study establishes control over the solution by proper feedback of the information on
variables which deviated significantly. As soon as one or more of the control ed variables change significantly,

Visit Our Website www.iaritoppers.com for Agri.Study Material

 Operations Research

the solution goes out of control. In such a situation the model may accordingly be modified.

Phase VI: Implementing the solution

This phase of the study deals with the implementation of the tested results of the model. This would basical y
involve a careful explanation of the solution to be adopted and its relationship with the operating realities.

General Methods of Deriving the Solution

There are three methods to derive the solution to an OR model.

Analytical method

Analytical methods involve expressions of the model by graphic solutions or by mathematical computations.
For example, area indicated by mathematical function may be evaluated through the use of integral calculus.
Solutions of various inventory models are obtained by using the analytical procedure. The analytical methods
involve al the tools of classical mathematics, such as calculus, finite dif erences, etc. The kind of mathematics
required for a particular OR study depends upon the nature of the model.

Numerical or iterative method

Numerical methods are concerned with the iterative or trial and error procedures, through the use of
numerical computations at each step. These numerical methods are used when some analytical methods fail to
derive the solution. The algorithm is started with a trial (initial) solution and continued with the set procedure
to improve the solution towards optimality. The trial solution is then replaced by the improved one and the
process is repeated until the solution converges or no further improvement is possible. Thus the numerical
methods are hit and trial methods that end at a certain step after which no further improvement to the solution
is possible.

Monte Carlo method

This method is essential y a simulation technique, in which statistical distribution functions are created by
generating a series of random numbers. Various steps associated with a Monte Carlo method are:

(a) For appropriate model of the system, sample observations are made and then the probability
distributions for the variables of interest are determined.

(b) Convert the probability distribution to a cumulative distribution.

(c) Select the sequence of random numbers with the help of random number tables.

(d) Determine the sequence of values of variables of interest with the sequence of random numbers
obtained in the previous step

(e) Fit an appropriate mathematical function to the values obtained in step (d)

Advantages/ Merits of OR Techniques

Optimum use of production factors

Linear programming techniques indicate how a manager can utilize most ef ectively his inputs/ factors and by
Visit Our Website www.iaritoppers.com for Agri.Study Material
 Operations Research

more ef iciently selecting and distributing these elements.

Improved quality of decision

The ef ect on the profitability due to changes in the production pattern wil be clearly indicated in the simplex
table. These tables give a clear picture of the happenings within the basic restrictions and possibilities of
behaviour of compound elements involved in the problem.

Preparation of future managers

These methods substitute a means for improving the knowledge and skil of young managers.

Modification of mathematical solution

OR presents a possible practical solution when one exists, but it is always a responsibility of the manager to
accept or modify the solution before its use. The ef ect of these modifications may be evaluated from the
computation steps and tables.

Alternative solutions

OR techniques suggest al the alternative solutions available for the same profit so that the management may
decide on the basis of its strategies.

Limitations of OR

Practical application

Formulation of an industrial problem to an OR set programme is a dif icult task.

Reliability of the proposed solution

A non-linear relationship is changed to linear for fitting the problem to linear programming. This may disturb
the solution.

Money and time cost

When the basic data is subject to frequent changes, the cost of changing programmes manual y is a costly
af air.

Combining two or more objective functions

Very frequently maximum profit does not come from manufacturing the maximum quantity of the most
profitable product at the most convenient machine and at the minimum cost, since this may lead to
underutilization of certain lines of production.

The aim is not to optimize individual objective function. It is, therefore, necessary to have a single objective
function which can cover several objective functions at the same time. Despite al the above limitations, OR is
a powerful tool and an analytical process that of ers the presentation of an optimal solution.

Visit Our Website www.iaritoppers.com for Agri.Study Material

 Operations Research

Module 2. Linear programming problem

Lesson 3
MATHEMATICAL FORMULATION OF THE LINEAR PROGRAMMING PROBLEM AND
ITS GRAPHICAL SOLUTION

Introduction

A large number of business and economic situations are concerned with problems of planning and al ocation
of resources to various activities. In each case there are limited resources at our disposal and our problem is
to make such a use of these resources so as to maximize production or to derive the maximum profit, or to
minimize the cost of production etc. Such problems are referred to as the problems of constrained
optimization. Linear programming (LP) is one of the most versatile, popular and widely used quantitative
techniques. Linear Programming is a technique for determining an optimum schedule chosen from a large
number of possible decisions. The technique is applicable to problem characterized by the presence of a
number of decision variables, each of which can assume values within a certain range and af ect their decision
variables. The variables represent some physical or economic quantities which are of interest to the decision
maker and whose domain are governed by a number of practical limitations or constraints which may be due
to availability of resources like men, machine, material or money or may be due quality constraint or may arise
from a variety of other reasons. The most important feature of linear programming is presence of linearity in
the problem. The word Linear stands for indicating that al relationships involved in a particular problem are
linear. Programming is just another word for “planning” and refers to the process of determining a particular
plan of action from amongst several alternatives. The problem thus reduces to maximizing or minimizing a
linear function subject to a number of linear
near inequalities

Definitions of Various Terms Involved in Linear Programming

Linear

The word linear is used to describe the relationship among two or more variables which are directly
proportional. For example, if the production of a product is proportionately increased, the profit also increase
proportionately, then it is a linear relationship. A linear form is meant a mathematical expression of the type,

Programming

The term “Programming” refers to planning of activities in a manner that achieves some optimal result with
resource restrictions. A programme is optimal if it maximizes or minimizes some measure or criterion of
ef ectiveness, such as profit, cost or sales.

Decision variables and their relationship

The decision (activity) variables refer to candidates (products, services, projects etc.) that are competing with
 Operations Research

one another for sharing the given limited resources. These variables are usual y inter-related in terms of

Visit Our Website www.iaritoppers.com for Agri.Study Material

 Operations Research

utilisation of resources and need simultaneous solutions. The relationship among these variables should be
linear.

Objective function

The Linear Programming problem must have a wel defined objective function for optimization. For example,
maximization of profits or minimization of costs or total elapsed time of the system being studied. It should be
expressed as linear function of decisionvariables.

Constraints

There are always limitations on the resources which are to be al

a ocated among various competing activities.
These resources may be production capacity, manpower, time, space or machinery. These must be capable
of being expressed as linear equalities or inequalities in terms of decision variables.

Alternative courses of action

There must be alternative courses of action. For example, there may be many processes open to a firm for
producing a commodity and one process can be substituted for another.

Non-negativity
Non negativity restriction

Al the variables must assume non-negative

negative values, that is, al variables must take on values equal to or greater
than zero. Therefore, the problem should not result in negative values for the variables.

Linearity and divisibility

Al relationships (objective functions and constraints) must exhibit linearity, that is, relationships among
decision variables must be directly proportional. For example, if our resources increase by some percentage,
then it should increase the outcome by the same percentage. Divisibility means that the variables are not
limited to integers. It is assumed that decision variables are continuous, i.e., fractional values of these variables
must be permissible in obtaining an optimal solution.

Deterministic

In LP model (objective functions and constraints), it is assumed that the entire model coef icients are
completely known (deterministic), e.g. profit per unit of each product, and amount of resources available are
assumed to be fixed during the planning period.

Formulation of a Linear Programming Problem

The formulation of the Linear Programming Problem (LPP) as mathematical model involves the fol owing key
steps:

Step 1. Identify the decision variables to be determined and express them in terms of algebraic symbols as
.

Step 2. Identify the objective which is to be optimized (maximized or minimized) and express it as a linear
function of the above defined decision variables.
 Operations Research
Step 3. Identify al the constraints in the given problem and then express them as linear equations or

Visit Our Website www.iaritoppers.com for Agri.Study Material

 Operations Research

inequalities in terms of above defined decision variables.

Step 4. Non-negativity restrictions on decision variables.

The formulation of a linear programming problem can be ilustrated through the fol owing examples:

Example 1:

A milk plant manufactures two types of products A and B and sel s them at a profit of Rs. 5 on type A and
Rs. 3 on type B. Each product is processed on two machines G and H. Type A requires one minute of
processing time on G and two minutes on H; type B requires one minute on G and one minute on H. The
machine G is available for not more than 6 hours 40 minutes, while machine B is available for 8 hours 20
minutes during any working day; formulate the problem as LP problem.

Solution :

Let X1 be the number of products of type A and X2 the number of products of type B. Since the profit on
type A is Rs. 5 per product, 5X1 wil be the profit on sel ing X1units of type A. Similarly, 3X2 wil be the
profit on sel ing X2 units of type B. Therefore, total profit on sel ing X1 units of A and X2 units of B is given by
Z = 5X1+3X2. Since this has to be maximized, hence the objective function can be expressed as

Max Z = 5X1+3X2 (Objective function)

Since machine G takes 1 minute time on type A and 1 minute time on type B, the total time in minutes
required on machine G is given by X1+X2. Similarly the total time in minutes required on machine H is given
by 2X1+X2. But machine G is not available for more than 6 hour 40 minutes (400 minutes), therefore
X1+X2 ≤ 400 (first constraint on machine G)
Also the machine H is available for 8 hours 20 minutes only, therefore
2X1+X2 ≤ 500 (second constraint on machine H)
Since it is not possible to produce negative quantities,
X1 ≥0, X2 ≥0 (non-negativity restrictions)

Example 2 :

A dairy plant packs two types of milk in pouches viz., ful cream and single toned. There are suf icient
ingredients to make 20,000 pouches of ful cream & 40,000 pouches of single toned. But there are only
45,000 pouches into which either of the products can be put. Further it takes three hours to prepare enough
material to fil 1000 pouches of ful cream milk & one hour for 1000 pouches of single toned milk and there
are 66 hours available for this operation. Profit is Rs. 8 per pouch for ful cream milk and Rs. 7 per pouch for
single toned milk. Formulate it as a linear programming problem.

Solution

Let number of ful cream and single toned milk pouches to be packed is X 1 and X2.

Let profit be Z

Visit Our Website www.iaritoppers.com for Agri.Study Material

 Operations Research

Objective function: Max. Z = 8X1+7X2

Subject to constraints: X1 + X2 45000 (1)

X1 20000 (2)

X2 40000 (3)

(4)

Non negativity restrictions

Non-negativity X1, X2 0

Example 3 :

Consider two diff erent types of food stuf

stu s say F1 and F2. Assume that these food stuf
stu s contain vitamin A
and B. Minimum daily requirements of vitamin A and B are 40mg and 50mg respectively. Suppose food stuf
F1 contains 2mg of vitamin A and 5mg of vitamin B while F2 contains 4mg of vitamin A and 2mg of vitamin B.
Cost per unit of F1 is Rs. 3 and that of F2 is Rs. 2.5. Formulate the minimum cost diet that would supply the
body at least the minimum requirements of each vitamin.

Solution :

Let number of units needed for food stuf s F1 and F2 to meet the daily requirements of vitamins A and B be
respectively X1 and X2..
Objective function: Minimize Z =3X1+2.5X2
Subject to constraints:
2X1+4X2 ≥ 40
5X1+2X2 ≥50
Non-negativity
negativity restrictions X1, X2 0

Mathematical Formulation of a General Linear Programming Problem

The general formulation of LP problem can be stated as fol ows. If we have n-decision
decision variables X1, X2,…,
Xn and m constraints in the problem , then we would have the fol
fo owing type of mathematical formulation of
LP problem

Optimize (Maximize or Minimize) the objective function:

(3.4.1)
subject to satisfaction of m
m- constraints :
 Operations Research

Where the constraint may be in the form of an inequality (≤ or ≥) or even in the form of an equality (=) and
final y satisfy the non-negativity
negativity restrictions

X1, X2, … , Xn ≥ 0 (3.4.3)

where Cj (j=1, 2,….,n); bi (i=1, 2,…,m) and aij are al constants and m<n, and the decision variables Xj ≥0,
j=1, 2,…,n. If bi is the available amount of resource i then aij is amount of resource i that must be aal ocated
(technical coef icient) to each unit of activity j.
NOTE: By convention, the values of RHS parameters bi (i=1, 2, 3,…,m) are restricted to non
non-negative
negative
values only. If any value of bi is negative then it is to be changed to a positive value by multiplying both sides
of the constraint by -1. This not only changes the sign of al LHS Coef icients and of RHS parameters but
also changes the direction of inequality sign.
3.4.1 Matrix Form of LP problem
The linear programming problem (3.4.1), (3.4.2) and (3.4.3) can be expressed in matrix form as
Maximize
(objective function )
subject to
(constraint equation )
(non-negativity
negativity restriction )
where ),
) and )

Graphical Solution of Linear Programming Problem

LP problems which involve only two variables can be solved graphical

graphica y. Such a solution by geometric
method involving only two variables is important as it gives insight into more general case with any number of
variables.

Feasible solution

A set values of the variables of a linear programming problem which satisfies the set of constraints and the
non--negative restrictions is cal
ca ed feasible solution of the problem.

Feasible region

The col
co ection of al
a feasible solutions is known as the feasible region. Any point which does not lie in the
feasible region cannot be a feasible ssolution
olution to the LP problems. The feasible region does not depend on the
 Operations Research

form of the objective function in any way. If we can represent the relations of the general LP problem on an n
dimensional space, we wil obtain a shaded solid figure (known as Convex-polyhedron) representing the
domain of the feasible solution.

Optimal solution

A feasible solution of a linear programming problem which optimizes its objective function is cal ed the
optimal solution of the problem. Theoretical y, it can be shown that objective function of a LP problem
assumes its optimal value at one of the vertices (cal ed extreme points) of this solid figure.

Steps to find graphical solution of the linear programming problem

Step 1: Formulate the linear programming problem.

Step 2: Draw the constraint equations on XY-plane.
Step 3: Identify the feasible region which satisfies al the constraints simultaneously. For less than or equal to
constraints the region is general y below the lines and for greater than or equal to constraints, the
region is above the lines.
Step 4: Locate the solution points on the feasible region. These points always occur at the vertices of the
feasible region.
Step 5: Evaluate the optimum value of the objective function.

The geometric interpretation and solution for the LP problem using graphical method is il ustrated with the
help of fol owing examples

Example 4 Find the graphical solution of problem formulated in example 1.

Max Z = 5X1+3X2
Subject to
X1+X2 ≤ 400
2X1+X2 ≤ 500
X1 ≥0 , X2 ≥0

Solution

Any point lying in the first quadrant has X1, X2 ≥ 0 and hence satisfies the non-negativity restrictions.
Therefore, any point which is a feasible solution must lie in the first quadrant. In order to find the set of points
in the first quadrant which satisfy the constraints, we must interpret geometrical y inequalities such as X1+X2≤
400. If equality holds then we have the equation X1+X2= 400 i.e., any point on the straight line satisfies the
equation. Any point lying on or below the line X1+X2= 400 satisfies the constraint X1+X2≤ 400. However,
no point lying above the line satisfies the inequality. In a similar manner, we can find the set of points satisfying
2X1+X2 ≤ 500 and the non-negativity restrictions are al the points in the first quadrant. The set of points
satisfying the constraints and the non-negativity restrictions is the set of points in the shaded region of the
figure OABC as shown in Fig 3.1 Any point in this region is a feasible solution, and only the points in this
region are feasible solutions. To solve the LP problem, we must find the point or points in the region of

Visit Our Website www.iaritoppers.com for Agri.Study Material

 Operations Research

feasible solutions which give the largest value of the objective function. Now for any fixed value of Z, Z =

Visit Our Website www.iaritoppers.com for Agri.Study Material

 Operations Research

5X1+3X2 is a straight line, for each dif erent values of Z, we obtain a dif erent line. Again, al the lines
corresponding to dif erent values of Z are paral el; clearly our interest is to find the line with the largest value
of Z which has at least one point in common with the region of feasible solutions. The line Z= 5X1+3X2 is
drawn for various values of Z. It can be seen that the point farthest from Z = 5X1+3X2 and yet in the feasible
region is B (100, 300) and thus point B maximizes Z while satisfying al the constraints. In other words, the
corner B of the region of feasible solutions is the optimal solution of the LP problem. Since B is the intersection
of the lines X1+X2=400 and 2X1+X2=500, it can be seen that at vertex B, X1=100 and X2=300. The
maximum value of Z = 5(100)+3(300)=1400.The fact that the optimum occurred at the vertex B of the
feasible region is not a coincidence but on the other hand represents a significant property of optimal solutions of
al LP problems. To find the optimal solution find the values of objective function at the various extreme points as
shown in the fol owing Table 3.1

Table 3.1 Computation of maximum value of objective function

Extreme Point Coordinates Profit Function Z = 5X1+3X2
O X1=0 , X2=0 Z=5(0)+3(0)=0
A X1=0 , X2=400 Z=5(0)+3(400)=1200
B X1=100, X2=300 Z=5(100)+3(300)=1400
C X1=250, X2=0 Z=5(250)+3(0)=1250

Note: A fundamental theorem in LP states that the feasible region of any LP is a convex polygon (that is the n
dimensional version of two dimensional polygon), with a finite number of vertices, and further for any LP
problem, there is at least one vertex which provides an optimal solution. Whenever a LP problem has more
than one optimal solution, we say that there are alternative optimal solutions. Physical y, this means that the
resources can be combined in more than one way to maximize profit.

Visit Our Website www.iaritoppers.com for Agri.Study Material

 Operations Research

Fig. 3.1 Feasible region

Example 5: Find the graphical solution of problem formulated in Example 2.

Solution :

Let number of fu
ful cream and single toned milk pouches to be produced is X 1and X2 and Let profit be Z

Objective function: Max. Z = 8X1+7X2

Subject to constraints: X1 + X2 45000 (1)

X1 20000 (2)

X2 40000 (3)

3X1 + X2 66000 (4)

Non negativity restrictions

Non-negativity X1, X2 0

First of al draw the graphs of these inequalities (as discussed in Example 5) which is shown in Fig. 3.2.
 Operations Research

Fig. 3.2 Feasible region

To find the optimal solution find the values of objective function at the various extreme points as shown in the
fol owing Table 3.2

Table 3.2 Computation of maximum value of objective function

Extreme Point Coordinates Profit Function Z = 8X1+7X2
O X1=0 , X2=0 Z=8(0)+7(0)=0
A X1=0 , X2=40000 Z=8(0)+7(40000)=280000
B X1=5000, X2=40000 Z=8(5000)+7(40000)=320000
C X1=10500, X2=34500 Z=8(10500)+7(34500)=325500
D X1=20000 , X2=6000 Z=8(20000)+7(6000)=202000
E X1=20000 , X2=0 Z=8(20000)+7(0)=160000

So maximum value of Z occurs at point C (10500, 34500) so it is the optimal solution. It can be concluded
that Dairy Plant must produce 10500 pouches of ful cream milk and 34500 pouches of single toned milk.
Let us consider the fol owing example to il ustrate graphical solution for minimization problem

Special cases in Linear Programming

Visit Our Website www.iaritoppers.com for Agri.Study Material

 Operations Research
Up til now we have discussed those problems which have a unique optimal solution. However, it is possible

Visit Our Website www.iaritoppers.com for Agri.Study Material

 Operations Research

for LP problem to have fo

fol owing special cases.
3.5.4.1
.4.1 Unbounded solution: A linear programming problem is considered to have an unbounded solution
if it has no limits on the constraints and further, the common feasible region is not bounded in any respect.
Example 6: Find the graphical solution of problem formulated in example 3.

Solution

Let number of units needed for food stuf s F1 and F2 to meet the daily requirements of vitamins A and B be
respectively X1 and X2..
Objective function: Minimize Z =3X1+2.5X2
Subject to constraints: 2X1+4X2 ≥ 40
5X1+2X2 ≥50
Non-negativity
negativity restrictions X1, X2 0
First of al draw the graphs of these inequalities (as discussed in example 5) .Since the inequalities are of the
greater than or equal to type, the feasible region is formed by considering the area to the upper right side of
each equation i.e. away from origin .The shaded area which is shown above CBD is satisfied by the two
constraints as shown in Fig. 3.3 is feasible region. The feasible region for minimization problem is unbounded
and unlimited. Since the optimal solution corresponds to one of the corner (extreme) points, we wil calculate
the values of objective function for each corner point viz. D (20, 0); B (7.5, 6.25) and C (0, 25). The
calculations are shown in Table 3.3

Table 3.3 Table showing the computation of minimum value of objective function
Extreme Point Coordinates Cost Function Min. Z = 3X1+2.5X2
D X1=20 , X2=0 Z=3(20)+2.5(0)=60
B X1=7.5 , X2=6.25 Z=3(7.5)+2.5(6.25)=38.125
C X1=0, X2=25 Z=3(0)+2.5(25)=62.5
 Operations Research

Fig. 3.3 Feasible region for minimization problem

The minimum cost is obtained at the corner point B(7.5,6.25) i.e. X1=7.5 and X2=6.25. Hence to minimize
the cost and to meet the daily requirements of vitamin A and B, number of units needed of food stuf s F1 and
F2 be 7.5 and 6.25 respectively.

Infeasible solution

In this there is no solution to an LP Problem that satisfies al the constraints. Graphical y, it means that no
feasible solution region exists. Such a condition indicates that the LP problem has been wrongly formulated.

Redundant constraint

In a properly formulated LP problem, each of the constraints wil define a portion of the boundary of the
feasible solution region. Whenever, a constraint does not define a portion of the boundary of the feasible
solution region, it is cal ed a redundant constraint. Let us consider the fol owing example to il ustrate this.
Example 7: Maximize Z=1170X1+1110X2
Subject to:
9X1+5X2 ≥ 500
7X1+9X2 ≥ 300
5X1+3X2 ≤ 1500
7X1+9X2 ≤1900
2X1+4X2 ≤ 1000
Visit Our Website www.iaritoppers.com for Agri.Study Material
 Operations Research

X1, X2 0

The feasible region of this LP problem is indicated in Fig. 2.4. In this the feasible region has been formulated
by two constraints.
9X1+5X2 ≥ 500
7X1+9X2 ≤1900
X1, X2 0

Fig 3.4 Feasible Region and Redundant Constraints

The remaining three constraints, although present, is not af ecting the feasible region in any manner. Such
constraints are known as redundant constraints.
 Operations Research

Module 2. Linear programming problem

Lesson 4
SIMPLEX METHOD

Introduction

In the previous chapter we considered the formulation of linear programming problems and the graphic
method of solving them. Although the graphical approach to the solution of such problems is an invaluable aid
to understand its basic structure, the method is of limited application in industrial problems as the number of
variables occurring there is often considerably large. The Simplex Method provides an ef icient technique
which can be applied for solving linear programming problems of any magnitude-involving two or more
decision variables. The Simplex Method is the name given to the solution algorithm for solving LP problems
developed by George B. Dantzig in 1947. It is iterative procedure having fixed computational rules that leads
to a solution to the problem in a finite number of steps. By using this one is capable of solving large LP
problems ef iciently.

Principle of Simplex Method

As the fundamental theorem of LP problem tel s us that at least one basic feasible solution of any LP problem
must be optimal, provided the optimal solution of the LP problem exists. Also, the number of basic feasible
solutions of the LP problem is finite and at the most nCm (where, n is number of decision variables and m is
the number of constraints in the problem). On the other hand, the feasible solution may be infinite in number.
So it is rather impossible to search for optimal solutions from amongst al feasible solutions. Furthermore, a
great labour wil also be required in finding out al the basic feasible solutions and select that one which
optimizes the objective function. In order to remove this dif iculty; a simple method was developed by Dantzig
(1947) which is known as Simplex Algorithm. Simplex Algorithm is a systematic and ef icient procedure for
finding corner point solutions and taking them for optimality. The evaluation of corner points always starts from
the point of origin. This solution is then tested for optimality i.e. it tests whether an improvement in the
objective function is possible by moving to adjacent corner point of the feasible function space. This iterative
search for a better corner point is repeated until an optimal solution if it exists, is determined.

Basic Terms Involved in Simplex Procedure

The fol owing terms relevant for solving a linear programming problem through simplex procedure are given
below:

Standard form of linear programming problem

The standard form of LP problem is to develop the procedure for solving general LP problem. The optimal
solution of the standard form of a LP problem is the same as original LP problem. The characteristics of
standard form are given in folowing steps:

Step 1. Al the constraints should be converted to equations except for the non-negativity restrictions which
remain as inequalities (≥0).
Visit Our Website www.iaritoppers.com for Agri.Study Material
 Operations Research

Step 2. The right side element of each constraint should be made non-negative.
non negative.

Step 3. Al variables must have non-negative

non negative values.

Step 4. The objective function

function should be of maximization form.

Slack variables

If a constraint has less than or equal sign, then in order to make it on equality we have to add something
positive to the left hand side. The non-negative
negative variable which is added to the left hand side of the constraint to
convert it into equation is call ed the slack variable. For example, consider the constraints.
3X1 + 5X2 2, 7X1 + 4X2 5, X1, X2 0
We add the slack variables S1 0, S2 0 on the left hand sides of above inequalities respectively to obtain
3X1+5X2+S1 =2
7X1+4X2+S2 = 5
X1, X2, S1, S2 0

Surplus variables

If a constraint has greater than or equal to sign, then in order to make it an equality we have to subtract
something non-negative
negative from its left hand side. The positive variable which is subtracted from the left hand side
of the constraint to convert it into equation is cal ed the surplus variable.
For example, consider the constraints.

3X1 + 5X2 2, 2X1 + 4X2 5, X1, X2 0

We subtract the surplus variables S3 0, S4 0 on the left hand sides of above inequalities
respectively to obtain

3X1+5X2 -S
S3 = 2

2X1 + 4X2 -S
- 4=5

X1, X2, X3, X4 0

Solution to LPP

Any set X = {X1, X2, X3,…, Xn+m} of variables is ca

cal ed a solution to LP problem, if it satisfies the set of
constraints only.

Feasible solution (FS)

Any set X = {X1, X2, X3,……,Xn+m} of variables is cal ed a feasible solution of L.P. problem, if it satisfies
the set of constraints as wel as non-negativity
negativity restrictions.

Basic solution (BS)

 Operations Research
For a system of m simultaneous linear equations in n variables(n>m), a solution obtained by setting (n-m)

Visit Our Website www.iaritoppers.com for Agri.Study Material

 Operations Research

variables equal to zero and solving for the remaining variables is cal ed a basic solution. Such m variables (of
course, some of them may be zero) are cal ed basic variables and remaining (n-m) zero-valued variables are
cal ed non-basic variables.

Basic feasible solution (BFS)

A basic feasible solution is a basic solution which also satisfies the non-negativity restrictions, that is al basic
variables are non-negative. Basic solutions are of two types
(a) Non-degenerate BFS: A non-degenerate basic feasible solution is the basic feasible solution which has
exactly m positive Xj (j=1, 2,…,m). In other words, al basic m variables are positive, and the remaining
(n –m) variables wil be al zero.
(b) Degenerate BFS: A basic feasible solution is cal ed degenerate, if one or more basic variables are zero-
valued.
Optimum basic feasible solution

A basic feasible solution is said to be optimum, if it also optimizes (maximizes or minimizes) the objective
function.

Unbound solution

If the value of the objective function Z can be increased or decreased indefinitely, such solutions are cal ed
unbounded solutions.

Computational Aspect of Simplex Method for Maximization Problem

Step 1: Formulate the linear programming model. If we have n-decision variables X1, X2,…, Xn and m
constraints in the problem , then mathematical formulation of L P problem is
Maximize Z=C1X1+ C2X2+…+CnXn
Subject to the constraints:
a11 X1+ a12 X2+ … +a1n Xn ≤ b1
a21 X1+ a22 X2 + … + a2n Xn ≤ b2
.
am1 X1+ am2 X2 + … + amn Xn ≤ bm

X 1, X 2, … X n ≥ 0
Step 2: Express the mathematical model of LP problem in the standard form by adding slack variables in the
left–hand side of the constraints and assign zero coef icient to these variables in the objective function.Thus we
can restate the problem in terms of equations as fol ows:
Maximize Z=C1X1+ C2X2 +…+ CnXn +0Xn+1 +0Xn+2+…+0Xn+m

Subject to the constraints:

a11 X1+ a12 X2+ …+ a1n Xn +Xn+1 = b1
a21 X1+ a22 X2+…+ a2n Xn +Xn+2 = b2
.
Visit Our Website www.iaritoppers.com for Agri.Study Material
 Operations Research

am1 X1+ am2X2+…+amn Xn +Xn+m = bm

X1, X2, … ,Xn, Xn+1,Xn+2 ,…, Xn+m ≥0

Step 3: Design the initial feasible solution .An initial basic feasible solution is obtained by setting X1=X2=…
=Xn =0. Thus, we get Xn+1=b1, Xn+2 =b2 ,…, Xn+m =bm.
Step 4: Construct the starting (initial) simplex tableau. For computational eff iciency and simplicity, the initial
basic feasible solution, the constraints of the standard LP problem as well as the objective function can be
displayed in a tabular form, cal
ca ed the Simplex Tableau as shown below.

Table 4.1 Initial simplex tableau

Cj (contribution per unit) → c1 c2 … cn 0 0 … 0
Cb Basic Value of Coef icient Matrix Identify Matrix
Minimum
Variables Basic
Ratio*
Variables
B b(=XB) X1 X2 … Xn Xn+1 Xn+2 … Xn+m
0 s1 b1 a11 a12 … a1n 1 0 … 0
0 s2 b2 a21 a22 … a2n 0 1 … 0
. . . . . … .
. . . . . … .
. . . . . … .
0 sm bm am1 am2 … amn 0 0 … 1
Contribution 0 0 … 0 0 0 … 0
loss per unit:
Net c1 c2 … cn 0 0 … 0
contribution
per units:

* Negative ratio is not to be considered.

The interpretation of the data in the above ‘Tableau

Tableau’ is given as under. Other simplex tableau wil have similar
interpretations.
(i)
( The first row, cal ed the objective row of the simplex table indicates the values of Cj (j subscript refer
to the column number) which are the coef icients of the (m + n) variables in the objective function.
These coef
coe icients are obtained directly from the objective function and the value Cj would remain the
same in the succeeding tables. The second row of the table provides the major column headings for
the table and these column headings remain unchanged in the succeeding tables of the Simplex
Method.
(i)
(i ) The first column label
labe ed CB, also known as objective column,
column lists the coeff icient of the current
basic variables in the objective function. The second column label ed ‘Basic variables’ points out the
basic variables in the basis, and in the initial simplex tableau these basic variables are the slack
variables. The third column labell ed ‘Solution
Solution values
values’ (= xB), indicates the resources or the solution
Operations Research
values of

Visit Our Website www.iaritoppers.com for Agri.Study Material

 Operations Research

the basic variables.

(ii)
(ii ) The body matrix (under non-basic
basic variables) in the initial simplex tableau consists of the coef
coe icients
of the decision variables in the constraint set.

(iv)
( The identity matrix in the initial simplex tableau represents the coeff icient of the slack variables that
have been added to the original inequalities to make them equation. The matrix under non-basic
variables in the simplex tableau is cal ed coefficient matrix. Each simplex tableau contains an identity
matrix under the basic variables.

(v)
( To find an entry in the Zj row under a column, we multiply the entries of that column by the

corresponding entries of CB – column and add the products, i.e., Z=

.

The Zj entry under the ““Solution

Solution column”
column” gives the current value of the objective function.

(vi)
( The final row labeled ∆j = Zj - Cj cal ed the index (or net evaluation) row, is used to determine whether
or not the current solution is optimal. The calculation of Zj - Cj row simply involves subtracting each Zj
value from the corresponding Cj value for that column, which is written at the top of that column. Each
entry in the ∆j row represents the net contribution (or net marginal improvement) to the objective
function that results by introducing one unit of each of the respective column variables.

Step 5: Test the Solution for Optimality.

Optimality Examine
ne the index row of the above simplex tableau. If al the
elements in the index row are positive then the current solution is optimal. If there exist some negative values,
the current solution can further be improved by removing one basic variable from the basis and replacing it by
some non-basic one.
Step 6: Revision of the Current Simplex Tableau
Tableau. At each iteration, the Simplex Method moves from the
current basic feasible solution to a better basic feasible solution. This involves replacing one current basic
variable (cal ed the departing variable) by a new non-basic
basic variable (cal ed the entering variable).

(a) Determine which variable to enter into the solution-mix

solution mix net. One way of doing this is by
identifying the column (and hence the variable) with the most negative number in the ∆j row of the
previous table.

(b) Determine the departing variable or variable to be replaced. Next we proceed to determine
which variable must be removed from the basis to pave way for the entering variable. This is
accomplished by dividing each number in the quantity (or solution values) column by the
corresponding number in the pivot column selected in (a), i.e., we compute the respective ratios
b1/a1j, b2/a2j, …, bm/amj (only for those aij’s; i=1,2,…, m which are strictly
trictly positive). These
quotients are written in the last column label
labe ed ‘Minimum Ratio’ of the simplex tableau. The row
corresponding to smal est of these non--negative ratios is cal ed the pivot (or key) row and the
corresponding basic variable wil leave the basis. Let the minimum of { b1/a1j, b2/a2j, …, bm/amj ;
 Operations Research

aij > 0} be bk/akj , then corresponding variables in the pivot row sk wil be termed as outgoing (or
departing) variable in the next tableau to be constructed just after we put an arrow → of type to
right of kth row of the simplex tableau 1.

(c) Identify the pivot number. The non-zero

zero positive number that lies at the intersection of the pivot
column and pivot row of the given table is cal ed the pivot (or key) number. We place a circle
around the number.
Step 7: Evaluate (update) the new solution. After identifying the entering and departing variable, find the
new basic feasible solution by constructing a new simplex tableau from the current one by using the fol
fo owing
steps:
(a) Compute new values for the pivot row by simply dividing every element of the pivot row by the
pivot number.
(b) New entries in the CB column and XB column are entered in the new table of the current solution
(c) Compute new values for each of the remaining rows by using the fo
fol owing formula
New row numbers=Number in old rows-{(corresponding
{(corresponding number above or below pivot
number)x(corresponding number in the row replaced in (a))}
(d) Test for optimality. Compute the Zj and index rows as previously demonstrated in the initial simplex
tableau. If al numbers in the index row are either zero or positive, an optimal solution has been made
attained. i.e., there is no variable which can be introduced in the solution to cause the objective
function value to increase.
4. Revise the solution. If any of the numbers in the index (∆j = Zj - Cj) row are negative, repeat the entire
steps 5 & 6 again until an optimal solution has been obtained.

The above procedure is il

il ustrated through the fol
fol owing example.

Example 1

A firm produces three products A, B, and C each of which passes through three dif erent departments
fabrication, finishing, packaging. Each unit of product A requires 3, 4 and 2 hours respectively, B requires 5, 4
and 4 hours respectively and C requires 2, 4 and 5 hours respectively in 3 departments respectively. Every
day 60 hours are available in fabrication department, 72 hours in finishing and 100 hours in packaging
department. If unit contribution of unit A is Rs. 5, Rs. 10 for B and Rs. 3 for C. Then determine number of
units of each product so that total contribution to cost is maximized and also determine if any capacity would
remain unutilized.

Solution:

Step 1 : Formulate this as LPP. Let X1, X2 and X3 be the number of units produced of the products A, B
and C respectively.

Objective function:
function Max Z = 5X1 + 10X2 + 3X3

Subject to constraints:
constraints 3X1 + 5X2 + 2X3 60
 Operations Research

4X1 + 4X2 + 4X3 72

2X1 + 4X2 + 5X3 100 X1, X2, X3

0

Step 2 :Now converting into standard form of LPP

Max Z= 5X1 + 10X2 + 3X3 + 0S1 + 0S2 + 0S3

3X1 + 5X2 + 2X3 + S1 = 60

4X1 + 4X2 + 4X3 + S2 = 72

2X1 + 4X2 + 5X3 + S3 = 100 X1, X2, X3, S1, S2, S3 0

where S1 ,S2 and S3 are slack variables.

Step 3: Find the initial feasible solution .An initial basic feasible solution is obtained by setting X1 =0, X2 = 0
and X3 = 0. Thus, we get S1 = 60, S2 = 72 and S3 =100.

Step 4: Construct the starting (initial) simplex tableau.

Cj → 5 10 8 0 0 0
B.V. CB XB X1 X2 X3 S1 S2 S3 Minimum
Ratio
R1 S1 0 60 3 (5) 2 1 0 0 60/5=12
R2 S2 0 72 4 4 4 0 1 0 72/4=18
R3 S3 0 100 2 4 5 0 0 1 100/4=25
Zj 0 0 0 0 0 0
−5 −10 −8 0 0 0

Step 5: The most negative value of ∆j is −10 hence X2 is the incoming variable (↑) and the least positive
). The element under column X2 and row R1 is the
minimum ratio is 12 hence S1 is the outgoing variable (

key element i.e. 5 so divide each element of row R1 by 5 (i.e. ). Subtract appropriate multiples of this
new row from the remaining rows, so as to obtain zeros in the remaining positions. Performing the row
operations

and
we get the second Simplex tableau as
 Operations Research

Cj → 5 10 8 0 0 0
B.V. CB XB X1 X2 X3 S1 S2 S3 M.R.
X2 10 12 3/5 1 2/5 1/5 0 0 12/2/5=30
S2 0 24 8/5 0 12/5 −4/5 1 0 24/12/5=10
S3 0 52 -
-2/5 0 17/5 −4/5 0 1 52/17/5=15.294
Zj 6 10 4 2 0 0
1 0 −4 2 0 0

Step 6: - hence X3 is the incoming variable ((↑) and the least positive
6 The most negative value of ∆j is -4
minimum ratio is 10 hence S2 is the outgoing variable (). The element under column X2 and row

R1 is the key element i.e. 5 so divide each element of row R b2 by 12/5 (i.e. Rc → Rb * 5/12). Subtract
appropriate multiples of this new row from the remaining rows, so as to obtain zeros in the remaining

positions. Performing the row operations

and
we get the third Simplex tableau as

Cj → 5 10 8 0 0 0
B.V. CB XB X1 X2 X3 S1 S2 S3
X2 10 8 1/3 1 0 1/3 −1/6 0
X3 8 10 2/3 0 1 −1/3 5/12 0
S3 0 18 −8/3 0 0 1/3 −17/12 1
Zj 26/3 10 8 2/3 5/3 0
11/3 0 0 2/3 5/3 0
j

It is apparent from this table that al ∆j =Zj − Cj are positive and therefore an optimum solutions is reached .
So X1 = 0, X2 = 8, X3 = 10

Z = 5X1 + 10X2 + 8X3 = 160

And also as S3 is coming out to be 18 so there are 18 hours unutilized in finishing department.
In case the objective function of the given LP problem is to be minimized, then we convert it into a problem of
maximization by using
Min. Z* = − Max. ((−Z). The procedure of finding optimal solution using Simplex Method is il
i ustrated through
the fol
fo owing example:

Example 4.2
 Operations Research

Minimize the objective function Z: X1 – 3X2 + 2X3

Subject to constraints 3X1 – X2 + 3 X3 7

−2X1 + 4X2 12

−4X1 + 3X2 + 8X3 10

X1, X2, X3 0

Solution

Converting this minimization problem into maximization problem

Objective function Max Z#: −X1 – 3X2 + 2X3 + 0S1 + 0S2 + 0S3

Constraints 3X1 – X2 + 3 X3 + S1 = 7

−2X1 + 4X2 + S2 = 12

−4X1 + 3X2 + 8X3 + S3 = 10

X1, X2, X3, S1, S2, S3 0

Starting Simplex Table

Cj −1 3 −2 0 0 0
B.V. CB XB X1 X2 X3 S1 S2 S3 Min. Ratio
S1 0 7 3 −1 3 1 0 0 -
S2 0 12 −2 4 1 0 1 0 3 →
S3 0 10 −4 3 8 0 0 1 10/3
1 −3 2 0 0 0
j

Now performing the row operations

Cj −1 3 −2 0 0 0 Min. Ratio
B.V. CB XB X1 X2 X3 S1 S2 S3
S1 0 10 5/2 0 3 1 1/4 0 4→
X2 3 3 −1/2 1 0 0 1/4 0 -
X3 0 1 −5/2 0 8 0 −3/4 1 -
−1/2 0 2 0 3/4 0
j

Now performing the row operations

 Operations Research

↑ Cj −1 3 −2 0 0 0
B.V. CB XB X1 X2 X3 S1 S2 S3
X1 −1 4 1 0 6/5 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
∆j 0 0 13/5 1/5 4/5 0

Now al
a
j are positive. Therefore, Optimal Solution is reached
Therefore, X1 = 4, X2 = 5 & X3 = 0

& Max Z# = −1 4 + 5 = 11

Or Minimum Z = −Z# = −11

 Operations Research

Module 3. Transportation problems

Lesson 5
DEFINITION AND MATHEMATICAL FORMULATION

Introduction

The Transportation Problem (TP) is a special type of LP problem where the objective is to minimize the cost
of distributing a single commodity from a number of supply sources (e.g. factories) to a number of demand
destinations (e.g. warehouses). The objective of the problem is to determine the amount to be transported
from each source to each destination so as to maintain the supply and demand requirements at the lowest
transportation cost. The characteristic of a TP are such that it is usual y solved by a specialized method rather
than by simplex method. A key problem in many projects is the al ocation of scarce resources among various
activities. TP refers to a planning model that al ocates resources, machines, materials, capital etc. in the best
possible manner so that the costs are minimized or profits are maximized.

Formulation of Transportation Problem

Transportation problem is an important class of the linear programming problem in which, the objective is to
transport various quantities of a single homogenous commodity that are stored at various origins to dif erent
destinations in such a way that the transportation cost is minimum. Transportation problem arises in situations
involving physical movements of goods e.g. milk and milk products from plants to cold storages, cold
storages to wholesalers, wholesalers to retailers and retailers to customers. The solution of a TP is to
determine the quantity to be shifted from each plant to each cold storage so as to maintain the supply and
demand requirements at the lowest transportation cost.

Let us suppose that there are m origins (dairy plants) supplying a certain homogeneous dairy product to n
destinations (cold storages). Let plant Pi (i = 1, 2, …,m) produce ai units and cold storage Wj (j = 1, 2,. . .,
n) requires bj units. Suppose that the cost of transporting from plant Pi to cold storage Wj is directly
proportional to the amount/quantity transported and let Cij be the cost of transporting one unit of product

from ith origin to jth destination and Xij be the amount/quantity transported from ith origin to jth destination.

The objective is to determine the number of units to be transported from ith origin to jth destination such that

the transportation cost

is minimized. For balanced transportation problem, it is assumed
that the total supply equals to the total demand. The transportation problem in a tabular form can be
represented as follows:
Table 5.1 Tabular re pre se ntation of transportation proble m
Dairy Dairy Plant
Ware house s (Cold storage )/De stinations
Plants/
Capacity
Source s W1 W2 ….. ….. Wj ….. Wn

P1 …. …. …… a1
 Operations Research
P2 …. …... ….. a2

: : : : : : : :

Pi : : : ai

: : : : : : : :

Pm …. …. …… am

Cold storage b1 b2 …. ….. bj ….. bn

Requirements

Where Cij and Xij are unit cost of transportation and quantity transported respectively in the cell (i, j). Then the

sum of the product of Xij and Cij of allocated cells

gives us the net cost in transporting Xij units from plant Pi to cold storage Wj.

The following example will help to demonstrate the formulation of the transportation problem:

Example 1 :

Milk in a milk shed area is col

co ected on three routes A, B and C. There are four chil ing centers P, Q, R and
S where milk is kept before transporting it to a milk plant. Each route is able to supply on an average one
thousand litres of milk per day. The supply of milk on routes A, B and C are 150, 160 and 90 thousand litres
respectively. Daily capacity in thousand litres of chill ing centers is 140, 120, 90 and 50 respectively. The cost
of transporting 1000 litres of milk from each route (source) to each chi
chil ing center (destination) dif ers
according to the distance. These costs (in Rs.) are shown in the fol owing table:

Chilling centers
Routes P Q R S
A 16 18 21 12
B 17 19 14 13
C 32 11 15 10

The problem is to determine how many thousand liters of milk is to be transported from each route on daily
basis in order to minimize the total cost of transportation.

Solution

A transportation problem can be formulated as a linear programming problem. To il ustrate this, let us see
how this example can be formulated as a linear programming problem. Let Xij represent the quantity

transported from ith route to jth chil ing center. These are the decision variables. In this example there are
 Operations Research

twelve decision variables. Let Cij represents the cost of transported thousand litres of milk from ith route to j
th chil
chi ing center. The objective is to find the values for the Xij so as to minimize total transportation cost. Thus
the LP objective function is:

The supply of milk on routes A, B and C and daily capacity in thousand litres of chil ing centers P, Q, R and
S impose constraints. The total quantity transported from route A must be equal to its capacity i.e. 150
thousand litres milk. Thus, for route A, the constraint is:

Similarly, the constraint for other two routes B and C can be expressed as under:

Similarly, we must satisfy the demand for each of the four chil ing centers. The units can be transported
through route A, B and C. Thus, for Chil
Chi ing centre P, the constraint is:

Similarly, constraints for the other three chi

chil ing centers Q, R and S

Finall y, al values of Xij must be greater than or equal to zero, as negative units cannot be transported. Thus,

for i=1,2,3 ; j=1,2,3,4.

Thus the linear programming formulation of the transportation problem becomes:

Subject to
Supply constraints De mand Constraints

and
for i=1,2,3 ; j=1,2,3,4.

Mathe matical Formulation of the Transportation Proble m

 Operations Research

Using the notations described in the previous section, the transportation problem consist of finding Xij (i =1,

2, . . ., m; j=1, 2, . . , n) in order to minimize the total transportation cost Z =

It is assumed that the total availabilities equals to the total requirements i.e.
(rim condition). The problem now is to determine non
non–negative
negative values of Xij

which satisfy both the availability constraints and requirement constraints

Such types of problems where supply and demand are exactly equal are known as Balanced
Transportation Problem
Problem.. Supply (from various sources) is written in the rows, while a column is an
expression for the demand of dif erent warehouses. In general, if a transportation problem has m rows and n
columns, then the problem is solvable if there are exactly (m + n –1)
– basic variables. However, a
transportation problem is unbalanced if the total supply and the total demand are not equal.
 Operations Research

Module 3. Transportation problem

Lesson 6
INITIAL BASIC FEASIBLE SOLUTION

Introduction

In the last lesson we have learnt about Transportation Problem (TP) and its formulation. Transportation
problem can be solved by simplex method and transportation method. In simplex method the solution is very
lengthy and cumbersome process because of the involvement of a large number of decision and artificial
variables. In this lesson we wil look for an alternate solution procedure cal ed transportation method in which
initial basic feasible solution of a TP can be obtained in a better way by exploiting the special structure of the
problem.

Some Definitions

The fol owing terms are to be defined with reference to Transportation Problem

Feasible solution (FS)

By feasible solution we mean a set of non-negative individual al ocations (Xij ≥ 0) which satisfies the row and
column conditions (rim condition).

Basic feasible solution (BFS)

A feasible solution is said to be basic if the number of positive al ocations equals m+n-1; that is one less than
the number of rows and columns in a transportation problem.

Optimal solution

A feasible solution (not necessarily basic) is said to be optimal if it minimizes the total transportation cost.

Solution for Transportation Problem

The solution algorithm to a transportation problem can be summarized into fol owing steps:
Step 1: Formulate the problem. The formulation of transportation problem is similar to a LP problem
formulation. Here the objective function is to minimize the total transportation cost and the
constraints are the supply and demand available at each source and destination, respectively.
Step 2: Obtain an initial basic feasible solution. This initial basic feasible solution can be obtained by using any
of the folowing five methods:
a) North West Corner Rule
b) Minimum cost method
c) Row Minimum Method
d) Column Minimum Method
e) Vogel’s Approximation Method
The solution obtained by any of the above methods must fulfil the fol owing conditions:
i. The solution must be feasible, i.e., it must satisfy al the supply and demand constraints. This is cal ed
rim condition.
Visit Our Website - www.InstantAgriculture.com 37 Download Our App - Instant Agriculture
 Operations Research

i. . The number of positive al ocations must be equal to m+n–1, where, m is number of rows and n
is number of columns.
The solution that satisfies both the above mentioned conditions is cal ed a non-degenerate basic feasible
solution.
Step 3: Test the initial solution for optimality. Using any of the fol owing methods one can test the optimality of
an initial basic solution:
i. Stepping Stone Method
i . Modified Distribution Method (MODI)
If the solution is optimal then stop, otherwise, find a new improved solution.
Step 4: Updating the solution. Repeat Step 3 until the optimal solution is obtained.

Methods of Obtaining an Initial Basic Feasible Solution

Five methods are described to obtain the initial basic feasible solution of the transportation problem. These
methods can be explained by considering the fol owing example
Example 1 :
Let us consider the formulation of TP as given in example 1 of the previous lesson which can be represented
by the fol owing e transportation table:

Each cel in the table represents the amount transported from one route to one chil ing center. The amount
placed in each cel is, therefore, the value of a decision. The smal er box within each cel contains the unit
transportation cost for that route.

6.4.1 North west corner rule (NWC) method

Step I: The first assignment is made in the cel occupying the upper left-hand (North West) corner of the
transportation table. The maximum feasible amount is al ocated there. That is X11 = Min (a1, b1) and
this value of X11 is then entered in the cel (1, 1) of the transportation table.
Step II: a) If b1>a1, we move down vertical y to the second row and make the second al ocation of
magnitude X21=Min. (a2, b1- X11) in the cel (2, 1).
b) If b1< a1 we move horizontal y to the second column and make the second al ocation of magnitude
X12 = Min. (a1–X11, b2) in the cel (1, 2).

Visit Our Website - www.InstantAgriculture.com 38 Download Our App - Instant Agriculture

 Operations Research

c) If b1= a1, there is a tie for the second al ocation. One can make the second al ocation of magnitude
X12 = Min. (a1–a1, b1) = 0 in the cel (1, 2) or X21=Min. (a2, b1–b1) = 0 in the cel (2, 1)
Step III: Repeat steps I & II by moving down towards the lower right corner of the transportation table
until al the rim requirements are satisfied.
6.4.1.1 Illustration of North –West Corner Rule:
Let us il ustrate the above method with the help of example 1given above. Fol owing North-West corner rule,
the first al ocation is made in the cel (1,1), the magnitude being X11=Min.(150,140)=140. The second
al ocation is made in cel (1,2) and the magnitude of al ocation is given by X12=Min. (150-140,120)=10.
Third al ocation is made in the cel (2,2), the magnitude being X22=Min. (160,120-10)=110. The magnitude
of fourth al ocation in the cel (2,3) is given by X23=Min. (160-110, 90)=50. The fifth al ocation is made in
the cel (3,3), the magnitude being X33= Min. (90-50, 90)=40. The sixth al ocation in the cel (3,4) is given by
X34=Min. (90-40, 50) = 50. Now al requirements have been satisfied and hence an initial basic feasible
solution to T.P. has been obtained.

The transportation cost according to the above al ocation is given by z = 16 × 140 + 18 × 10 + 19 × 110 +
14 × 50 + 15 × 40 + 10 × 50 = 6310
6.4.2 Matrix Minima ( Least Cost Entry) Method

Various steps of Matrix Minima method are given below:

Step I: Determine the smal est cost in the cost matrix of the transportation table. Let it be Cij. Al ocate Xij =
Min. (ai, bj) in the cel (i, j).

Step II: a) If Xij = ai, cross of the ith row of the transportation table and decrease bi by ai and go to Step
III.
b) If Xij = bj, cross of the jth column of the transportation table and decrease ai by bj and go to
Step III.
c) If Xij = ai = bj, cross of either the ith row or the jth column but not the both.

Step III: Repeat steps I & II for the resulting reduced transportation table until al the requirements are
satisfied. Whenever the minimum cost is not unique, make an arbitrary choice among the minima.

Visit Our Website - www.InstantAgriculture.com 39 Download Our App - Instant Agriculture

 Operations Research

6.4.2.1 Illustration of Matrix Minima ( Least Cost Entry) Method:

Let us il ustrate this method by considering example 1given earlier in this lesson . The first al ocation is made in
the cel (3, 4) where the cost of transportation is minimum, the magnitude being X34=50. This satisfies the
requirement of S chil ing centre. Therefore cross of the fourth column. The second al ocation is made in the
cel (3,2) having minimum cost among remaining cel s and its magnitude being X32=Min.(120,90-50)=40.
This satisfies the supply of route C. Cross of the third row. Among the remaining cel s, the minimum cost is
found in cel (2, 3) so the third al ocation is done in cel (2, 3) and its magnitude being X23=Min.(90,160)=90.
The requirement of the chil ing center R is fulfil ed, hence third column is crossed of . Out of the remaining
cel s the minimum cost is found in cel (1,1) and its magnitude is X11=Min(140,150)=140, as the requirement
of chil ing center P is exhausted hence column first is deleted. Among remaining two cel s minimum cost is
found in cel (1,2) and its magnitude X12=Min(80,150-140)=10 and the last al ocation is done in the cel (2,2)
and its magnitude is X22=Min(80-10,70)=70. Now al requirements have been satisfied and hence an initial
basic feasible solution to T.P. has been obtained and given in the fol owing table.

The transportation cost according to the above al ocation is given by z = 16 × 140 + 18 × 10 + 19 × 70 +

11 × 40 + 14 × 90 + 10 × 50 = 5950

6.4.3 Row Minima Method

Various steps of Row Minima method are given below:

Step I: The smal est cost in the first row of the transportation table is determined, let it be Cij. Al ocate the
maximum feasible amount Xij = Min. (ai, bj) in the cel (i, j), so that either the capacity of origin Oi is
exhausted or the requirement at destination Dj is satisfied or both.
Step II: a) If Xij = a1, so that the availability at origin O1 is completely exhausted, cross of the first row of
the table and move down to the second row.
b) If Xij=bj so that the requirement at destination Dj is satisfied. Cross of the jth column and
reconsider the first row with the remaining availability of origin O1.
c) If Xij=a1=bj , the origin capacity of O1 is completely exhausted as wel as the requirement at
destination D1 is completely satisfied. An arbitrary choice is made. Cross of the first row and

Visit Our Website - www.InstantAgriculture.com 40 Download Our App - Instant Agriculture

 Operations Research

make the second al ocation X1k = 0 in the cel (1, k) with C1k being the new column cost in the
first row. Cross of the first row and move down to the second row.

Step III: Repeat steps I & II for the resulting reduced transportation problem until the requirements are
satisfied.

6.4.3.1 Illustration of Row Minima Method:

Let us il ustrate this method by considering Example 1given earlier in this lesson. In the first row smal est cost
is in the first row which is in the cel (1,4) al ocate minimum possible amount X14=Min(50,150)=50. This
exhausts requirement of S chil ing centre and thus we cross of the fourth column from the transportation
table. In the resulting transportation table, the minimum cost in the first row is 16 in the cel (1, 1), the second
al ocation is made in the cel (1, 1) with X11=Min (140,150-50) =100. This satisfies the supply of Route A,
hence first row is deleted. In the second row of the remaining transportation table, the minimum cost is
present in cel (2,3) so third al ocation is done X23=Min(90,160)=90.This satisfies the requirement of R
chil ing centre, hence third column is deleted. The next al ocation is made in the cel (2,1) with X21=Min(140-
100,160-90)=40.This exhausts the requirement of Route P and so we cross-out the first column .The fifth
al ocation is made in cel X22=Min(120,70-40)=30.This exhausts the supply of route B and hence it is
crossed out. In third row the cel left is (3, 2). Hence the last al ocation of amount X32=Min (120-30, 90)
=90 is obviously made in the cel (3,2). Now al requirements have been satisfied and hence an initial basic
feasible solution of TP has been obtained and given in the fol owing table.

The transportation cost according to the above al ocation is given by z = 16 × 100 + 12 × 50 + 17 × 40 +

19 × 30 + 14 × 90 + 11 × 90 = 5700

The Column Minima Method

This method is same as Row minima method except that we apply the concept of minimum cost in column
instead of row.

Vogel’s Approximation Method (VAM)

The Vogel’s Approximate Method takes into account not only the least cost Cij but also the costs that just
exceed Cij. Various steps of the method are given below:

Visit Our Website - www.InstantAgriculture.com 41 Download Our App - Instant Agriculture

 Operations Research

Step I: For each row of the transportation table identify the smal est and the next to smal est cost. Determine
the dif erence between them for each row. These are cal ed penalties (opportunity cost). Put them
along side of the transportation table by enclosing them in parenthesis against the respective rows.
Similarly compute these penalties for each column.
Step II: Identify the row or column with the largest penalty among al the rows and columns. If a tie occurs,
use any arbitrary tie breaking choice. Let the greatest penalty correspond to ith row and let cij be the

smal est cost in the ith row. Al ocate the largest possible amount Xij = min. (ai, bj) in the (i, j) of the

cel and cross of the ith row or jth column in the usualmanner.
Step III: Re-compute the column and row penalty for the reduced transportation table and go to step II.
Repeat the procedure until al the requirements are satisfied.
Remarks:
1. A row or column ‘dif erence’ indicates the minimum unit penalty incurred by failing to make an al
ocation to the smal est cost cel in that row or column.
2. It wil be seen that VAM determines an initial basic feasible solution which is very close to the
optimum solution that is the number of iterations required to reach optimum solution is minimum in this
case.
6.4.2 Illustration of Vogel’s Approximation Method (VAM):

Let us il ustrate this method by considering Example 1 discussed before in this lesson .

For each row and column of the transportation table determine the penalties and put them along side of the
transportation table by enclosing them in parenthesis against the respective rows and beneath the
corresponding columns. Select the row or column with the largest penalty i.e. (7) (marked with an arrow)
associated with second column and al ocate the maximum possible amount to the cel (3,2) with minimum cost
and al ocate an amount X32=Min(120,90)=90 to it. This exhausts the capacity of route C. Therefore, cross
of third row. The first reduced penalty table wil be:

Visit Our Website - www.InstantAgriculture.com 42 Download Our App - Instant Agriculture

 Operations Research

In the first reduced penalty table the maximum penalty of rows and columns occurs in column 3, al ocate the
maximum possible amount to the cel (2,3) with minimum cost and al ocate an amount X23=Min(90,160)=90
to it . This exhausts the capacity of chil ing center R .As such cross of third column to get second reduced
penalty table as given below.

In the second reduced penalty table there is a tie in the maximum penalty between first and second row.
Choose the first row and al ocate the maximum possible amount to the cel (1,4) with minimum cost and
al ocate an amount X14=Min(50,150)=50 to it . This exhausts the capacity of chil ing center S so cross of
fourth column to get third reduced penalty table as givenbelow:

The largest of the penalty in the third reduced penalty table is (2) and is associated with first row and second
row. We choose the first row arbitrarily whose min. cost is C11 = 16. The fourth al ocation of X11=min. (140,
100) =100 is made in cel (1, 1). Cross of the first row. In the fourth reduced penalty table i.e. second row,
minimum cost occurs in cel (2,1) fol owed by cel (2,2) hence al ocate X21=40 and X22=30. Hence the
whole al ocation is as under:

Visit Our Website - www.InstantAgriculture.com 43 Download Our App - Instant Agriculture

 Operations Research

The transportation cost according to the above al ocation is given by z = 16 × 100 + 12 × 50 + 17 × 40 +

19 × 30 + 14 × 90 + 11 × 90 = 5700

NOTE: General y, Vogel’s Approximation Method is preferred over the other methods because the initial
BFS obtained is either optimal or very close to the optimal solution.

Visit Our Website - www.InstantAgriculture.com 44 Download Our App - Instant Agriculture

 Operations Research

Module 3. Transportation problem

Lesson 7
OPTIMAL SOLUTION

Introduction

After examining the initial basic feasible solution, the next step is to test the optimality of basic feasible
solution. Though the solution obtained by Vogel’s method is not optimal, yet the procedure by which it was
obtained often yields close to an optimal solution. So to say, we move from one basic feasible solution to a
better basic feasible solution, ultimately yielding the minimum cost of transportation. There are two methods of
testing optimality of a basic feasible solution. The first of these is cal ed the Stepping Stone method and the
second method is cal ed Modified Distribution method (MODI).By applying either of these methods, if the
solution is found to be optimal, then problem is solved. If the solution is not optimal, then a new and better
basic feasible solution is obtained. It is done by exchanging a non-basic variable for one basic variable i.e.
rearrangement is made by transferring units from an occupied cel to an empty cel that has the largest
opportunity cost and then shifting the units from other related cel s so that al the rim requirements are
satisfied.

Some Definitions

Non-degenerate solution

A basic feasible solution of an m x n transportation problem is said to be non-degenerate, if it has the

fol owing two properties:
(1) Starting BFS must contain exactly (m + n -1) number of individual al ocations.
(2) These al ocations must be in independent positions.
Here by independent positions of a set of al ocations we mean that it is always impossible to form closed
loops through these al ocations. The fol owing table show the non-independent and independent positions
indicated by the fol owing diagram:

Degeneracy

If the feasible solution of a transportation problem with m origins and n destinations has fewer than m+n-1
positive Xij (occupied cel s), the problem is said to be a degenerate transportation problem. Degeneracy can
occur at two stages:
a) At the initial stage of Basic Feasible Solution.
Visit Our Website - www.InstantAgriculture.com 45 Download Our App - Instant Agriculture
 Operations Research

b) During the testing of the optimal solution.

To resolve degeneracy, we make use of artificial quantity.

Closed path or loop

This is a sequence of cel s in the transportation tableau such that

a) each pair of consecutive cel s lie in either the same row or the same column.
b) no three consecutive cel s lie in the same row or column.
c) the first and last cel s of a sequence lie in the same row or column.
d) no cel appears more than once in the sequence.

Stepping Stone Method

In this method, the net cost change can be obtained by introducing any of the non-basic variables
(unoccupied cel s) into the solution. For each such cel find out as to what ef ect on the total cost would be if
one unit is assigned to this cel . The criterion for making a re-al ocation is simply to know the desired ef ect
upon various costs. The net cost change is determined by listing the unit costs associated with each cel and
then summing over the path to find the net ef ect. Signs are alternate from positive (+) to negative (-)
depending upon whether shipments are being added or subtracted at a given point. A negative sign on the net
cost change indicates that a cost reduction can be made by making the change while a positive sign wil
indicate a cost increase. The stepping stone method for testing the optimality can be summarized in the
fol owing steps:
Step
1. Determine an initial basic feasible solution.

2. Make sure that the number of occupied cel s is exactly equal to m+n-1, where m is number of rows and
n is number of columns.
3. Evaluate the cost–ef ectiveness of transporting goods through the transportation routes not currently in
solution. The testing of each unoccupied cel is conducted by fol owing four steps given as under:
a. Select an unoccupied cel , where transportation should be made. Beginning with this cel , trace a closed
path using the most direct route through at least three occupied cel s and moving with only horizontal and
vertical moves. Further, since only the cel s at the turning points are considered to be on the closed path,
both unoccupied and occupied boxes may be skipped over. The cel s at the turning points are cal ed
Stepping Stone on the path.

b. Assign plus (+) and minus (-) sign alternatively on each corner cel of the closed path traced starting a
plus sign at the unoccupied cel to be evaluated.
c. Compute the net change in the cost along the closed path by adding together the unit cost figures found
in each cel containing a plus sign and then subtracting the unit cost in each square containing the minus
sign.
d. Repeat sub step (a) through sub step (b) until net change in cost has been calculated for al unoccupied
cel s of the transportation table.

4. Check the sign in each of the net changes .If al net changes computed are greater than or equal to zero, an
optimal solution has been reached. If not, it is possible to improve the current solution and decrease total
transportation cost.

Visit Our Website - www.InstantAgriculture.com 46 Download Our App - Instant Agriculture

 Operations Research

5. Select the unoccupied cell having the highest negative net cost change and determine the maximum number
of units that can be assigned to a cell marked with a minus sign on the closed path, corresponding to this
cel
ce . Add this number to the unoccupied cel and to other cel s on the path marked with a plus sign.
Subtract the number from cel s on the closed path marked with a minus sign.

6. Go to step 2 and repeat the procedure until we get an optimal solution.

To demonstrate the application of this method let us take example 1 discussed in lesson 6 for consideration.
Example 1: Find optimal solution of the problem given in example 1 of lesson 6 by stepping stone method.

Solution :

The initial basic feasible solution using Least Cost Method was found in example 1 of previous lesson with the
fol owing al ocations.

The Transportation cost according to the above aal ocation is given by

Beginning at first unoccupied ce

cel (A,Q) trace a closed path and this closed path is

. Assign plus (+) and minus (-) sign alternatively on each corner cel of the closed
path traced starting a plus sign at the unoccupied ce
cel .Evaluate the cost–ef ectiveness of transporting milk on
dif erent transportation routes of each unoccupied ce
cel and given in the fol owing table:
Unoccupied cel
ce Closed Path Net Cost change (in Rs.)
(A,R) 21-14+19
14+19-18= 8
(A,S) 12-10+11
10+11-18= -5
5
(B,P) 17-16+18
16+18-19= 0
(B,S) 13-10+11
10+11-19= -5
5
(C,P) 32-16+18
16+18-11=
11= 23
(C,R) 15-11+19
11+19-14= 9

Select the unoccupied cel having the highest negative net cost change i.e. ce
cel (A,S) . The maximum number
of units that can be transported to a cel marked with a minus sign on the closed path is 10. Add this number
to the unoccupied cel and to other cel s on the path marked with a plus sign. Subtract the number from cel s
on the closed path marked with a minus sign. New solution is given in the fol
fo owing table.
 Operations Research

The transportation cost according to the above aal ocation is given by

Beginning at the first unoccupied cel (A,Q) and repeating the steps given above. Evaluate the cost–
ef ectiveness of transporting milk on dif
di erent transportation routes of each unoccupied cel given in the
folowing table:

Unoccupied ce
cel Closed Path Net Cost change (in
Rs.)
(A,Q) 18-12+10
12+10-11= 5
(A,R) 21-12+10
12+10-11+19-14=
14=
13
(B,P) 17-19+11
19+11-10+12-16=
16=
-5
(B,S) 13-10+11
10+11-19= -5
(C,P) 32-16+12
16+12-10= 18
(C,R) 15-11+19
11+19-14= 9

Select the unoccupied cel having the highest negative net cost change i.e. ce
cel (B,S) . The maximum number
of units that can be transported to a cel marked with a minus sign on the closed path is 40. Add this number
to the unoccupied cel and to other cel s on the path marked with a plus sign. Subtract the number from cel s
on the closed path marked with a minus sign. New solution is given in the fol
fo owing table.
 Operations Research

The transportation cost according to the above aal ocation is given by

Beginning at the first unoccupied cel (A,Q) and repeating the steps given above. Evaluate the cost–
ef ectiveness of transporting milk on diff erent transportation routes of each unoccupied cel and given in the
folowing table:

Unoccupied cel
ce Closed Path Net Cost change (in Rs.)

(A,Q) 18
18-12+13-19=
19= 0
(A,R) 21
21-12+13-14=
14= 8
(B,P) 17
17-16+12-13=
13= 0
(C,P) 32
32-16+12-13+14
13+14-11=
11= 18
(C,R) 15
15-11+19-14=
14= 9
(C,S) 10
10-13+19-11=
11= 5

Since al the unoccupied cel s have positive values for the net cost change, hence optimal solution has reached
and optimal cost is Rs. 5700.

MODI (Modified Distribution) Method

The MODI (Modified Distribution) method is an eff icient method of testing the optimality of a transportation
solution. In stepping stone method each of the empty cel s is evaluated for the opportunity cost by drawing a
closed loop. In situations where a large number of sources and destinations are involved, this would be a very
time consuming exercise. This method avoids this kind of extensive scanning and reduces the number of steps
required in the evaluation of the empty cel
ce s. This method al ows us to compute improvement indices quickly
for each unused square without drawing al of the closed paths. Because of this, it can often provide
considerable time savings over other methods for solving transportation problems. It provides new means of
finding the unused route with the largest negative improvement index. Once the largest index is identified, we
are required to trace only one closed path. This path helps determine the maximum number of units that can
be transported via the best unused route. Steps involved for finding out the optimal solution of transportation
 Operations Research

problem are as fol

fo ows:
1. Determine an initial basic feasible solution using any one of the five methods
discussed earlier. We start with a basic feasible solution consisting of (m + n –
1) al ocations in independent positions.
2. Determine a set of (m +n) numbers ur (r = 1, 2, 3, . . . , m) and vs (s = 1, 2, 3, . . .

,n), such that, for each occupied ce

cel (r, s)
3.
Compute the opportunity cost using .
4. Check the sign of each opportunity cost. If the opportunity costs of al
a the
unoccupied cel s are either positive or zero, the given solution is the optimum
solution. On the other hand, if one or more unoccupied cel has negative opportunity
cost, the given solution is not an optimum solution and further savings in
transportation cost are possible.
5. Select the unoccupied cel with the smal est negative opportunity cost as the cel to
be included in the next solution.
6. Draw a closed path or loop for the unoccupied cell selected in the previous step,
using the most direct route through at least three occupied cel s and moving with only
horizontal and vertical moves.
7. Assign alternate plus and minus signs at the unoccupied cel s on the corner points of
the closed path with a plus sign at the cel being evaluated.
8. Determine the maximum number of units that should be transported to this
unoccupied cel . The smal est value with a negative position on the closed path
indicates the number of units that can be transported to the entering cel . Now, add
this quantity to al the cel s on the corner points of the closed path marked with plus
signs and subtract it from those cel s marked with minus signs. In this way an
unoccupied cel becomes an occupied cel .

9. Repeat the whole procedure until an optimum solution is obtained.

Let us il ustrate this method by taking example 1 into consideration.

Example 2 : Find optimal solution of the problem given in example 1 of lesson 6 by MODI method.

Solution :

The initial basic feasible solution using Least Cost Method was found in example 1 of previous lesson with the
fol owing al ocations.
 Operations Research

The transportation cost according to the above aal ocation is given by

Step 1: Since the number of occupied cel

ce s are m+n
m+n-1=3+4-1=6
1=6 the initial solution is non –degenerate.
degenerate. Thus
optimal solution can be obtained.
Step 2: We first set up an equation for each occupied ce
cel :

Step 3: After the row and column numbers have been computed, the next step is to evaluate the opportunity
cost for each of the unoccupied cel s by using the relationship

Unoccupied cel
ce Opportunity Cost

(A,R) d13 = 21 – (0+13)= 8

(A,S) d14 = 12-
12 (0+7) = -5
(B,P) d21 = 17 – (1+16) = 0
(B,S) d24 = 13 - (1+17) = -5
(C,P) d31 = 32 – (-7+16)
7+16) = 23
(C,R) d33 = 15 – (-7+13)
7+13) = 9

Step 4: Select the unoccupied cel having the highest negative net cost change i.e. cell (A,S). The maximum
number of units that can be transported to a cel marked with a minus sign on the closed path is 10. Add this
number to the unoccupied cel
ce and to other cel s on the path marked with a plus sign. Subtract the number
from cel s on the closed path marked with a minus sign. New solution is given in the fol
fo owing table
 Operations Research

The transportation cost according to the above aal ocation is given by

To test this solution for further improvement, we recalculate the values of ur and vs based on the second

solution for each of the occupied cells.

Again, we again calculate the opportunity costs for each unoccupied ce

cel by using the relationship

Unoccupied cel
ce Opportunity Cost
(A,Q) d12 = 18-(0+13)
18 (0+13) = 5
(A,R) d13 = 21-(0+8)
21 (0+8) = 13
(B,P) d21 = 17 – (6+16) = -5
(B,S) d24 =13 – (6+12) = -5
(C,P) d31 = 32-(-2+16)
32 2+16) = 18
(C,R) d33 = 15 – (-2+8)
2+8) = 9

Since the opportunity cost in the cel (B,P) and (B,S) are negative , the current basic feasible solution is not
optimal and can be improved . Choosing the unoccupied cel (B,S), we trace a closed path that begins and
ends at this cel . The maximum number of units that can be transported to a cel marked with a minus sign on
the closed path is 40. Add this number to the unoccupied cel and to other cel s on the path marked with a
plus sign. Subtract the number from cel s on the closed path marked with a minus sign. New solution is given
in the folowing table.
 Operations Research

To test this solution for further improvement, we recalculate the values of ur and vs based on the second

solution for each of the occupied cells.

Again, we calculate the opportunity costs for each unoccupied ce

cel by using the relationship

Unoccupied cel
ce Opportunity Cost
(A,Q) d12 = 18-(0+18)
18 (0+18) = 0
(A,R) d13 = 21-(0+13)
21 (0+13) = 8
(B,P) d21 = 17 – (1+16) = 0
(C,P) d31 = 32-(-7+16)
32 7+16) = 23
(C,R) d32 = 15 – (-7+13)
7+13) = 9
(C,S) d34 = 10 – (-7+12)
7+12) = 5

In the above table the opportunity costs for each unoccupied cel
ce is positive value, we conclude that solution
is optimal one and the transportation cost is

Maximization in a Transportation Problem

Although the transportation problems which were dealt so far were of minimization, we may also come across
of transportation problems with maximization objectives. When origins are related to destinations by a profit
function instead of cost, the objective is to be maximize instead of minimize. The most convenient way to deal
with the maximization case is to transform the profits to relative costs which are carried out by subtracting the
profit associated with each ce
cel from the largest profit in the matrix. Once the optimal solution for the
minimization problem is obtained, the value of objective function is determined with reference to the original
 Operations Research

profit matrix.

Visit Our Website - www.InstantAgriculture.com 54 Download Our App - Instant Agriculture

 Operations Research

Module 4. Assignment problem

Lesson 8
INTRODUCTION AND MATHEMATICAL FORMULATION

Introduction

In earlier module, transportation problem and the technique of solving such a problem was discussed. In this
lesson, the Assignment Problem, which is a special type of transportation problem, is introduced. Here the
objective is to minimize the cost or time of completion of a number of jobs by a number of persons. In other
words, when the problem involves the al ocation of n dif erent facilities to n dif erent tasks, it is often termed
as Assignment Problem. The assignment problem deals in al ocating the various origins (jobs) to equal number
of destinations (persons) on a one to one basis in such a way that the resultant ef ectiveness is optimized
(minimum cost or maximum profit). This is useful in solving problems such as assigning men to dif erent
operations in a milk plant, milk tankers to delivery routes, machine operators to machines, jobs to persons in
a dairy plant etc.

Definition of Assignment Problem

Assignment problem is special class of the transportation problem in which the supply in each row represents
the availability of a resource such as man, vehicle, product and demand in each column represents dif erent
activities to be performed, such as jobs, routes, milk plants respectively is required. The name ‘Assignment
Problem ’originates from the classical problem where the objective is to assign a number of origins (jobs) to
equal number of destinations(persons) at a minimum cost (or Maximum profit). Suppose there are n jobs to
be performed and n persons are available for doing these jobs. Assume that each person can do each job at
a time, though with varying degree of ef iciency. Let Cij be the cost if ith person is assigned the jth job, the
problem is to find an assignment so that the total cost for performing al jobs is minimum. One of the important
characteristics of assignment problem is that only one job (or worker) is assigned to one machine (or
project). Hence, the number of sources are equal to the number of destinations and each requirement and
capacity value is exactly one unit.

The assignment problem can be stated in the form n x n cost matrix [Cij] of real number as given below

Sources Jobs
(Milk plants)
J1 J2 …… …… Jn
Jj
P1 C11 C12 …… C1j …… C1n

P2 C21 C22 …… C2j …… C2n

: : : …… : …… :

Pi Ci1 Ci2 …… Cij …… Cin

: : : …… : …… :

Visit Our Website - www.InstantAgriculture.com 55 Download Our App - Instant Agriculture

 Operations Research

Pn Cn1 Cn2 …… Cnj …… Cnn

Formulation of an Assignment Problem

Let us consider the case of a milk plant which has three jobs to be done on the three available machines.
Each machine is capable of doing any of the three jobs. For each job the cost depends on the machine to
which it is assigned. Costs incurred by doing various jobs on dif
di erent machines are given below

Machine
Job
I II III
A 7 8 6
B 5 4 9
C 2 5 6

The problem of assigning jobs to machines, one to each, so as to minimize total cost of doing al the jobs, is
an assignment problem. Each job machine combination which associates al jobs to machines on one –to-one
basis is cal ed an assignment. In the above example let us write al
a the possible assignments

Number Assignment Total Cost

1 Job A-Machine
Machine I, Job B –Machine
Machine II, Job C-Machine
Machine III 7+8+6=21
2 Job A-Machine
Machine I, Job B –Machine
Machine III, Job C--Machine II 7+9+5=21
3 Job A-Machine
Machine II, Job B –Machine
Machine III, Job C-Machine
C I 8+9+2=19
4 Job A-Machine
Machine II, Job B –Machine
Machine I, Job C-Machine
Machine III 8+5+6=19
5 Job A-Machine
Machine III, Job B –Machine
Machine I, Job C--Machine II 6+5+2=13
6 Job A-Machine
Machine III, Job B –Machine
Machine II, Job C-Machine
C I 6+4+2=12

As per the above assignment, the assignment number 6 having total cost 12 is minimum therefore needs to be
selected. But selecting assignment in this manner is quite time consuming.

Mathematical Formulation of Assignment Problem

Using the notations described above, the assignment problem consist of finding the values of Xij in order to
minimize the total cost

Subject to restrictions
 Operations Research

where Xij denotes the jth job to be assigned to the ith person. An assignment problem could thus be solved
by Simplex Method.

We state below, the fol owing theorems which have potential applications in finding out of the optimal solution
for assignment problems:

Theorem 1: Reduction Theorem

It states that in an assignment problem, if we add or subtract a constant to every element of any line (row or
column) of the cost matrix [Cij], then an assignment that minimizes the total cost on one matrix also minimizes
the total cost on the other matrix.

Theorem 2: In an assignment problem with cost (Cij), if al Cij≥ 0 then a feasible solution (Xij) which

satisfies,
is optimal for the problem
Remarks:
There are situations when a particular assignment may not be permissible. In such situations assign a very high
cost (say M) for such an assignment and proceed as usual.

1.
1 If the assignment problem involves maximization, convert the ef ective matrix to an opportunities loss
matrix by subtracting each element from the highest element of the matrix. Minimization of the
resulting matrix is the same as the maximization of the original matrix.

Similarity of Assignment Problem to Transportation Problem

The assignment problem is a particular case of transportation problem in which a number of operations are to
be assigned to an equal number of operators, where each operator performs only one operation. The
objective is to maximize overal
overa profit or minimize overal
overa cost for a given assignment schedule.

The Assignment Problem is a special case of the transportation problem in which m=n, Al
A ai and bj are unity

i.e., The availability and requirement at ith origin and jth destination are unity, and each Xij is limited to one of
the two values 0 and 1. Under these circumstances, exactly n of Xij can be non-zero (i.e., unity), one in each
row of the table and one in each column.

An assignment problem is a completely degenerate form of a transportation problem. The units available at
each origin and units demanded at each destination are al equal to one. This means exactly one occupied cel
in each row and each column of the transportation table. i.e., only n occupied cel s in place of the required (n
+ n – 1) = (2n – 1).
 Operations Research

Module 4. Assignment problem

Lesson 9
SOLUTION OF ASSIGNMENT PROBLEM

Introduction

Although assignment problem can be solved either by using the techniques of Linear Programming or by the
transportation method yet the assignment method developed by D. Konig, a Hungarian mathematician known
as the Hungarian method of assignment problem is much faster and ef icient. In order to use this method, one
needs to know only the cost of making al the possible assignments. Each assignment problem has a matrix
(table) associated with it. Normal y, the objects (or people) one wishes to assign are expressed in rows,
whereas the columns represent the tasks (or things) assigned to them. The number in the table would then be
the costs associated with each particular assignment. It may be noted that the assignment problem is a
variation of transportation problem with two characteristics firstly the cost matrix is a square matrix and
secondly the optimum solution for the problem would be such that there would be only one assignment in a
row or column of the cost matrix.

Solution of Assignment Problem

The assignment problem can be solved by the following four methods:
a) Complete enumeration method
b) Simplex Method
c) Transportation method
d) Hungarian method

Complete enumeration method

In this method, a list of al possible assignments among the given resources and activities is prepared. Then an
assignment involving the minimum cost, time or distance or maximum profits is selected. If two or more
assignments have the same minimum cost, time or distance, the problem has multiple optimal solutions. This
method can be used only if the number of assignments is less. It becomes unsuitable for manual calculations if
number of assignments is large

Simplex Method

This can be solved as a linear programming problem as discussed in section 8.1.3 of the last lesson and as
such can be solved by the simplex algorithm.

Transportation method

As assignment is a special case of transportation problem, it can also be solved using transportation model
discussed in module 3. The solution obtained by applying this method would be degenerate. This is because
the optimality test in the transportation method requires that there must be m+n-1= (2n-1) basic variables.
For an assignment problem of order n x n there would be only n basic variables in the solution because here n
assignments are required to be made. This degeneracy problem of solution makes the transportation method
computational y inef icient for solving the assignment problem.

Visit Our Website - www.InstantAgriculture.com 58 Download Our App - Instant Agriculture

 Operations Research

Hungarian assignment method

The Hungarian method of assignment provides us with an ef icient means of finding the optimal solution. The
Hungarian method is based upon the fol owing principles:
(i) If a constant is added to every element of a row and/or column of the cost matrix of an assignment
problem the resulting assignment problem has the same optimum solution as the original problem or
vice versa.
(i ) The solution having zero total cost is considered as optimum solution.
Hungarian method of assignment problem (minimization case) can be summarized in the fol owing steps:

Step I: Subtract the minimum cost of each row of the cost (ef ectiveness) matrix from al the elements of the
respective row so as to get first reduced matrix.

Step II: Similarly subtract the minimum cost of each column of the cost matrix from al the elements of the
respective column of the first reduced matrix. This is first modified matrix.

Step III: Starting with row 1 of the first modified matrix, examine the rows one by one until a row
containing exactly single zero elements is found. Make any assignment by making that zero in or enclose the
zero inside a. Then cross (X) al other zeros in the column in which the assignment was made. This eliminates
the possibility of making further assignments in that column.

Step IV: When the set of rows have been completely examined, an identical procedure is applied
successively to columns that is examine columns one by one until a column containing exactly single zero
element is found. Then make an experimental assignment in that position and cross other zeros in the row in
which the assignment has been made.

Step V: Continue these successive operations on rows and columns until al zeros have been either assigned
or crossed out and there is exactly one assignment in each row and in each column. In such case optimal
assignment for the given problem is obtained.

Step VI: There may be some rows (or columns) without assignment i.e. the total number of marked zeros is
less than the order of the matrix. In such case proceed to step VII

Step VII: Draw the least possible number of horizontal and vertical lines to cover al zeros of the starting
table. This can be done as fol ows:

1. Mark (√) in the rows in which assignments has not been made.
2. Mark column with (√) which have zeros in the marked rows.
3. Mark rows with (√) which contains assignment in the marked column.
4. Repeat 2 and 3 until the chain of marking is completed.
5. Draw straight lines through marked columns.
6. Draw straight lines through unmarked rows.
By this way we draw the minimum number of horizontal and vertical lines necessary to cover al zeros at least
once. It should, however, be observed that in al n x n matrices less than n lines wil cover the zeros only when
there is no solution among them. Conversely, if the minimum number of lines is n, there is a solution.
Step VIII: In this step, we

Visit Our Website - www.InstantAgriculture.com 59 Download Our App - Instant Agriculture

 Operations Research

1. Select the smal est element, say X, among al the not covered by any of the lines of the table; and
2. Subtract this value X from al of the elements in the matrix not covered by lines and add X to al those
elements that lie at the intersection of the horizontal and vertical lines, thus obtaining the second
modified cost matrix.
Step IX: Repeat Steps IV, V and VI until we get the number of lines equal to the order of matrix I, til an
optimum solution is attained.
Step X: We now have exactly one encircled zero in each row and each column of the cost matrix. The
assignment schedule corresponding to these zeros is the optimum assignment. The above technique is
explained by taking the fol owing examples

Example 1:

A plant manager has four subordinates, and four tasks to be performed. The subordinates dif er in ef iciency
and the tasks dif er in their intrinsic dif iculty. This estimate of the times each man would take to perform each
task is given in the ef ectiveness matrix below.

I II III IV
A 8 26 17 11
B 13 28 4 26
C 38 19 18 15
D 19 26 24 10

How should the tasks be al ocated, one to a man, so as to minimize the total man hours?

Solution:

Step I : Subtracting the smal est element in each row from every element in that row, we get the first reduced
matrix.

0 18 9 3
9 24 0 22
23 4 3 0
9 16 14 0

Step II: Next, we subtract the smal est element in each column from every element in that column; we get the
second reduced matrix.
Step III: Now we test whether it is possible to make an assignment using only zero distances.
0 14 9 3
9 20 0 22
23 0 3 0
9 12 14 0

Visit Our Website - www.InstantAgriculture.com 60 Download Our App - Instant Agriculture

 Operations Research

(a) Starting with row 1 of the matrix, we examine rows one by one until a row containing exactly single
zero elements are found. We make an experimental assignment (indicated by) to that cel . Then we
cross al other zeros in the column in which the assignment was made.
(b) When the set of rows has been completely examined an identical procedure is applied successively to
columns. Starting with Column 1, we examine columns until a column containing exactly one
remaining zero is found. We make an experimental assignment in that position and cross other zeros
in the row in which the assignment was made. It is found that no additional assignments are possible.
Thus, we have the complete ‘Zero assignment’,
A – I, B – III, C – II, D – IV

The minimum total man hours are computed as

Optimal assignment Man hours

A–I 8
B – III 4
C – II 19
D – IV 10
Total 41 hours

Example 2:

A dairy plant has five milk tankers I, II, III, IV & V. These milk tankers are to be used on five delivery routes
A, B, C, D, and E. The distances (in kms) between dairy plant and the delivery routes are given in the
fol owing distance matrix

I II III IV V
A 160 130 175 190 200
B 135 120 130 160 175
C 140 110 155 170 185
D 50 50 80 80 110
E 55 35 70 80 105

How the milk tankers should be assigned to the chil ing centers so as to minimize the distance travel ed?

Solution :
Step I: Subtracting minimum element in each row we get the first reduced matrix as

30 0 45 60 70
15 0 10 40 55
30 0 45 60 75
0 0 30 30 60
Visit Our Website - www.InstantAgriculture.com 61 Download Our App - Instant Agriculture
 Operations Research

20 0 35 45 70

Step II: Subtracting minimum element in each column we get the second reduced matrix as
30 0 35 30 15
15 0 0 10 0
30 0 35 30 20
0 0 20 0 5
20 0 25 15 15

Step III: Row 1 has a single zero in column 2. We make an assignment by putting ‘ ‘around it and delete
other zeros in column 2 by marking ‘X’. Now column1 has a single zero in column 4 we make an assignment by
putting ‘ ‘and cross the other zero which is not yet crossed. Column 3 has a single zero in row 2; we make an
assignment and delete the other zero which is uncrossed. Now we see that there are no remaining zeros; and
row 3, row 5 and column 4 has no assignment. Therefore, we cannot get our desired solution at this stage.

Step IV: Draw the minimum number of horizontal and vertical lines necessary to cover al zeros at least once
by using the fol owing procedure
1. Mark (√) row 3 and row 5 as having no assignments and column 2 as having zeros in rows 3 and 5.
2. Next we mark (√) row 2 because this row contains assignment in marked column 2. No further rows
or columns wil be required to mark during this procedure.
3. Draw line L1 through marked col.2.
4. Draw lines L2 & L3 through unmarked rows.
Step V: Select the smal est element say X among al uncovered elements which is X = 15. Subtract this value
X=15 from al of the values in the matrix not covered by lines and add X to al those values that lie at the
intersections of the lines L1, L2 & L3.
Applying these two rules, we get a new matrix

15 0 20 15 0
15 15 0 10 0
15 0 20 15 5
0 15 20 0 5

Visit Our Website - www.InstantAgriculture.com 62 Download Our App - Instant Agriculture

 Operations Research

5 0 10 0 0
Step VI: Now reapply the test of Step III to obtain the desired solution.

The assignments are

Total Distance 200 + 130 + 110 + 50 + 80 = 570

Visit Our Website - www.InstantAgriculture.com 63 Download Our App - Instant Agriculture

 Operations Research

Module 5. Inventory control

Lesson 10
INTRODUCTION AND GENERAL NOTATIONS

Introduction

Inventory, in broad sense, is a stock of physical assets having some economic value, which can be either in
the form of material, money or men and is defined as any idle resource of an enterprise. It is a physical stock
of goods kept for the future purposes. The term is general y used to indicate raw material, work-in-progress
(intermediate good), finished goods, packaging material and other stock in order to meet an expected
demand or distribution in future as wel as day to day functioning of any organization. Though inventory of
materials is an idle resource, it is not meant for immediate use, it is always essential to maintain inventory of
goods for smooth functioning of an enterprise. For example, suppose a milk plant does not have Inventory of
materials at al . If this plant receives a sales order, then
i) It wil have to order out raw material required to complete the order.
ii) It has to wait til the material arrives and then start production.
iii) Customers have to wait too long for the delivery of finishedgoods.
iv) The plant may have to purchase raw materials at very high prices which wil increase the cost of
production and decrease the margin of profit.

Centuries ago, inventories were viewed as measures of the wealth and power of a country or of an individual.
Now-a-days, inventories are viewed as a large potential risk rather than as a measure of wealth due to the
fast developments and changes in product life, therefore, it has necessitated the use of scientific techniques in
the management of inventories known as inventory control. It is the technique of maintaining stock-items at
desired levels. Dif erent departments within the same organization adopt dif erent attitudes towards inventory.
This is mainly because the particular functions performed by a department influence the department’s
motivation. For example, the sales department might desire large stock in reserve to meet virtual y every
demand that comes. The production department similarly would ask for large stocks of materials so that the
production system runs uninterrupted. On the other hand, the finance department would always argue for a
minimum investment in stocks so that the funds could be used elsewhere for other better purpose.

Inventory (Production Management)

The study or function of directing the movements of goods through the entire manufacturing cycle from the
raw material to the inventory of finished goods orderly mannered to meet the objective of maximum
customers service with minimum investment and ef icient plant operation. This process is known as inventory
control. Inventory may be defined as stock of goods, commodities or other economic resources that are
stored or reserved in order to ensure smooth and ef icient running of business. Inventory, for example may
include raw material which are kept in stock for in the production of goods (raw material inventory), Semi
finished goods or goods in process which are stored during the production process (Work in progress
inventory), finished goods awaiting shipment from plant, wholesaler etc. (finished goods inventory). Inventory
also includes furniture, machinery fixtures etc. The term inventory is general y classified into two categories:

Visit Our Website - www.InstantAgriculture.com 64 Download Our App - Instant Agriculture

 Operations Research

Direct inventory

Direct inventories play a direct role in the manufacturing and become a bigger part of finished goods. They
are further classified into three groups:

i) Raw material inventories are provided :

a) for economical bulk purchasing.
b) to enable production rate changes.
c) to provide production buf er againsttransportation.
d) seasonal fluctuations.
ii) Work in progress inventories are provided:
a) to enable economical lot production.
b) to consider the variety of production for replacement of base.
c) to maintain uniform production even if amount of sales may vary.
iii) Finished goods inventories are provided:
a) for maintaining of selfdelivery.
b) to al ow stabilization of production level.
c) for sales promotion.

Indirect inventory

They include those items which are necessary for manufacturing but do not become component of finished
production, such as oil, grease, petrol, lubricant, of ice material, maintenance material etc.

Types of Inventory

Fluctuation inventory

These have to be carried because sales and production time can’t be predicted accurately. There is
fluctuation in the demand and lead times that af ect the production of items such type of results stock or safety
stock are cal ed fluctuation inventory.

Anticipation inventory

These are built in advance for the season of large sale a promotional programme or a plant shut down period.
In this inventory are store for future requirement.

Cycle or lot size inventory

In practical situations it seldom happens that the rate of consumption is same as rate of production so the
items are purchased in large quantity than they are required. This results in cycle or lot size inventory.

Transportation inventory

Such inventory exists because the material is required to move from one place to another place when the
transportation time is long and the items under transport can’t be served to customer. Therefore these
inventories exist solely because of time.

Decoupling inventory

Visit Our Website - www.InstantAgriculture.com 65 Download Our App - Instant Agriculture

 Operations Research

Such inventory is needed for meeting of demand for decoupling the dif erent parts of the production system.

Inventory Decisions

The fol owing are the decisions made for every item of inventory.

i) How much amount of an item should be ordered when the inventory of that item is to be replenished.
i ) When to replenish the inventory of that item

How to Develop Inventory Control Model?

i) First take physical stock of al inventory items in an inventory organization.

ii) Second, classify al items in various categories.
iii) Each of above classification may be further divided into dif erent groups.
iv) After classification of inventories, each item should be assigned a suitable code.
v) Since the number of items is very large, separate inventory management model should be developed
for each category of items A-B-C or V-E-D classification.
vi) Estimates annual demand of each inventory item and their prevailing prices.
vii) Estimate lead time, safety stock and reorder level, if supply is not instantaneous.
viii) Develop the inventory mode.
ix) Final y, review the position and make suitable alterations, if required due to current situations or
constraints.

Costs Involved in Inventory Control Models

Holding cost

Costs associated with carrying or holding goods in stock is known as carrying or holding cost which is
denoted by C1 or Ch per unit of goods for a unit of time, respectively. However cost is assumed to be
varying directly the size of inventory as wel as the time for which the item is in stock. The fol owing
components constitute the holding cost.
i) Invested capital costs (Interest charged on capital investment):- This is the most important
component so its rate of interest should be investigated careful y.
ii) Record keeping and administrative costs: - This signifies the need of keeping funds for
maintaining of records and necessary administration.
iii) Handling cost:- It includes al costs associated with movement of stock such as cost of labour,
overhead and other machinery required for this purpose.
iv) Storage cost: - This involves the rent of storage space or depreciation in interest even if own space
is used.
v) Depreciation, deterioration or obsolescence cost: - Such cost is due to the itemin stock being
out of fashion or the item undergoes chemical changes during storage (e.g. rusting of iron), breakage and
spoilage.
vi) Insurance cost – The amount of money paid in the form of insurance premium is known as
insurance cost. This cost is also a component of stock holdingcost.
vii) Purchase Cost or Production Cost: - Purchase price per unit item is af ected by the purchase due

Visit Our Website - www.InstantAgriculture.com 66 Download Our App - Instant Agriculture

 Operations Research

to discount or price break. Production cost per unit item depends upon length of production time.
viii) Selling price or Salvage Cost: - When the demand for an item is af ected by its quantities in stock,
the decision model of the problem depends upon profit maximum and in revenue (sales tax etc) from the
sale of the items. General y salvage cost is combined with storage cost and not considered
independently.
Carrying cost can be determined as
Carrying cost= (Cost of carrying one unit of item for a given length of time) x (Average number of
units carried for a given length of time)

Shortage cost or stock out cost

These are penalty costs that are incurred as a result of either delay in meeting the demand or inability to meet
it at al due to running out of stock are known as shortage or stock out cost. It is denoted as C2 or Cs per
unit of goods for specified period. This cost arises due to shortage of goods, sales may be lost, good wil may
be lost either by delay in meeting the demands or being quite unable to meet the demand.

Setup cost

These include the fixed cost associated with obtaining goods through placing of an order or purchasing or
manufacturing or setting up of a machine before start of production. So they include cost of purchase,
requisition, quality control etc. These are also known as order cost or replenishment cost. It is denoted by
C3.

Objective of Inventory Control

i) It helps in smooth and ef icient running of business.

ii) It provides service to the customers immediately at a short notice.
iii) Due to absence of stock the company has to pay high price because of piecewise purchasing.
iv) It also acts as a buf er stock when raw material is received late or when orders are likely to be
rejected.
v) It reduces product cost.
vi) It helps to maintain the economy by absorbing some of the fluctuations and the demand for an item
fluctuates or is seasonal.

Visit Our Website - www.InstantAgriculture.com 67 Download Our App - Instant Agriculture

 Operations Research

Module 3. Inventory control

Lesson 11
ECONOMIC LOT SIZE MODELS WITH KNOWN DEMAND

Introduction

In the last lesson we have seen that maintenance of proper Inventory Control System helps in keeping the
investment in Inventories as low as possible and yet (i) ensures availability of materials by providing adequate
protection against supply uncertainty and consumption of materials and (i ) al ows ful advantages of
economies of bulk purchases and transportation costs. The basic Inventory Control problem therefore lies in
determining firstly when should an order for materials be placed and secondly how much should be produced
at the beginning of each time interval or what quantity of an item should be ordered each time. In this lesson
we wil learn how to develop inventory model.

Variables in Inventory Problem

The variables used in any inventory model are of two types: Control ed and Uncontrol ed variables

Controlled variables

The fol owing are the variables that may be considered separately or in combination:
 How much quantity acquired
 The frequency or timing of acquisition .How often or when to replenish the inventory?
 The completion stage of stocked items.

Uncontrolled variables

The fol owing are the principal variables that may be control ed:
 The holding costs, shortage or penalty cost, set up costs.
 Demand: It is the number of units required per period and may be either known exactly or is known
in terms of probabilities or is completely unknown. Further if the demand is known, it may be either
fixed or variable per unit of time. The model which has fixed demand is known as deterministic
model.

 Lead Time: This is the time of placing an order and its arrival in stock as shown in Fig. 11.1. If the
lead time is known and not equal to zero and if demand is deterministic then one should order in
advance by an amount of time equal to lead time. If the lead time is zero then there is no need to order
in advance. If lead time is variable then it is known as probabilistical y.

Visit Our Website - www.InstantAgriculture.com 68 Download Our App - Instant Agriculture

 Operations Research

Fig 11.1 Inventory with constant demand rate and constant lead time

 Amount delivered (supply of goods) : The supply of goods may be instantaneous or spread over a
period of time .If a quantity q is ordered for purchase, the amount delivered may very around g with
known probability densityfunction .

Symbols used in Inventory Models

Holding cost per item per unit time

Shortage cost per unit quantity per unit time for the back-log
back case and per unit only for no back log
case.
Set up cost per production run
Demand Rate
Production Rate
Scheduling time period which is not prescribed.

Scheduling time period (prescribed)

Order Level (or stock level)
Total Demand
Quantity already present in the beginning
Lead time
Classification of Inventory Models

The inventory models are broadly classified as either deterministic (variables are known with certainty) or
stochastic (variables are probabilistic). The deterministic models are furth
further divided into Static demand
models and dynamic demand models. In this lesson we wil
wi discuss deterministic inventory model i.e.
economic lot size system with uniform demand.

Economic Order Quantity (EOQ) Model

This concept was developed by F. Haris in 1916. The concept is as lot size (q) increases the carrying charges
(C1) wil increase while the ordering cost (C3) wil decrease. On the other hand as the lot size (q) decreases,
 Operations Research

the carrying cost (C1) wil decrease but the ordering costs wil increase. Economic Ordering Quantity (EOQ)
is the size of order which minimizes total annual cost of carrying inventory and cost of ordering under the
assumed conditions of certainly and annual demands are known.

The economic lot size system with uniform demand

In this model we want to derive an Economic Lot Size Formula for the optimum production quantity q per
cycle (per production run) of a single product so as to minimize the total average variable cost per unit time.
The model is based on the fol
fo owing basic assumptions:
i)
i the demand for the item is uniform at the rate of R quantity units per unit time
ii)
ii the lead time is known and fixed. Thus, when the lead time is zero, the delivery of item is
instantaneous.
iii)
iii production rate is infinite i.e. production is instantaneous
iv)
iv shortages are not al owed
v)
v holding cost is
in per quantity unit per unit time
vi)
vi set up cost is per set up.
This can be solved by two fo
fol owing methods

Algebraic method

The algebraic method is based on the fo

fol owing principle:
Inventory carrying costs=Annual ordering (set up) costs
(11.1)
Since the demand is uniform and known exactly and supply is instantaneous, the reorder point is that when
inventory fal s to zero.
Average Inventory =1/2(maximum level +minimum level) =(q+0)/2=q/2

Total inventory carrying costs are determined by using the fol owing formula:

= × ×
=1/2q×C×I=1/2qC1 (11.2)

where C1=CI is holding or carrying cost per unit for unit time.
Total annual ordering costs are obtained as fol ows:

×
=(R/q) ×C3=(R/q)C3 (11.3)

Now summing up the total inventory carrying cost and total ordering cost we get total inventory cost as

C(q)= 1/2qC1+(R/q)C3 , this is cost equation . (11.4)

 Operations Research

The total inventory cost C(q) is minimum when the inventory carrying costs become equal to the total
ordering costs. Therefore

1/2qC1=(R/q)C3 or
(11.5)

Or optimal

To find the minimum of total inventory cost C(q) , we substitute the value of q from (11.5) in cost equation
(11.4) we get

(11.6)

Optimum inventory cost (Cmin) =

To obtain the optimum interval of ordering (t*) we have

q=R × t (11.7)

(11.8)
Calculus method

Let each production cycle be made at fixed interval ‘t’ and therefore the quantity q already present in the
beginning (when the business was started) should be

(11.9)
where R is the demand rate.
Since the stock in smal time wil be , therefore the stock in total time t wil be

=Area of the inventory as shown in Fig. 11.2

 Operations Research

Fig. 11.2 Inventory profile of EOQ model

q=Rt

The rate of replenishment = slope of line OP

Thus the cost of holiday inventory is

(11.10)
Set up cost is
per production run for interval t. (11.11)

Total cost C(t) is summing up the costs in (11.10) and (11.11) and dividing by t we get the average total cost
given by ;

(11.12)
The condition for minimum or maximum of C(t) is

(11.13)

(11.14)

(11.15)
which is known as optimal lot size formula.
 Operations Research

Therefore

per unit time is obtained by putting in .The above procedure is il ustrated through
fol owing examples
Example 1: A manufacture has to supply his customer with 600 units of a product per year, shortages are
not al
a owed and the storage costs amounts to 0.60 per unit per year. The set up cost per run is Rs. 80.00.
Find the optimum run size and the minimum average yearly cost.
Solution :
In this the demand rate ( R )=600 per unit per year

Holding cost per item per unit time=Rs. 0.60 per unit per year

Set up cost per production run=Rs. 80 per production run

Optimum lot size is

=0.67 years =8 months

Thus the manufacturer should produce 400 units of his product at an interval of 8 months.
Example 2 : A company uses 3000 units of a product, its carrying cost is 30%of average inventory.
Ordering cost is Rs. 100 per order. Unit cost is Rs. 20.Calculate EOQ and the total cost .
Solution :

Total Demand=3000 units

carrying cost =30% of Rs. 20=Rs.6
Ordering cost =Rs.100

Optimum lot size is

The total cost is equal to
Total cost=Material cost +Total variable cost
=(3000 × 20) + = 60000 + 1897.36 = Rs.61897.36
Limitations of EOQ Formulae

In spite of several assumptions made in the derivation of above EOQ formulae, fol owing are the limitations
while considering applications of these formulae:
Operations Research

i) In the EOQ model we assumed that the demand for the item under consideration is constant, while in
practicalsituations demand is neither known with certainty nor it is uniform.

i) ) It is dif icult to measure the ordering cost and also it is not linearly related to number of orders.

ii) ) In EOQ model it is assumed that the annual demand can be estimated in advance which is just a
guess in practice.

iv) In EOQ model it is assumed that the entire inventory which is ordered arrives simultaneously. In many
situations it may not be true.

v) In EOQ, it is assumed that the demand is uniform; which may not hold in practice.

vi) In EOQ model, the replenishment time is assumed to be zero which is not possible in real life always.

Visit Our Website - www.InstantAgriculture.com 74 Download Our App - Instant Agriculture

 Operations Research

Module 6. Replacement theory

Lesson 12
INTRODUCTION AND ELEMENTARY CONCEPTS

Introduction

The replacement problems are concerned with the situations that arise when some items such as machines,
men, electric appliance etc. need replacement due to their decreased ef iciency, failure or breakdown. The
deteriorating ef iciency or complete breakdown may be either gradual or al of a sudden. A replacement is
cal ed for whenever new equipment of ers more ef icient or economical service than the existing one. The
problem in such situation is to determine the best policy to be adopted with respect to replacement of the
equipment. In case of items whose ef iciency go on decreasing according to their age, we have to spend more
and more money on account of increased operating cost, increased repair cost, increased scrap, etc. In such
cases the replacement of an old item with a new one is the only alternative to prevent such increased
expenses. Thus, it becomes necessary to determine the age at which replacement is more economical rather
than continuing with the same.

Types of Replacement Situations

The replacement situations may be classified into four categories:

a) Replacement of items that become worse with time e.g. milk plant machinery, tools, vehicles,
equipment etc.
b) Replacement of items which do not deteriorate with time but break down completely after certain
usage e.g. electric tubes, machinery parts etc.
c) Replacement of items that becomes obsolete due to new developments.
d) The existing working staf in an organization gradual y reduces due to death, retirement and other
reasons.
The problem is to decide the best policy to adopt with regard to replacement. The need for replacement
arises in a number of dif erent situations so that dif erent types of decisions may have to be taken. For
example:
a) It may be necessary to decide whether to wait for certain items to fail, which might cause some loss, or
to replace the same in advance, even at a higher cost.
b) An item can be considered individual y to decide whether or not to replace immediately.
c) It is necessary to decide whether to replace by the same item or by an improved type of item.

Types of Failure

There are two types of failure: i) Gradual failure i ) Sudden failure

Gradual failure

It means slow or progressive failure as the life of the item increases, its ef iciency decreases resulting in
decreased productivity, increased operating cost and decrease in the value of the item, e.g.
machines/equipment etc.

Visit Our Website - www.InstantAgriculture.com 75 Download Our App - Instant Agriculture

 Operations Research

Sudden failure

In this type of failure the items do not deteriorate markedly with service but which ultimately fail after some
period of usage, thus precipitating cost of failure. Sometimes sudden failure of an item may cause loss of
production or may also account for damaged or faulty products. The period between instal ation and failure is
not constant for any particular type of equipment but wil fol ow some probability distribution which may be
progressive, retrogressive or random in nature.

Progressive failure

Under this mechanism, the probability of failure increases with the increase in the life of an item.

Retrogressive failure

Certain items have more probability of failure in the beginning of their life and as time passes, the chances of
failure become less. In other words, the ability of the unit to survive the initial period of life increases its
expected life.

Random failure

Under this mechanism, constant probability of failure is associated with items that fail from random causes
such as physical shocks, not related to age.

Assumptions

Fol owing assumptions are essential y required for replacement decisions:

i) The quality of the output remains constant.
i) ) Replacement and maintenance costs remain constant.
ii) ) The operational ef iciency of the equipment remains constant.
iv) There is no change in technology of the asset under consideration.

OR Methodology of Solving Replacement Problem

OR provides a methodology for tackling replacement problem which is discussed below:

i) Identify the items to be replaced and also their failure mechanism.
i) ) Col ect the data relating to the depreciation cost and the maintenance cost for the items which fol
ow gradual failure mechanism. In case of sudden failure of items, col ect the data for replacement
cost of the failed items.
ii) ) Select a suitable replacement model as discussed in Lesson 13.

Visit Our Website - www.InstantAgriculture.com 76 Download Our App - Instant Agriculture

 Operations Research

Module 6. Replacement theory

Lesson 13
REPLACEMENT OF ITEMS DETERIORATING WITH TIME

Introduction
In any establishment, sooner or later equipment needs to be replaced, particularly when new equipment gives more efficient or
economical service than the old one. In some cases, the old equipment might fail and work no more or is worn out. In such situations it
needs more expenditure on its maintenance than before. The problem in such situation is to determine the best policy to be adopted
with respect to replacement of the equipment. The replacement theory provides answer to this question in terms of optimal
replacement period. Replacement theory deals with the analysis of materials and machines which deteriorate with time and fix the
optimal
al time of their replacement so that total cost is the minimum.

Replacement Decisions
The problem is to decide the best policy to adopt with regard to replacement. The need for replacement arises in a number of different
following situations so that different types of decisions may have to be taken.
 It may be necessary to decide whether to wait for a certain item to fail which might cause some loss or to replace earlier at the
expense of higher cost of the item.
 The item can be considered individually to decide whether to replace now or if not when to reconsider the item in question.
 It is necessary to decide whether to replace by the same item or by a different type of item.

Types of Replacement Problems

i) Replacement policy for items, efficiency of which declines gradually with time without change in money value.
ii) Replacement policy for items, efficiency of which declines gradually with time but with change in money value.
iii)) Replacement policy of items breaking down suddenly
a) Individual replacement policy
b) Group replacement policy
iv)) Staff replacement
In this lesson we confine ourselves to first two situations only

Replacement of Items that Deteriorate with Time

There are certain items which deteriorate gradually with usage and such items decline in efficiency over a period of time. Generally, the
maintenance cost of certain items always increase gradually with time and a stage comes when the maintenance cost becomes so large
that it is better and economical to replace the item with a new one. There may be number of alternatives and we may have a
comparison between various alternatives by considering the costs due to waste, scrap, loss of output, damage to equipment and safety
risks etc.

Replacement of items whose maintenance cost increases with time and the value of money remains same
during the period
The following costs are considered in such decisions:
C: Capital cost of a certain item say a machine
s(t): The selling or scrap value of the item after t years
f(t): Operating (or maintenance) cost of the item at time t
n : Optimal replacement period of the item
The operating cost function f(t) is assumed to be strictly positive. It may be continuous or discrete.

When t is a continuous variable

Now the annual cost of the machine at time t is given by and since the total maintenance cost incurred on the
machine during n years is
Total cost T incurred on machine during n years is given by
 Operations Research

Thus the average annual cost incurred on the machine per year during n years is given by

To determine the value of optimal period (n), the principle of minima will be employed.

Clearly

Therefore, it can be seen that is a minimum for T provided that is non–decreasing

decreasing . Hence, if
time is measured continuously, then the averageannual cost will be minimized by replacing the item when the average cost becomes
equal to the current maintenance cost.

When time is a discrete variable

Here the period of time is considered as fixed and n takes values then

By using finite differences, A(n) will be a minimum for that value of n for which

or
For this, we write

Similarly,it can be shown

This suggests the replacement policy that if time is measured in discrete units, then the average annual cost will be minimized by
replacing item when the next period’s maintenance cost become greater than the current average cost. Hence, replace the equipment
at the end of n years, if the maintenance cost in the (n+1)th year is more than the average total cost in the (n)th year and the (n)th
year’s maintenance cost is less than the previous year’s average total cost. The following examples will illustrate this methodology
Example 1: A milk plant is considering replacement of a machine whose cost price is Rs. 12,200 and the scrap value Rs. 200. The
running (maintenance and operating) costs in Rs. are found from experience to be as follows:
 Operations Research

Year:
Running Cost:
When should the machine be replaced?

Solution: The computations can be summarized in the following tabular form:

Table 13.1 Calculations for average cost of machine

(In Rupees)
Year Running Cumulative Depreciation Total Cost Average
Cost Running Cost TC Cost
(5) = (3) + (4)
Cost
(1) (2) (3) (4) (6) = (5)/(1)
1 200 200 12000 12200 12200
2 500 700 12000 12700 6350
3 800 1500 12000 13500 4500
4 1200 2700 12000 14700 3675
5 1800 4500 12000 16500 3300
6 2500 7000 12000 19000 3167
7 3200 10200 12000 22200 3171
8 4000 14200 12000 26200 3275

From the table it is noted that the average total cost per year, A(n) is minimum in the 6th year (Rs. 3167). Also the average cost in 7th
year (Rs.3171) is more than the cost in 6th year. Hence the machine should be replaced after every 6 years.

Example 2:
A Machine owner finds from his past records that the maintenance costs per year of a machine whose purchase price is Rs. 8000 are
as given below:
Year:
Maintenance Cost:

Resale Price:
Determine at which time it is profitable to replace the machine.

Solution .Table 13.2 shows the average cost per year during the life of machine. Here The computations can be
summarized in the following tabular form:

Table 13.2 Calculations for average cost of machine

f(t) Cumulative Scrap value Total cost
Year
maintenance
cost
1 1000 1000 4000 5000 5000
2 1300 2300 2000 8300 4150
3 1700 4000 1200 10800 3600
4 2200 6200 600 13600 3400
5 2900 9100 500 16600
6 3800 12900 400 20500 3417
7 4800 17700 400 25300 3614
8 6000 23700 400 31300 3913
 Operations Research
The above table shows that the value of during fifth year is minimum. Hence the machine should be replaced after every fifth year.
 Operations Research
Example 3:
The cost of a machine is Rs. 6100 and its scrap value is only Rs.100. The maintenance costs are found to be
Year:
Maintenance Cost (in Rs.):
When should the Machine be replaced?

Solution :

The computations can be summarized in the following tabular form:

Table 13.3 Calculations for average cost of machine
Replace Cumulative Total cost
at the end maintenance
of year cost

1 100 100 6100 6100

2 250 350 6350 3175
3 400 750 6750 2250
4 600 1350 7350 1737.50
5 900 2250 8250 1650
6 1250 3500 9500
7 1600 5100 11100 1585.7
8 2000 7100 13100 1637.50

It is now observed that the machine should be replaced at the end of sixth year otherwise the average cost per year will start to
increase.

Replacement of items whose maintenance cost increases with time and the money value changes at a constant rate
To understand this let us define the following terms:
Money Value: Since money has a value over time, therefore the explanation of the statement: ‘Money is worth 10% per year’ can be
given in two ways:
(a) In one way, spending Rs.100 today would be equivalent to spend Rs.110 in year’s time. In other words if we plan to spend
Rs.110 after a year from now, we could spend Rs.100 today and an investment which would be worth Rs.110 next year.
(b) Alternatively if we borrow Rs.100 at the interest of 10% per year and spend Rs.100 today, we have to pay Rs.100 after one
year (next year).
hus, we conclude that Rs.100 is equal to Rs.110 a year from now. Consequently Rs. 1 from a year now is equal to (1+0.1)-1
Thus,
rupee today.
Present Worth Factor: As we have seen, a rupee a year from now will be equivalent to (1+0.1)-1 rupee today at the discount rate of
10% per year. So, one rupee in n years from now will be equal to (1+0.1)-n.Therefore, the quantity (1+0.1)-n is called the Present
Worth Factor (PWF) or Present Value (PV) of one rupee spent in n years from now. In general, if r is the rate of interest, then (1+r)-n
is called PWF or PV of one rupee spent in n years from now onwards. The expression (1+r)-n is known as compound amount factor
of one rupee spent in n years.

Discount Rate: Let r be the rate of interest. Therefore present worth factor of unit amount to be spent after one year is .
Then v is known as the discount rate. The optimum replacement policy for replacement of item where maintenance costs increase with
time and money value changes with constant rate can be determined by following method:
Suppose that the item (which may be machine or equipment) is available for use over a series of time periods of equal intervals (say
one year). Let
C = Purchase price of the item to be replaced
 Operations Research

Rt= Running (or maintenance) cost in the tth year

r=Rate of interest

is the present worth of a rupee to be spent in a year hence.

Let the item be replaced at the end of every nth year. The year wise present worth of expenditure on the item in the successive cycles
of n years can be calculated as follows:

Year 1 2----
---- n n+1 n+2 ---- 2n 2n+1

Present worth C+R1 R2v Rnvn--1 (C+R1)vn R2vn+1 Rnv2n-1 (C+R1)v2n

Assuming that machines has no resale value at the time of replacement, the present worth of the machine in n years will be given by

Summing up the right-hand

right hand side, column-wise
column wise

, using sum of an infinite G.P.

f(n) and f(n+1) given above at n = 0,1,2…, are called the weighted average cost of previous n years with weights 1,v, v2, ----vn-
1respectively. P(n) is the amount of money required now to pay all the future costs of acquiring and operating the equipment when it is
renewed every n years. However, if P (n) is less than P (n+1) then replacing the equipment each n year is preferable to replacing each
n years is preferable to replacing each (n+1) years. Further, if the best policy is replacing every n years, then the two inequalities P
(n+1) – P (n) > 0 and P (n-1) - P (n) < 0 must hold, without giving the proof we shall state the following two inequalities which holds
good at n, the optimal replacement interval.

As a result of these two inequalities, rules for minimizing costs may be stated as follows:

1.. Do not replace if the operating cost of next period is less than the weighted average of previous costs.

2.. Replace if the operating cost of the next period is greater than the weighted average of the previous costs.

Working Procedure
The step-by-step
step step procedure for solving the problem is stated as under:
1. Write in a column the running/maintenance costs of machine or equipment for different years, Rn.
 Operations Research

2. In the next column write the discount factor indicating the present value of a rupee received after (i-1) years,

3. The two column values are multiplied to get present value of the maintenance costs, i.e., .
4. These discounted maintenance costs are then cumulated to the ith year to .
5. The cost of machine or equipment is added to the values obtained in Step 4 above to

Obtain C+ .

6. The discount factors are then cumulated to get .

7. The total costs obtained in (Step 5) are divided by the corresponding value of the accumulated discount factor for each of the years.

8. Now compare the column of maintenance costs which is constantly increasing with the last column. Replace the machine in the latest
year that the last column exceeds the column of maintenance costs.

Example 4

A milk plant is offered an equipment A which is priced at Rs.60,000 and the costs of operation and maintenance are estimated to be
Rs.10,000 for each of the first 5 years, increasing every year by Rs. 3000 per year in the sixth and subsequent years. If money carries
the rate of interest 10% per annum what would the optimal replacement period?
Solution :
Table 13.4 Determination of optimal replacement period
At the Operating & Discounted Discounted Cumulative Discounted total Cumulative Weighted
end of maintenance factor operation & Discounted cost discounted average annual
year cost maintenance operation & factor cost
(n) Rn cost maintenance
cost

(1) (2) (3) (4)=(2)x(3) (5) (6)=(5)+60000 (7) (8)=(6)+(7)

1 10000 1.0000 10000.00 10000.00 70000.00 1.00 70000.00
2 10000 0.9091 9091.00 19091.00 79091.00 1.91 41428.42
3 10000 0.8264 8264.00 27355.00 87355.00 2.74 31933.83
4 10000 0.7513 7513.00 34868.00 94868.00 3.49 27207.75
5 10000 0.6830 6830.00 41698.00 101698.00 4.17 24389.18
6 13000 0.6209 8071.70 49769.70 109769.70 4.79 22913.08
7 16000 0.5645 9032.00 58801.70 118801.70 5.36 22184.36
8 19000 0.5132 9750.80 68552.50 128552.50 5.87 21905.89
9 22000 0.4665 10263.00 78815.50 138815.50 6.33 21912.82
10 25000 0.4241 10602.50 89418.00 149418.00 6.76 22106.52

From Table 13.4 we find the weighted cost is minimum at the end of 8th year, hence the equipment should be replaced at the end of
8th year.
Example 5 :
A Manufacturer is offered two machines A and B. Machine A is priced at Rs. 5000 and running cost is estimated at Rs. 800 for each
of the first five years, increasing by Rs. 200 per year in the sixth and subsequent years. Machine B, with the same capacity as A, costs
Rs. 2500, but has running cost of Rs. 1200 per year for six years, thereafter increasing by Rs. 200 per year. If money is worth 10%
per year, which machine should be purchased? (Assume that the machines will eventually be sold for scrap at a negligible price).

Solution

Since money is worth 10% per year, therefore discount rate is

 Operations Research
Table 13.5 Computation of weighted average cost for machine A
At the Operating & Discounted Discounted Cumulative Discounted total Cumulative Weighted
end of maintenance factor operation & Discounted cost discounted average annual
year cost maintenance operation & factor cost
(n) Rn cost maintenance
cost

(1) (2) (3) (4)=(2)x(3) (5) (6)=(5)+ (7) (8)=(6)+(7)

5000

1 800 1.0000 800 800 5800 1 5800

2 800 0.9091 727 1527 6527 1.9091 3419.035
3 800 0.8264 661 2188 7188 2.7355 2627.819
4 800 0.7513 601 2789 7789 3.4868 2233.98
5 800 0.6830 546 3336 8336 4.1698 1999.098
6 1000 0.6209 621 3957 8957 4.7907 1869.61
7 1200 0.5645 677 4634 9634 5.3552 1799.025
8 1400 0.5132 718 5353 10353 5.8684 1764.13
9 1600 0.4665 746 6099 11099 6.3349 1752.043
10 1800 0.4241 763 6862 11862 6.759 1755.053

From table 13.5 we conclude that for machine A 1600<1752.043<1800. Since the running cost of 9th year is 1600and that of 10th
year is 1800 and 1800>1752.043, it would be economical to replace machine A at the end of nine years.

Table 13.6 Computation of weighted average cost for machine B

At the Operating & Discounted Discounted Cumulative Discounted total Cumulative Weighted
end of maintenance factor operation & Discounted cost discounted average annual
year cost maintenance operation & factor cost
(n) Rn cost maintenance
cost

(1) (2) (3) (4)=(2)x(3) (5) (6)=(5)+ (7) (8)=(6)+(7)

2500
1 1200 1.0000 1200.00 1200.00 3700.00 1.00 3700.00
2 1200 0.9091 1090.92 2290.92 4790.92 1.91 2509.52
3 1200 0.8264 991.68 3282.60 5782.60 2.74 2113.91
4 1200 0.7513 901.56 4184.16 6684.16 3.49 1916.99
5 1200 0.6830 819.60 5003.76 7503.76 4.17 1799.55
6 1200 0.6209 745.08 5748.84 8248.84 4.79 1721.84
7 1400 0.5645 790.30 6539.14 9039.14 5.36 1687.92
8 1600 0.5132 821.12 7360.26 9860.26 5.87 1680.23
9 1800 0.4665 839.70 8199.96 10699.96 6.33 1689.05
10 2000 0.4241 848.20 9048.16 11548.16 6.76 1708.56

In table13.6 we find that 1800<1689<2300 so it is better to replace the machine B after 8th year. The equivalent yearly average
discounted value of future costs is Rs. 1748.60 for machine A and it is 1680.23for machine B. Hence, it is more economical to buy
machine B rather than machine A.
 Operations Research

Module 7. Sequencing problem

Lesson 14
INTRODUCTION AND GENERAL NOTATIONS

Introduction

Sequencing problems are concerned with an appropriate order (sequence) for a series of jobs to be done on
a finite number of service facilities (like machines) in some wel -defined
defined technological order so as to optimize
some ef iciency measure such as total elapsed time or overal cost etc. In such cases, the ef ectiveness is a
function of the order or sequence in which the tasks are performed. The eef ectiveness may be measured in
terms of cost, time or mileage etc. A sequencing problem could involve jobs in a manufacturing plant, aircraft
waiting for landing and clearance, maintenance scheduling in a factory, programmers to be run on a computer
centre, customers in a bank, and so forth.

Sequencing

In Sequencing we are concerned with a situation where the ef ectiveness measure is a function of the order or
sequence in which a series of tasks or jobs are performed. Suppose we have n jobs (1,2,3,---,n),
(1,2,3, ,n), each of
which has to be processed or performed one at the time on each of m machines A,B,C,---.
A,B,C, The order
(sequence) of processing each job through the machines as wel
we as the actual or expected time required by
the jobs on each of the machine is also given. The eef ectiveness in terms of cost, times or mileage etc. can be
measured for any given sequence of jobs at each machine and our aim is to select the most suitable sequence
(which optimizes the ef ectiveness measure) among al theoretical possible sequences whose number wil be

(n!)m. Although theoreticall y it is always possible to select the best sequence by testing each one but it is
practical
practica y impossible due to large number of computations. Hence we have to compute ef ectiveness for

each of (n!)m sequences before selecting the most suitable one. But this is practical y impossible to do. Let
us explain as to what the sequencing problem is.

14.2.1 Problem of sequencing

Definition: Suppose there are n jobs (1, 2, ---, n) each of which has to be processed one at a time at each
of m machines A, B, C, …………. The order of processing each job through the machines is given (for
example, Job 1 is processed through machines A, C, B, - in that order). The time required for each job at
each machine is also given. The problem is to find among (n!)m number of al possible sequences (or
combinations) that sequence (or order) for processing the jobs so that the total elapsed time for al
a the jobs
wil be a minimum. Mathematical
Mathematica y, let us define:

Time for ith job on machine A

Time for ith job on machine B
Time from start of the first job to completion of the last job
We wish to determine for each machine a sequence (i1,i2, … … … … in) where (i1,i2, … … …
… in) is a permutation of the integers (1,2, .. … … n) which wi
wil minimize total elapsed time T.
 Operations Research

Basic Terminology and Notations

The fol owing terminology and notations wil be used in this unit:

Number of machines

It means service facilities through which a job must pass before it is completed. In a milk plant, milk packed in
pouches has to be processed through boiling, cooling, packaging etc. In this milk packing in pouches
constitutes the job and various processes constitute the number of machines.

Processing time

It means the time required by each job on each machine. It is denoted by Tij which means processing time

required by the ith machine (i=1, 2, 3---, n; j=1, 2, 3---, m).

Processing order

It refers to the order in which various machines are required for completing the job.

Idle time on a machine

This is the time a machine remains idle during the total elapsed time. Let Xij denote the idle time of machine j

between the end of (i-1) th job and the start of ith job.

Total elapsed time

This is the time between starting the first job and completing the last job including the idle time (if any) in a
particular order by the given set of machines. This wil be denoted by T.

No passing rule

It means that passing is not al owed i.e. same order of jobs is maintained over each machine. If each of the n
jobs is to be processed through two machines A and B in the order AB then this means that each job wil go
to machine A and then to B.

Basic Assumptions of Sequencing Problem

• No machine can process more than one operation at a time.

• Each job once started on a machine is to be performed up to its completion on that machine.
• A job is an entity i.e. even though the job represents a lot of individual parts, no lot may be processed
by more than one machine at a time.
• The time intervals for processing are independent of the order in which the operations are performed.
• There is only one of each type of machine.
• A job is processed as soon as possible subject only to ordering requirements.
• Al jobs are known and are ready to start processing before the period under consideration begins.
• The processing times are on dif erent machines are independent of the order of the job in which they
are to be processed.
• The time taken by the jobs in moving from one machine to another is very negligible and is taken as
equal to zero.

Visit Our Website - www.InstantAgriculture.com 86 Download Our App - Instant Agriculture

 Operations Research

Module 7. Sequencing problem

Lesson 15
SOLUTION OF A SEQUENCING PROBLEM

Introduction
When a number of jobs are given to be done and they require processing on two or more machines, the main
concern of a manager is to find the order or sequence to perform these jobs. We shal consider the sequencing
problems in respect of the jobs to be performed in a factory and study the method of their solution. Such
sequencing problems can be broadly divided in two groups. In the first one, there are n jobs to be done, each of
which requires processing on some or al of the k dif erent machines. We can determine the ef ectiveness of each
of the sequences that the technological
technologica y feasible (that is to say, those satisfying the restrictions on the order in
which each job must be processed through the machines) and choose a sequence which optimizes the
ef ectiveness. To il ustrate, the timings of processing of each of the n jobs on each of the k machines, in a certain
given order, may be given and the time for performing the jobs may be the measure of ef
e ectiveness. We shal
select the sequences for which the total time taken in processing al the jobs on the machines would be the
minimum.
In this unit we will look into solution of a sequencing problem. In this lesson the solutions of fol owing cases wil
be discussed:
a) n jobs and two machines A and B, al jobs processed in the order AB.
b) n jobs and three machines A, B and C al jobs processed in the order ABC
c) Problems with n jobs and m machines.
15.1.1 Processing of n jobs through two machines
The simplest possible sequencing problem is that of n job two machine sequencing problem in which we want to
determine the sequence in which n-job should be processed through two machines so as to minimize the total
elapsed time T. The problem can be described as:
a) Only two machines A and B are involved;
b) Each job is processed in the order AB.
c) The exact or expected processing times are known and are
provided in the fol owing table

Machine Job(s)
1 2 3 -- - i -- - n
A A1 A2 A3 -- - Ai -- - An
B B1 B2 B3 -- - Bi -- - Bn

The problem is to find the sequence (or order) of jobs so as to minimize the total elapsed time T. The solution of
the above problem is also known as Johnson’s procedure which involves the fol
fo owing steps:
 Operations Research

Step 1. Select the smal est processing time occurring in the list

if there is a tie, either of the sma

smal est processing
times can be selected.
Step 2. If the least processing time is Ar , select the r th job first. If it is Bs, do the sth job last as the
given order is AB
Step 3. There are now (n-1) jobs left to be ordered. Repeat steps I and II for the remaining set of
processing times obtained by deleting the processing time for both the machines corresponding
to the job already assigned.
Step 4. Continue in the same manner til
ti the entire jobs have been ordered. The resulting ordering wil
minimize the total elapsed time T and is cal ed the optimal sequence.
Step 5. After finding the optimal sequence as stated above find the total elapsed time and idle times on
machines A and B as under:
Total elapsed The time between starting the first job in the optimal sequence on machine A and completing the
ime = last job in the optimal machine B.
dle time on (Time when the last job in the optimal sequence on sequences is completed on machine B)-
machine A = (Time when the last job in the optimal sequences is completed on machine A)
dle time on (Time when the first job in the optimal sequences is completed on machine A)+
machine B =

The Johnson’s procedure can be il ustrated by fol owing examples:

Example 1;
There are nine jobs, each of which must go through two machines P and Q in the order PQ, the processing times
(in hours) are given below:

Job(s)
Machine
A B C D E F G H I

P 2 5 4 9 6 8 7 5 4
Q 6 8 7 4 3 9 3 8 11

Find the sequence that minimizes the total elapsed time T. Also calculate the total idle time for the machines in
this period.
Solution :
The minimum processing time on two machines is 2 which correspond to task A on machine P. This shows that
task A wil be preceding first. After assigning task A, we are left with 8 tasks on two machines
Machine
 Operations Research

P 5 4 9 6 8 7 5 4
Q 8 7 4 3 9 3 8 11

Minimum processing time in this reduced problem is 3 which correspond to jobs E and G (both on machine Q).
Now since the corresponding processing time of task E on machine P is less than the corresponding processing
time of task G on machine Q therefore task E wil be processed in the last and task G next to last. The situation
wil be dealt as

A G E

The problem now reduces to fol

fo owing 6 tasks on two machines with processing time as fo
fol ows:
Machine B C D F H I
P 5 4 9 8 5 4
Q 8 7 4 9 8 11

Here since the minimum processing time is 4 which occurs for tasks C and I on machine P and task D on
machine Q. Therefore, the task C which has less processing time on P wil be processed first and then task I and
task D wil be placed at the last i.e., sequence cel
cel .
The sequence will appear as fol
fo ows:
A C I D E G

The problem now reduces to the fol

fo owing 3 tasks on two machines

Machine B F H
P 5 8 5
Q 8 9 8

In this reduced table the minimum processing time is 5 which occurs for tasks B and H both on machine P. Now
since the corresponding time of tasks B and H on machine Q are same i.e. 8. Tasks B or H may be placed
arbitrarily in the sequence cel
ce s. The remaining task F can then be placed in the sequence cel .
Thus the optimal sequences are represented as

A I C B H F D E G
or
A 1 C H B F D E G

Further, it is also possible to calculate the minimum elapsed time corresponding to the optimal sequencing
 Operations Research

Job Machine A Machine B

Sequence Time In Time Out Time In Time Out
A 0 2 2 8
I 2 6 8 19
C 6 10 19 26
B 10 15 26 34
H 15 20 34 42
F 20 28 42 51
D 28 37 51 55
E 37 43 55 58
G 43 50 58 61

Hence the total elapsed time for this proposed sequence staring from job A to completion of job G is 61 hours
.During this time machine P remains idle for 11 hours (from 50 hours to 61 hours)and the machine Q remains idle
for 2 hours only (from 0 hour to 2 hour ) .

Processing of n Jobs through Three Machines

The type of sequencing problem can be described as fo fol ows:
a) Only three machines A, B and C are involved;
b) Each job is processed in the prescribed order ABC
c) No passing of jobs is permitted i.e. the same order over each machine is maintained.
d) The exact or expected processing times

are known and are denoted by the fo

fol owing table

Job(s)
Machine 1 2 3 -- - i -- - n

A A1 A2 A3 -- - Ai -- - An
B B1 B2 B3 -- - Bi -- - Bn
C C1 C2 C3 Ci Cn

Our objective wil

wi be to find the optimal sequence of jobs which minimizes the total elapsed time. No general
procedure is available so far for obtaining an optimal sequence in such case. However, the Johnson’s procedure
can be extended to cover the special cases where either one or both of the foll owing conditions hold:

a)
a The minimum processing time on machine A ≥ the maximum processing time on machine B.
b)
b The minimum processing time on machine C ≥ the maximum processing time on machine B.

The method is to replace the problem by an equivalent problem involving n jobs and two machines. These two
 Operations Research

fictitious machines are denoted by G and H and the corresponding time Gi and Hi are defined by

Now this problem with prescribed ordering GH is solved by the method with n jobs through two machines, the
resulting sequence wil also be optimal for the original problem. The above methodology is il ustrated by fol owing
example:

Example 2:
There are five jobs (namely 1,2,3,4 and 5), each of which must go through machines A, B and C in the order
ABC. Processing Time (in hours) are given below:

Jobs 1 2 3 4 5
Machine A 5 7 6 9 5
Machine B 2 1 4 5 3
Machine C 3 7 5 6 7

Find the sequence that minimum the total elapsed time required to complete the jobs.

Solution

Here since the condition of is satisfied the given problem

can be converted into five jobs and two machines problem.

Jobs
1 7 5
2 8 8
3 10 9
4 14 11
5 8 10

The Optimal Sequence wil

wil be
2 5 4 3 1

Total elapsed Time wil

wi be
Machine A Machine B Machine C
Jobs
In Out In Out In Out

2 0 7 7 8 8 15
5 7 12 12 15 15 22
4 12 21 21 26 26 32
3 21 27 27 31 32 37
 Operations Research

Min. total elapsed time is 40 hours.

Idle time for Machine A is 8 hrs. (32
(32-40)
40)
Idle time for Machine B is 25 hours (0-7,
7, 8-12,
8 15-21,
21, 26
26-27, 31--32 and 34-40)
40)
Idle time for Machine C is 12 hours (0-8,
8, 22-26.)

Problems with n Jobs and m Machines

Let there be n jobs, each of which is to be processed through m machines, say in the order
Let T ij be the time taken by the ith machine to complete the jth job.
The iterative procedure of obtaining an optimal sequence is as fol ows:

Step I: Find i)
i)) )
and
i i)
for j=1,2,---,
j=1,2, , n
Step II: Check whether
a.
for i=2,3,----
----,m-1
Or
b.
for i=2,3,---
---,m-1
Step III: If the inequalities in Step II are not satisfied, method fails, otherwise, go to next step.
Step IV: Convert the m machine problem into two machine problem by introducing two fictitious machines G
and H, such that

Determine the optimal sequence of n jobs through 2 machines by using optimal sequence algorithm.
Step V: In addition to condition given in Step IV, if
is a fixed positive
constant for al , then determine the optimal sequence of n jobs and two
machines
hines and in the order by using the optimal sequence algorithm.
Example 3:
Find an optimal sequence for the fol owing sequencing problem of four jobs and five machines when passing is
not al
a owed, of which processing time (in hours) is given below:

Job Machine
A B C D E
1 7 5 2 3 9

2 6 6 4 5 10

3 5 4 5 6 8
4 8 3 3 2 6

Also find the total elapsed time.

 Operations Research

Here

respectively
Since satisfied therefore the problem can be converted into 4
jobs and 2 fictitious machines G and H as fol
fo ows:

Fictitious Machine
Job

1 17 19
2 21 25
3 20 23
4 16 14

The above sequence wil

wi be:
1 3 2 4

Total Elapsed Time Corresponding to Optimal Sequence can be obtained as fo

fol ows:

Machine A Machine B Machine C Machine D Machine E

Job In Out In Out In Out In Out In Out
1 0 7 7 12 12 14 14 17 17 26
3 7 12 12 16 16 21 21 27 27 35
2 12 18 18 24 24 28 28 33 35 45
4 18 26 26 29 29 32 33 35 45 51

Thus the minimum elapsed time is 51 hours.

Idle time for machine A= 25 hours(26
hours(26-51)
Idle time for machine B= 33 hours(0
hours(0-7,16--18,24-26,29
26,29-51)
Idle time for machine C= 37 hours(0
hours(0-12,14
12,14-16,21-24,28
24,28-29,32-51)
51)

Idle time for machine D= 35 hours (0-14,17

14,17-21,27-28,35
28,35-51)
Idle time for machine E= 18 hours (0-17,26
17,26-27)
 Operations Research

Module 8. Queuing theory

Lesson 16
INTRODUCTION AND CLASSIFICATION OF QUEUES

Introduction

The study of waiting lines, called queuing theory is one of the oldest and most widely used Operations Research techniques.
Waiting lines are the most frequently encountered problem in our daily life. The queuing theory, also called the waiting line
theory, owes its development to A. K. Erlang’s efforts to analyze telephone traffic congestion with a view to satisfying the
randomly arising demand for the services of the Corpenhagen automatic telephone system, in the year 1909. The theory is
applicable to situation where the ‘customers’ arrive at some ‘service stations’ for some service; wait (occasionally not); and
then leave the system after getting the service.

A flow of customers from finite/ infinite population towards the services facility form a queue (waiting line) on account of lack of
capability to serve them all at a time. The queues may consist of customers for buying milk and other milk products at a milk
parlour, machines waiting to be repaired, trucks or vehicles waiting at the milk plant, patients in a hospital who need treatment
and so on. In the absence of a perfect balance between the service facilities and the customers, waiting is required either of the
services facilities or for the customer’s arrival. In general a queue is formed when either units requiring services-commonly
referred as customers, wait for service or the service facilities, stand idle and wait for customers. The queuing models are
basically relevant to service oriented organizations and suggest ways and means to improve the efficiency of the service.
Queuing methodology indicates the optimal usage of existing manpower and other resources to improve the service. Waiting
lines can’t be eliminated completely but suitable techniques can be used to reduce the waiting line of an object in the system.

Queuing System

The mechanism of a queuing process is very simple. Customers arrive at services counter are attended by one or more of the
servers. As soon as a customer is served, he departs from the system. Thus a queuing system can be described as composed of
customers arriving for service, waiting for service if it is not immediate, and if having wanted for service, leaving the system after
being served.

Component of a Queuing System

A queuing system can be described by the following components:

16.3.1 Input process (or Arrival pattern)

This is considered with the pattern in which the customers arrive and join the system. An input source is characterized by

Visit Our Website - www.InstantAgriculture.com 94 Download Our App - Instant Agriculture

 Operations Research

a) Size of the calling population.

b) Pattern of arrivals at the system.
c) Behaviour of the arrivals.
Customers requiring service are generated at different times by an input source, commonly known as population. The rate at
which customers arrive at the service facility is determined by the arrival process.

Size of the calling population

The size represents the total number of potential customers who will require service. The source of customers can be either
finite or infinite. It is considered infinite if the number of people being very large e.g. all people of a city or state (and others)
could be the potential customers at a milk parlour. Whereas there are many situations in industrial conditions where we cannot
consider the population to be infinite—it is finite. The customers may arrive for service individually or in groups. Single arrivals
are illustrated by a customer visiting a milk parlour, students arriving at a library counter etc. On the other hand, families visiting
restaurants, ships discharging cargo at a dock are examples of bulk or batch arrivals.

Pattern of arrivals at the system

Customers arrive in the system at a service facility according to some known schedule (for example one patient every 15
minutes or a candidate for interview every half hour) or else they arrive randomly. Arrivals are considered at random when they
are independent of one another and their occurrence cannot be predicted exactly. The queuing models wherein customers’
arrival times are known with certainty are categorized as deterministic models and are easier to handle. On the other hand, a
substantial majority of the queuing models are based on the premise that the customers enter the system stochastically, at
random points in time. The arrival process (or pattern) of customers to the service system is classified into two categories:
static and dynamic. These two are further classified based on the nature of arrival rate and the control that can be exercised
on the arrival process.

In static arrival process, the control depends on the nature of arrival rate (random or constant). Random arrivals are either at a
constant rate or varying with time. Thus to analyze the queuing system, it is necessary to describe the probability distribution of
arrivals. From such distributions average time between successive arrivals, is obtained also called inter-arrival time (time
between two consecutive arrivals), and the average arrival rate (i.e. number of customers arriving per unit of time at the service
system).

The dynamic arrival process is controlled by both service facility and customers. The service facility adjusts its capacity to
match changes in the demand intensity, by either varying the staffing levels at different timings of service, varying service charges
(such as telephone call charges at different hours of the day or week) at different timings, or allowing entry with appointments.
Frequently in queuing problems, the number of arrivals per unit of time can be estimated by a probability distribution known as
the Poisson distribution, as it adequately supports many real world situations
Se rvice Me chanism (or Se rvice Patte rn)

The service is provided by a service facility (or facilities). This may be a person (a bank teller, a barber, a machine (elevator,
gasoline pump), or a space (airport runway, parking lot, hospital bed), to mention just a few. A service facility may include one
person or several people operating as a team. There are two aspects of a service system
a) the configuration of the service system
b) the speed of the service.

Configuration of the service system

The customers’ entry into the service system depends upon the queue conditions. If at the time of customers’ arrival, the server
is idle, then the customer is served immediately. Otherwise the customer is asked to join the queue, which can have several
configurations. By configuration of the service system we mean how the service facilities exist. Service systems are usually
classified in terms of their number of channels, or numbers of servers.
i) Single Server – Single Queue -- The models that involve one queue – one service station facility are called single

Visit Our Website - www.InstantAgriculture.com 95 Download Our App - Instant Agriculture

 Operations Research

server models where customer waits till the service point is ready to take him for servicing. Students arriving at a
library counter are an example of a single server facility.
ii) Single Server – Several Queues – In this type of facility there are several queues and the customer may join any one
of these but there is only one service channel.
iii) Several (Parallel) Servers – Single Queue – In this type of model there is more than one server and each server
provides the same type of facility. The customers wait in a single queue until one of the service channels is ready to
take them in for servicing.
iv) Several Servers – Several Queues – This type of model consists of several servers where each of the servers has a
different queue. Different cash counters in an electricity office where the customers can make payment in respect of
their electricity bills provide an example of this type of model. Different ticket issue encounters in a trade fair and
different boarding pass encounters at an airport are also other possible examples of this type of model.
v) Service facilities in a series – In this, a customer enters the first station and gets a portion of service and then moves
on to the next station, gets some service and then again moves on to the next station. …. and so on, and finally leaves
the system, having received the complete service. For example in a milk plant packaging of milk pouches consist of
boiling, pasteurization, cooling and packaging operations, each of which is performed by a single server in a series.
Speed of service

In a queuing system, the speed with which service is provided can be expressed in either of two ways—as service rate and as
service time. The service rate describes the number of customers serviced during a particular time period and the service time
indicates the amount of time needed to service a customer. Service rates and times are reciprocal of each other and either of
them is sufficient to indicate the capacity of the facility. Thus if a cashier can attend, on an average 5 customers in an hour, the
service rate would be expressed as 5 customers/hour and service time would be equal to 12 minutes/customer. Generally, we
consider the service time only. If these service times are known exactly, the problem can be handled easily. But, as generally
happens, if these are different and not known with certainty, we have to consider the distribution of the service times in order
to analyze the queuing system. Generally, the queuing models are based on the assumption that service times are exponentially
distributed about some average service time.

Queue discipline

In the queue structure, the important thing to know is the queue discipline. The queue discipline is the rule determining the
formation of queue, manner in which customers form the queue are selected for service. There are a number of ways in which
customers in the queue are served. Some of these are:
Static queue disciplines
These are based on the individual customer's status in the queue. The most common queue disciplines are:
i) First-Come-First-Served (FCFS): If the customers are served in the order of their arrival, then this is known as the
FCFS service discipline. For example, this type of queue discipline is observed at a milk parlour, railway station etc.
FCFS is also known as First In First Out (FIFO).
ii) Last-Come-First-Served (LCFS): Sometimes, the customers are serviced in the reverse order of their entry so that
the ones who join the last are served first and the system is referred to as LCFS. For example, in a big godown the
items which come last are taken out first. Similarly, the people who join an elevator last are the first ones to leave it.
Dynamic queue disciplines
These are based on the individual customer attributes in the queue. Few of such disciplines are:
i) Service in Random Order (SIRO): Under this rule customers are selected for service at random, irrespective of their
arrivals in the service system. In this, every customer in the queue is equally likely to be selected. The time of arrival of
the customers is, therefore, of no relevance in such a case.
ii) Priority Service: Under this rule customers are grouped in priority classes on the basis of some attributes such as
service time or urgency or according to some identifiable characteristic, and FCFS rule is used within each class to
provide service. Treatment of VIPs in preference to other patients in a hospital is an example of priority service.

Visit Our Website - www.InstantAgriculture.com 96 Download Our App - Instant Agriculture

 Operations Research

Custome r’s behaviour

Another thing to consider in the queuing structure is the behaviour or attitude of the customers entering the queuing system. On
this basis, the customers may be classified as being patient, or impatient. If a customer, on arriving at the service system stays in
the system until served, no matter how much he has to wait for service is called a patient customer whereas the customer, who
waits for a certain time in the queue and leaves the service system without getting service due to certain reasons such as a long
queue in front of him is called an impatient customer. The customers generally behave in four ways
i) Balking: A customer may leave the queue because the queue is too long or the estimated waiting time is too long or
waiting space is inadequate, for desired service and may decide to return for service at a later time. In queuing theory
this is known as balking.
ii) Reneging: A customer, after joining the queue, waits for some time and leaves the service system due to intolerable
delay or due to impatience.
iii) Jockeying: A customer who switches from one queue to another, hoping to receive service more quickly, is said to be
jockeying.
iv) Priorities: In certain applications some customers are served before others regardless of their order of arrival. These
customers have priority over others.

Visit Our Website - www.InstantAgriculture.com 97 Download Our App - Instant Agriculture

 Operations Research

Module 8. Queuing theory

Lesson 17
SOLUTION OF QUEUING MODELS

Introduction

The ultimate objective of the analysis of queuing systems is to understand the behavior of their underlying
processes so that informed and intel igent decisions can be made in their management. In a specified queuing
system the problem is to determine the probability distribution of queue length, waiting time of customers and
the busy period. Queuing theory uses queuing models to represent the various types of queuing systems that
arise in practice. Formulae for each model indicate how the corresponding queuing system should perform,
including the average amount of waiting time under a variety of circumstances. Therefore, these queuing models
are helpful in determining how to operate a queuing system in the most ef ective way. Providing too much
service capacity to operate the system involves excessive costs. But not providing enough service capacity
results in excessive waiting and al its unfortunate consequences. The models enable finding an appropriate
balance between the cost of service and the amount of waiting.

Characteristics of A Queuing System

Queuing models enable the analyst to study the ef ect of manipulating decision variables on the operating
characteristics of a service system. The most commonly used characteristics are stated as under:

Queue length

The average number of customers in the queue waiting to get service is known as ‘queue length’. ‘Short
queues’ could mean either good customer service or large waiting space while ‘long queues’ could indicate
low service ef iciency or a little waiting space.

System length

It is the average number of customers in the system waiting to be served and those being served. Long queues
imply congestion, potential customer dissatisfaction and need for more capacity.

Waiting time in queue

Waiting time is the average time that a customer has to wait in the queue to get service. Long waiting times may
indicate a need to adjust the service rate of the system or change the arrival rate of customers.

Total time in system

The average time that customer spends in the system from entry in the queue to completion of service. If this
time is more then there may be a need to change the priority discipline, increase productivity or adjust the
capacity.

Server idle time

The relative frequency with which the service system is idle which is directly related to cost. Queuing theory
analysis involves the study of systems’ behaviour over time.

Transient and Steady States

Visit Our Website - www.InstantAgriculture.com 98 Download Our App - Instant Agriculture
 Operations Research

When a service system is started, it progresses through a number of changes. However, it attains stability after
some time. Before the service operations start, it is very much influenced by the initial conditions (number of
customers in the system) and the elapsed time. This period of transition is termed as transient state.. A system
is said to be in transient-state
transient state when its operating characteristics are dependent on time.

However, after su
suf icient time has passed, the system becomes independent of the initial conditions and of the
elapsed time (except under very special conditions) and enters a steady state condition. A steady state
condition is said to prevail when the behaviour of the system becomes independent of time. Let denote
the probability that are n units in the system at time t. We know that the change of with respect to t is

described by the derivative . Then the queuing system is said to be stable eventuall y, in the sense that the
probability is independent of time, i.e. remains the same as time . Mathematical
Mathematica y, in a steady
state,

This implies that

From the practical point of view period of the steady state behaviour of the system, queuing system under the
existence of steady state condition are being considered.

17.3 Notations and Symbols

The notations used in the analysis of a queuing system are as fol
fo ows:

n= Number of customers in the system (waiting and in service)

Pn (t) = Transient state probability that n ca
call ing units are in the queuing system at time t
En= The state in which there are n cal ing units in the system
Pn = Steady state probability of having n units in the system
λ= Average (expected) customer arrival rate or average number of arrivals per unit of time
in the queuing system
μ= Average (expected) service rate or average number of customers served per unit time at
the place of service
ρ=
 Operations Research

Traf
Tra ic intensity or server utilization factor
s= Number of service channels (service facilities or servers)
N= Maximum number of customers al owed in the system.
syste
L = Average (expected) number of customers in the system (waiting and in service)
s
L = Average (expected) number of customers in the queue (queue length)
q
L = Average (expected) length of non-empty
empty queue
b
W = Average (expected) waiting time in the system (waiting and in service)
s
W = Average (expected) waiting time in the queue
q
Pw = Probability that an arriving customer has to wait
17.3.1 Kendall’s notation for representing queuing models

General
Genera y queuing model can be specified by the symbolic representation
where,
a: Probability distribution of the arrival (or inter-arrival)
arrival) time
b: Probability distribution of the service time.
c: Number of channels (or service stations)
d: Capacity of the system
e: Queue discipline

The first three characteristics (a|b|c) in the above notation were introduced by D. Kendal
Kenda in 1953. Later in
1966, A. Lee added the fourth (d) and fifth (e) characteristics to the notation. Traditiona
Traditional y, the exponential
distribution in queuing
queu problems is denoted by M. Thus, (M|M|1): (∞|FIFO)
∞|FIFO) indicates a queuing system when
the inter-arrival
inter times and service times are exponentia
exponential y distributed having one server in the system with first in
first out discipline and the number of customers al owed in the system can be infinite.

Traffic Intensity (or Utilization factor)

An important measure of simple queue is its traffic intensity, where

The unit of traf

traf ic intensity is Erlang.

A necessary condition for a system to have settled down to steady state is that

i.e. arrival rate < service rate. If

, the arrival rate is greater than the service rate and consequently , the
number of units in the queue tends to increase indefinitely as the time passes on, provided the rate of service is
not af
a ected by the length of queue.
Queuing Models

The queuing models are categorized as ‘deterministic’ or ‘probabilistic’. If each customer arrives at known
 Operations Research

intervals and the service time is known with certainty, the queuing model wi
wil be deterministic in nature. When
both arrival and service rate are unknown and assumed to be random variable then this type of queuing model
is known as probabilistic.

Probability Distributions in Queuing systems

syste

The arrival of customers at a queuing system varies between one system and another, but in practice one
pattern of completely random arrivals is observed.

Distribution of arrivals ‘the Poisson process’ (pure birth process)

The models in which only arrivals are counted and no departure take place are cal ed pure birth models. In
terms of queuing, birth-death
death process that is increased by birth or arrival in the system and decreased by death
or departure of serviced customer from the system. If the arrivals are completely random, then the probability
distribution of number of arrivals in a fixed time–interval,
interval, fol ows Poisson probability distribution with
parameter (mean) . Thus,

Distribution of Inter–Arrival
Inter times (Exponential Process)
Proce

Inter–arrival
Inter– times are defined as the time intervals between two successive arrivals. It wil be proved that if the
arrival process fol
fo ows Poisson distribution, an associated random variable defined as the time between
successive arrivals (inter-arrival
arrival time) fol ows an exponential distribution and vice-versa.
vice versa.

The expected (or mean) time of inter arrival is given by

where is
the mean arrival rate.
Thus, t has the exponential distribution with mean , we would intuitively expect that if the mean arrival rate

is then the mean time between arrival is . Conversely, it can also be independent and have the
exponential distribution then the arrival rate fol ows the Poisson distribution.

Classification of Queuing Models (LISTING OF FOUR MODELS)

The queuing models are classified as fo

fol ows:
Model I : (M|M|1): ∞|FCFS)
(∞|FCFS)
Model II : (M|M|s): (∞|FCFS)
∞|FCFS)
Model III : (M|M|1): (N|FCFS)
Model IV : ∞|FCFS)
(M|M|s): (∞|FCFS)
In this lesson Model I and II have been discussed as Model III and IV are beyond the scope of this course.

Model I (single channel queuing model with Poisson arrivals and exponential service times)

This model is symbolical

symbolica y represented as by (M|M|1): (∞|FCFS).
∞|FCFS). This denotes Poisson arrival (exponential
inter--arrival), Poisson departure (exponential service time), single server, Infinite capacity and ‘First come,
First served’ service discipline. This model is also cal ed ‘birth and death model’. This model is one of the most
 Operations Research

widely used and simplest models. It assumes the fo

fol owing conditions:
(i)
( Arrivals are served on a ‘first come-first
come served’ (FCFS) basis. (i
) Every arrival waits to be served regardless of the length of the line.
(ii)
( ) Arrivals are independent of preceding arrivals, but the average number of arrivals does
not change over time.
(iv)
( Arrivals are described by a Poisson probability distribution and come from an infinite population.
(v)
( Service times also vary from one customer to the next and are independent of one another, but their
average rate is known.
(vi)
( Service times occur according to the negative exponential probability distribution.
(vi
(v ) The average service rate is greater than the average arrival rate.

To obtain the system of steady

steady- state equations

The probability that there will be n units (n>0) in the system at time (t+∆t) may be expressed as the sum of
three independent compound properties by using the fundamental properties of probability, Poisson arrivals,
and of exponential service times.
(i) The product of three probabilities (Fig. 17.1)
a)
a that there are n units in the system at time
b)
b that there is no arrival in time
c)
c that there is no service in time is given by

(i) ) The product of three probabilities (Fig. 17.2)

a)
a that there are (n-1) units in the system at time t = Pn-1(t)

b)
b that there is one arrival in time
 Operations Research

c)
c that there is no service in time is given by

(ii) ) The product of probabilities (see Fig. 17.3)

a)
a that there are (n+1) units in the system at time t = Pn+1(t)

b)
b that there is no arrival in time

c)
c that there is one service in time
is given by

Now, by adding above three independent compound probabilities, we obtain the probability of n units in the
system at time (t + ∆t), i.e.,

(17.1)

where

The equation may be written as

Now, taking limit as on both sides,

or
 Operations Research

(17.2)

In a similar fashion, the probability that there wil be no unit (i.e. n = 0) in the system at time (t + ∆t) wil
wi be
the sum of the fol
fo owing two independent probabilities:

(i) P[that there is no unit in the system at time t and no arrival in time

(i
( ) P[that there is one unit in the system at time t, one unit serviced in ∆t and no arrival in time ∆t]=

Now, adding these two probabilities we get

(17.3)

Now, taking limit as ∆t→0 on both sides,

(17.4)

Consequently the equations (17.2) and (17.4) can be written as

(17.5)

(17.6)
Equations (17.5) and (17.6) constitute the system of steady state dif erence equations for the model.

To solve the system of difference equations

We solve the diff erence equations given in (17.5) and (17.6) by the method of successive substitution

Since P0 = P0 and putting n=0 in equation (17.5) we get

---, (17.7)
Now using the fact that
 Operations Research

or

or
or (17.8)

Now, substituting the value of P0 from (17.8) in (17.7) , we get

(17.9)

The equations (17.8) and (17.9) give the probability distribution of queue length

Measure of model 1

Expected (average) number of units in system (Ls) is given by

(17.10)

Expected (average) queue length (Lq) is given by:

Since there are (n-1)

(n 1) units in the queue excluding one being serviced

Substituting the value of P0 from equation 17.8

(17.11)

Expected (average) waiting time in the queue (excluding service time) (W q) is given by:

(17.12)
 Operations Research

Expected (average) waiting time in the system (including service time) (W s) is given by:

(17.13)

Expected (average) length of non-empty

non empty queue, (L/L > 0) is given by:

Expected variance of queue length is given by:

Example 1:

Customers arrive at a milk parlour being manned by a single Individual at rate of 25 per hour. The time
required to serve a customer
omer has exponential distribution with a mean of 30 per hour. Discuss the various
characteristics of the queuing system, assuming that there is only one server.

Solution

Arrival rate (λ) = 25 per hour, Service rate (μ) = 30 per hour

Expected number of units in system (Ls) =

Expected queue length ( Lq) =

Expected waiting time in the queue

Example 2 :

In a service department manned by one server, on an average 8 customers arrive every 5 minutes while the
server can serve 10 customers in the same time assuming Poisson distribution for arrival and exponential
 Operations Research

distribution for service rate. Determine:

a)
a Average number of customers in the system.

b)
b Average number of customers in the queue.

c)
c Average time a customer spends in the system.

d)
d Average time a customer waits before being served.

Solution

Arrival rate (λ) =

=1.6 customers per minute.

Service rate (μ)=

2 customers per minute.

a)
a Average number of customer in the system.

b)
b Average number of customer in the queue.

c)
c Average time a customer spends in the system.

d)
e) Average time a customer waits before being served

17.7.2 Model II (A) general erlang queuing model (Birth-Death

Death Process)

∞|FCFS), but this is a general model in which the rate of arrival

This model is also represented by (M|M|1): (∞|FCFS),
and the service depend on the length n of the line.

17.7.2.1 To obtain the system of steady state equation

Let arrival rate service rate ; [depending upon n]

 Operations Research

(17.14)

(17.15)
Now dividing equations (17.14) and (17.15) by ∆t, taking limits as and fol owing the same procedure
as in Model I, obtain

(17.16)

(17.17)

The above written two equations are dif erential equations which could be solved if a set of initial values P0 (0),
P1 (0),…, is given. Such a system of equations can be solved if the time dependent solution is required. But,
for many problems it suf ices to look at the steady state solution.

In the case of steady state, the boundary conditions are

So the equations (17.16) and (17.17) become,

(17.18)

(17.19)

The equations (17.18) and (17.19) constitute the system of steady state dif erence equations for this model.

To solve the system of dif erence equations

Since P0 = P0

… … …

(17.20)

Now, in order to find P0, use the fact that

 Operations Research

where (17.21)

The result obtained above is a general one and by suitably defining μn and λn many interesting cases could be
studied. Now two particular cases may arise:

Case 1

In this case, the series S becomes

Therefore, from equation 17.21 and 17.20

Here, it is observed that this is exactly the case of Model 1.

Case 2

The case, in which the arrival rate λn depends upon n inversely and the rate of service μ
n is independent of n,
is call ed the case of ‘Queue with Discouragement’.
In this case, the series S becomes

The equation 17.21 gives

… … …

It is observed in this case that Pn fol ows the Poisson distribution, where is constant, however ρ ˃ 1
or ρ < 1 but must be finite. Since the series S is convergent and hence sum able in both the cases.

17.7.3 Model II (B) : (M|M

(M M|1):(∞ |SIRO)
SIRO)
 Operations Research

This model is actual

actua y the same as Model I, except that the service discipline fol ows the Service in Random
Order (SIOR) rule in place of FCFS rule. Since the derivation of Pn in Model I is independent of any specific
queue discipline, so for the SIOR rule also

Consequently, the average number of customers in the customers in the system wil be the same whether queue
discipline fol ows SIRO rule or FCFS rule.
 Operations Research

Module 9. Project planning and network analysis

Lesson 18
INTRODUCTION AND BASIC DEFINITIONS IN NETWORK ANALYSIS

Introduction

A Project such as setting up of a new milk plant, research and development in an organization, development
of a new milk product, marketing of a product etc. is a combination of interrelated activities (tasks) which
must be executed in a certain order before the entire task can be completed. The activities are interrelated in
a logical sequence in such a way that same activities can not start until some others are completed. An activity
in a project usual y viewed as job requiring resources for its completion. The objectives of project
management can be described in terms of a successful project which has been finished on time, within the
budgeted cost and to technical specifications and to the satisfaction level of end users. Normal y for any
project, one may be interested in answering questions such as

i) What wil be the expected time of project completion?

i) ) What is the ef ect of delay of any activity on the overal completion of project?
ii) ) How to reduce the time to perform certain activities in case of availability of additional funds?
iv) What is the probability of completion of project in time?

The OR techniques used for planning, scheduling and control ing large and complex projects are often
referred to as network analysis. A network is a graphical representation consisting of certain configuration of
arrows and nodes for showing the logical sequence of various tasks to be performed to achieve the project
objectives. Around five decades ago the planning tool was Gantt bar chart which specifies start and finish
time for each activity on a horizontal time scale. The disadvantage is that there is no interdependency among
the many activities which control the progress of the project. Now-a-days we use a technical tool for
planning, scheduling and control ing stages of the projects known as Critical Path Method (CPM) and Project
Evaluation & Review Technique (PERT). The techniques of PERT and CPM prove extremely valuable in
assisting the mangers in handling such projects and thus discharging their project management responsibilities
both at planning and control ing stages of the projects. Commonly used project management techniques are:
a) Critical Path Method (CPM) and
b) Project Evaluation and Review Technique (PERT)

Historical Development

CPM/PERT or Network Analysis as the technique is sometimes cal ed, developed along two paral el
streams, one industrial and the other military. CPM was developed in 1957 by J. E. Kel y of Remington
Rand and M. R. Walker of E. I. Du Pont de Nemours & Co. PERT was devised in 1958 for the POLARIS
missile program by the Program Evaluation Branch of the Special Projects of ice of the U. S. Navy, helped
by the Lockheed Missile Systems division and the Consultant firm of Booz-Al en & Hamilton.

Both are basical y time oriented methods laid to determination of a time schedule for project. The major
dif erence between these two techniques is that PERT is a Probabilistic approach for the determination of
time estimates of dif erent activities not exactly known to us. In the case of CPM, dif erent estimates are
Visit Our Website - www.InstantAgriculture.com 111 Download Our App - Instant Agriculture
 Operations Research

known as they are deterministic in nature. But now a days both these techniques are used for one purpose.
Initial y the PERT technique was applied to research and development projects while the CPM was used
towards construction projects.

Methodology in CPM/PERT Technique

The methodology involved in network scheduling by CPM/PERT for any project consists of the fol owing
four stages:

Planning

It is started by splitting the total project into smal projects. The smal er projects are further divided into
dif erent activities and are analyzed by a department or section. The relationship of each activity with respect
to other activities are defined and established.

Scheduling

The objective of scheduling is to give the earliest and the latest al owable start and finish time of each activity
as wel as its relationship with other activities in the project. The schedule must pinpoint the critical path i.e.
time activities which require special attention if the project is to be completed in time.

Allocation of resources

Al ocation of resources is performed to achieve the desired objective. Resource is a physical variable such as
labour, finance, space, equipment etc. which wil impose a limitation for completion of a project.

Controlling

The final phase in the project management is control ing. After making the network plan and identification of
the Critical path, the project is control ed by checking progress against the schedule, assigning and scheduling
manpower and equipment and analyzing the ef ects of delays. This is done by progress report from time to
time and updating the network continuously. Arrow diagram and time charts are used for making periodic
progress reports.

Basic Terminology used in Network Analysis

Network analysis is the general name given to certain specific techniques which can be used for the planning,
management and control of projects. A fundamental method in both PERT and CPM is the use of network
systems as a means of graphical y depicting the current problems or proposed projects in network diagram.
A network diagram is the first thing to sketch an arrow diagram which shows inter-dependencies and the
precedence relationship among activities of the project. Before il ustrating the network representation of a
project, let us define some basic definitions:

Activity

Any individual operation, which utilizes resources and has a beginning and an end is cal ed an activity. An
arrow is used to depict an activity with its head indicating the direction of progress in the project. It is of four
types:
a) Predecessor activity: activity that must be completed immediately prior to the start of another
activity.
Visit Our Website - www.InstantAgriculture.com 112 Download Our App - Instant Agriculture
 Operations Research

b) Successor activity: activity which cannot be started until one or more of other activities are completed
but immediately succeed them are cal ed successor activity.
c) Concurrent: Activity which can be accomplished concurrently is known as concurrent activity. An
activity can be predecessor or successor to an event or it may be concurrent with the one or more of
the other activities.
d) Dummy activity: An activity which does not consume any kind of resources but merely depicts the
technological dependence is cal ed a dummy activity. Dummy activity is inserted in a network to classify
the activity pattern in the fol owing situations:
i) To make activities with common starting and finishing points distinguishable.
i) To identify and maintain the proper precedence relationship between activities those are not
connected by events.

Let’s consider a situation where A and B are concurrent activities and activity D is dependent on B and C is
dependent on both A and B. Such a situation can be handled by use of dummy activity.

When two or more activities are exactly paral el such that they would start at the same node (event) and finish
at the same node. A dummy would be inserted between the end of one of the activities and the common
finishing node.

This is to ensure that each activity has a unique description when refer to by its start and finish node number.
Dummy are often used to improve the layout of network. When they may not strictly necessary to represents
the logic involved. This often happens at the start or finish of a network where a number of activities either
start from a certain point or converge to particular point.

Event

Visit Our Website - www.InstantAgriculture.com 113 Download Our App - Instant Agriculture
 Operations Research

The beginning and end points of an activity are cal ed events or nodes or connector. This is usual y
represented by circle in a network.

Here, A is known as the activity.

The events can be further classified into three categories:
a) Merge Event: When two or more activities come from an event it is known as merge event.
b) Burst Event: When more than one activity leaves an event is known as burst event.
c) Merge & Burst Event: An activity may be merged and burst at the same time.

Difference between event and activity

An event is that particular instant of time at which some specific part of project is to be achieved while an
activity is the actual performance of a task. An activity requires time and resources for its completion. Events
are general y described by such words as complete, start, issue, approves, taste etc. while the word like
design, process, test, develop, prepare etc. shows that a work is being accomplished and thus represent
activity. While drawing networks, it is assumed that
a) The movement is from left to right and
b) Head event has a number higher than the tail event.
Thus the activity (i-j) always means that job which begins at event (i) is completed at event (j).

Network representation is based on the fol owing two axioms.

a) An event is not said to be complete until al the activities flowing into it are completed.
b) No subsequent activities can begin until its tail event is reached or completed.

Visit Our Website - www.InstantAgriculture.com 114 Download Our App - Instant Agriculture
 Operations Research

Module 9. Project planning and network analysis

Lesson 19
RULES FOR DRAWING NETWORK ANALYSIS

Introduction

A fundamental ingredient in both PERT and CPM is the use of network systems as a means of graphically depicting a project.
When a network is being constructed, certain conventions are followed to represent a project graphically, for it is essential that the
relationship between activities and events are correctly depicted. Drawing a network diagram is a relatively easy task, and can be
accomplished by listing each task on a piece of paper and representing the sequence in which the tasks take place. Lines and
arrows are drawn between the pieces of paper to show which tasks follow on from others. The diagram aims to portray how the
tasks relate to one another i.e. which tasks have to be completed before others begin and which tasks can be performed
simultaneously. Creating the network is an iterative process and may involve a number of revisions before an optimum solution is
found.

Sequencing

The initial step in project scheduling process is the determination of all specific activities that comprise the project and their
interdependence relationships. In order to make a network following points should be taken into consideration.
What job or jobs precede it?
What job or jobs run concurrently?
What job or jobs follow it?
What controls the start and finish of a job?
Since all further calculations are based on the network, it is necessary that a network be drawn with full care. There are many
ways to draw a network, in this lesson we will describe the method which follows the precedence table. The following example of
preparation of Paneer (Cottage cheese) shows the basic steps required in drawing a network.

Example 1:

For preparation of Paneer (Cottage Cheese) the following list represents major activities
i) Receive whole cow/buffalo milk
ii) Standardize milk to obtain desired level of fat percentage
iii) Take citric acid and prepare 1% solution
iv) Heat the citric acid to 70 o C
v) Bring the standardized milk to boil on medium heat
vi) Cool the milk to 70 o C and add slowly the solution of citric acid till yellowish whey separates
vii) Strain the mixture through a clean muslin cloth.
viii) Hold it under running water for a minute and then press out the excess water.
ix) Hang the muslin for 15-20 minutes so that all the whey is drained out.
x) Prepare mould to form Paneer block
xi) Fill the mass into the block and tie the muslin
xii) Place it under something heavy for up to two hours
xiii) Cut the paneer into chunks and used as required.
Based on above list of different activities a precedence table may be formed which is given in Table 19.1

Table 19.1 Precedence table

Activity Description Preceding Activity

A Receive whole cow/buffalo milk -

B Standardize milk to obtain desired level of fat percentage A
C Take citric acid and prepare 1% solution -

Visit Our Website - www.InstantAgriculture.com 115 Download Our App - Instant Agriculture
 Operations Research
D Heat the citric acid to 70 o C C
E Bring the standardized milk to boil on medium heat B
F Cool the milk to 70 oC and add slowly the solution of citric acid D,E
till yellowish whey separates.
G Strain the mixture through a clean muslin cloth. F
H Hold it under running water for a minute and press out the G
excess water.
I Hang the muslin for 15-20 minutes and drain out all the whey. H
J Prepare mould to form Paneer block H
K Fill the mass into the block and tie the muslin J
L Place it under something heavy for up to two hours. K
M Cut the paneer into chunks and use as required. L

In the above table due consideration has been given to precedings of an activity. While drawing the network, other factors will be
considered. The activity A has no preceding activity and it is represented by an arrow line (Fig. 19.1). Likewise activity C has no
preceding activity and both activities A and C can be done simultaneously so they are shown as concurrent activities. Activities B
and D are preceded by the activities A and C respectively. The complete network is shown in Fig 19.1

Fig. 19.1 Network diagram

Guidelines for Drawing Network Diagram

There are number of rules in connection with the handling of events and activities of a project network which are given below:

a) Each activity is represented by one and only one arrow in the network. This implies that no single activity can be
represented twice in the network. This is to be distinguished from the case where one activity is broken into segments. In
such a case each segment may be represented by a separate arrow.
b) No two activities can be identified by the same beginning and end event. In such cases, a dummy activity is introduced to
resolve the problem as shown in Fig. 19.2

Fig. 19.2 Using a dummy activity

c) In order to ensure the correct precedence relationship in arrow diagram following question must be checked whenever

Visit Our Website - www.InstantAgriculture.com 116 Download Our App - Instant Agriculture
 Operations Research
any activity is added to a network.
What activity must be completed immediately before this activity can start?
What activities must follow this activity?
What activities must occur simultaneously with this activity?
d) Thus a network should be developed on the basis of logical or technical dependence.
e) The arrows depicting various activities are indicative of logical precedence only; hence length and bearing of the arrows
are of no significance.
f) The flow of the diagram should be from left to right.
g) Two events are numbered in such a way that the event of higher number can happen only after the event of lower number
is completed.
h) Arrows should be kept straight and not curved. Avoid arrow which cross each other.
i) Avoid mixing two directions vertical and standing arrows may be used ifnecessary.
j) Use dummy activity freely in rough graph but final network should have only reluctant dummy.
k) The network has only one entry point called the start event and one point of emergence called end event.
l) Angle between the arrows should be as large as possible.

Error in Drawing Network

There are three types of errors which are common in network diagrams

Dangling error

To disconnect an activity before the completion of all activities in a network diagram is known as dangling.

Fig. 19.3 Dangling error

In Fig. 19.3 the activity 5 to 8, 6 to 7 are known as dangling error. These are not last activities in the network.

Looping error

Looping error is also known as cyclic error in the network. Drawing an endless loop in a network diagram is known as error of
looping as shown in Fig. 19.4

Fig. 19.4 Appearance of a loop in the network

Visit Our Website - www.InstantAgriculture.com 117 Download Our App - Instant Agriculture
 Operations Research

Reductancy Error:
Unnecessarily inserting the dummy activity in a network diagram is known as error of reductancy as shown in Fig 19.5 in which
putting an dummy activity from 10 to 12 is a reductancy error.

Fig. 19.5 Reductancy error

19.5 Labeling of a Network Diagram

For network representation it is necessary that various nodes be properly labeled. For convenience, labeling is done on a network
diagram. A standard procedure called i-j rule developed by D.R.F Fulkerson is most commonly used for this purpose.
Fulkerson’s i-j Rule:
Step 1: First, a start event is one which has arrows emerging from it but not entering it. Find the start event and label it as
number1.
Step 2: Delete all arrows emerging from all numbered events. This will create at least one new start event out of the preceding
events.
Step 3: Number all new start events ‘2’, ‘3’ and so on. No definite rule is necessary but numbering from top to bottom may
facilitate other users using the network when there are more than one new start event.
Step 4: Go on repeating step no. 2 & 3 until the end reached.
These rules are illustrated by taking into consideration the Example 19.1 and network diagram as shown in Fig.19.6

Fig. 19.6 Network diagram of preparation of paneer

Visit Our Website - www.InstantAgriculture.com 118 Download Our App - Instant Agriculture
 Operations Research

Module 9. Project planning and network analysis

Lesson 20
CRITICAL PATH METHOD (CPM)

Introduction
After the project network plan is constructed and activity times are known, the time analysis of the network
becomes essential for planning various activities of the project as wel as obtaining answers to questions like
when the various activities are scheduled to be performed, how long it wil take the project work to be
completed and what are the crucial activities. In this lesson we wil learn about computation of time estimates and
determination of critical path.

Time Estimate Analysis

Time estimates analysis is a critical path in network analysis once the network of a project is constructed. Time
analysis of network becomes essential for planning various activities of the project. An activity time is a forecast
of the time and activity expected today from its starting point to it completion point (under normal conditions).
The main objective of time analysis is to prepare a planning schedule of a project. Planning schedule should
include the fol owing factors.
i) Total completion time of the project.
i ) Earliest time and each activity start.
ii) ) Latest time each activity can be started without delaying the total project
iv) Float for each activity i.e. the amount of time by which the completion of an activity can be delaying the
total project completion
v) Identification of critical activities and critical path.

Basic Notations
The fol owing notations for basic scheduling computations wil be used
( i−j) = Activity (i, j ) with tail event i and head event j.
TE or Ei = Earliest occurrence time of event (j).
TL or Lf = Latest al owable occurrence time of event(j).
Di j = Estimated completion time of activity (i, j).
(ES) i j = Earliest starting time of activity (i, j)
(Ef) i j = Earliest finishing time of activity (i, j).
(Ls)i j = Latest start time for activity (i, j).
(Lf) i j = Latest finish time for activity (i, j).

Forward Pass Computation (For Earliest Event Time)

Before starting computations, the occurrence time of the initial network event is fixed. The forward pass
computation computes the earliest start time (Es) and earliest finish time (Ef) for each activity. The earliest time
indicates the earliest time that a given activity can be scheduled and earliest finish time indicates the time by
which the activity can be completed at the earliest. This is done in fol owing three steps:

Step 1: The computations begin from the start node and move towards the end node. For the convenience in

Visit Our Website - www.InstantAgriculture.com 119 Download Our App - Instant Agriculture
 Operations Research

the forward pass computation start with an assumed earliest occurrence time equal to zero for the
initial project events i.e. E1=0

Step 2: i) Earliest starting time of activity (i, j) is the earliest event time of the tail event i.e.
ii) Earliest finish time of activity (i, j) is the addition of earliest starting time and the activity time i.e.

Step 3: Earliest event time for activity j is the maximum of the earliest finish time of al activities ending into that
event. That is

The computed values of Ei are put over the respective circles representing each event.

Backward Pass Computations (For latest allowable time)

The idea of the backward pass is to compute the latest al owable times of starting and finishing of each of the
activities of a project without delaying the completion of the project. These can be computed by reversing the
method of calculation used for earliest event time. This is done by using fol owing steps
Step 1: For ending event it is presumed that E = L where al E’s are computed by previous method.
Step 2: Latest finish time for activity (i, j) is equal to the latest event time of event j. i.e., (Lf) ij=L.
Step 3: Latest starting time for activity (i, j) is equal to latest completion time of (i, j) minus activity time (Ls)ij =
(Lf)ij – Dij
Step 4: Latest event time for event i is a minimum of the latest start time of al activities originating from that
event.

Al the computed L values are put over respective circles representing each event.

Determination of Floats and Slack Times

When the network diagram is completely drawn, properly labeled and earliest (E) and latest (L) event times are
computed, the next objective is to determine the floats and slack times defined as fol ows. There are three kinds
of floats as given below :

Total float
The amount of time by which the completion of an activity could be delayed beyond the earliest expected
completion time without af ecting the overal
overa project duration time. In other words, total float of an activity (i-j) is
the dif
di erence between latest start time and earliest start time of that activity. Hence total float for activity (i-j) ,
denoted by (Tf )ij is given as
(Tf)ij = (Latest start-Earliest
Earliest start) time for activity (i-j)
(Tf)ij = (Ls)ij – (Es)ij = (Lf – Dij) - Ei
 Operations Research

It refers to the amount of free time associated with an activity which can be used before, during or after the
performance of this activity. This is the most important type of float because this is concerned with the overal
project duration. Total float on critical activities is always taken as zero. The value of total floats for any activity
is useful for drawing the folowing conclusions.

Free float
The time by which the completion of an activity can be delayed beyond the earliest finish time without af ecting
the earliest start of a subsequent (succeeding) activity. This is that value of the float which is consumable when
the succeeding activities are started at their earliest starting times. Mathematical y, free float for activity (i-j)
denoted by (Ff)ij is calculated as (Ff)ij = (Ej – Ei) –Dij
Free float for (i-j)=(Earliest time for event j–Earliest time for event i)– Activity time for (i-j)
Thus free float is concerned with the commencement of subsequent activity.
(Tf)ij = (Lj− Ei) - Dij ,but Lj ≥ Ei as latest event time is always greater than equal to earliest event time.
Therefore for al activities free float can take values from zero up to total float but wil not exceed total float. Free
float is always useful for rescheduling the activities with minimum disruption of earliest plan.
Independent float
The amount of time by which the start of an activity can be delayed without af ecting the earliest start time of any
immediately fol owing activities, assuming that the preceding activity has finished at its latest finish time.
Mathematical y, independent float of an activity (i ,j) denoted by (If)ij can be calculated by the formula (If)ij =(Ej
– Ei) – Dij. The negative independent float is always taken as zero. This float is concerned with prior &
subsequent activity. The independent float thus provides a measure of variation in starting time of a job without
af ecting preceding and succeeding activities.
Note:

 It can be observed that Independent float ≤ Free float ≤ Total float.

 The concept of float is useful for the management in representing underutilized resources and flexibility
of the schedule and the extent to which the resources wil be utilized on dif erent activities.

 Float can be used for redeployment of resources to label the same or to reduce the project duration.
Whenever a float in a particular activity is utilized the float of not only that activity but that of other
Visit Our Website - www.InstantAgriculture.com 121 Download Our App - Instant Agriculture
 Operations Research

activities would also change.

Slack of an Event

The basic dif erence between slack and float times is that slack is used for events only whereas float is applied
for activities. For any given event, the event slack is defined as the dif erence between the latest event and
earliest event times. Mathematical y, for a given activity (i-j)

Head event slack = Lj – Ej and Tail event slack = (Li – Ei)

Al the floats defined earlier can be defined in terms of head or tail events slack as under:
Total float = Lj – Ei – Di
Free float = (Ej – Ei – Dij) = (Lj – Ei – Dij) – (Lij – Eij) = Total float – Head event slack

Independent float =Ej – Li – Dij= (Ej – Ei – Dij)–

(Li–Ei)

= Free float – Tail event slack

Determination of Critical Path

After determining the earliest and latest scheduled times for various activities, the next step is to find the minimum
time required for the completion of whole project. Before defining this let us first discuss about the meaning of
critical event and critical activity.

Critical event
The slack of an event is the dif erence between latest and earliest event time i.e. Slack (i) = Li – Ei. The event
with zero slack time is cal ed critical event. In other words, the event (i) is said to be critical when Li = Ei

Critical activity
Since the dif erence between the latest Start time & earliest start time of an activity is usual y cal ed as total float.
Activity with zero total float are known as critical activities. In other words, an activity is said to be critical if its
delay in its start wil cause a further delay in the completion date of entire project.

Non-critical activity
A non-critical activity is such that the time between its earliest start and its latest completion date is longer than its
actual duration.

Critical path
The sequence of critical activity in a network is cal ed a critical path. This path is the longest path in the network
from the starting event to the end of event and defines the minimum time required to complete the project. The
term path is defined as a sequence of activities such that it begins at the starting event and end at the final event.

Visit Our Website - www.InstantAgriculture.com 122 Download Our App - Instant Agriculture
 Operations Research

The length of the path is the sum of the individual time of the activities lying on the path. If the activities on critical
path are delayed by a day, the project would also be delayed by a day unless the time of the future critical
activity is reduced by a day by dif erent means. The critical path is denoted by double or darker lines in order to
distinguish from the other non criticalpath.

Main features of critical path

 If the project has to be shortened then some of the activities on that path must also be shortened. The
application of additional resources on other activities wil not give the desired results unless that critical
path is shortened first.

 The variation in actual performance from the expected duration time wil be completely reflected in one to
one fashion in anticipation completion of the whole project.

 It plays an important role in scheduling and control ing large projects.

 It identifies al the critical activities of the project.

The computation procedure used for the time analysis of the project is described in the examples 1and 2

Example 1
Draw a network diagram of the fol owing schedule of activities and find its critical path. Also calculate slack time
for each event

Activity 1-2 1-3 1-4 2-6 3-7 3-5 4-5 5-9 6-8 7-8 8-9
Duration 2 2 1 4 5 8 3 5 1 4 3
(in days)

Solution :
First construct the network diagram which is depicted in fig. 20.1

Visit Our Website - www.InstantAgriculture.com 123 Download Our App - Instant Agriculture
 Operations Research

Fig. 20.1 Network diagram

To determine the critical path, compute the earliest start E i and latest finish Lj for each activity (i, j)
Forward Pass Computation (For earliest time event): As shown in fig. 20.1 assuming E1=0 and
E2=E1+D12=0+2=2, E3=E1+D13=0+2=2 and E4=E1+D14=0+1=1

E6=E2+D26=2+4=6, E7=E3+D37=2+5=7

From this computation it can be inferred that this project wi

wil take 15 days to complete.
Backward Pass computations (For latest allowable time) In backward computation method assign the latest
al owable time determined in forward pass computation method i.e. put L9=15

L8=L9-D98=15-3=12,
3=12, L6=L8-D86=12-1=11,
1=11, L7=L8-D
- 87=12-4=8
4=8

L5=L9-D95=31-6=25,
6=25,

L4=L5-D54=10--3=7 L2=L6-D
- 62=11-4=7
4=7

The path 1-3-5-99 is the critical path which is shown in Fig. 20.1 with double lines joining all those events where
Ei=Lj The total duration of project is equal to 2+8+5=15 days
Computation of Float: For each non –critical
critical activity, the total float, free float and independent float calculations
are given Table 20.1

Table 20.1 Calculations of time estimates and floats

Activity Duration Start Finish Float
(i-j) Dij
(1) (2) Earliest Latest Earliest (5) Latest Total (7) Free (8) Independent (9)=
(3) (4)= (6) =(4) - (3) =(5) - (3) (8) - [(3)-(2)]
[(3)
-(2)
= (3) + (2)
(6)−
− (2)

1-2 2 0 5 2 7 5 0 0
1-3 2 0 0 2 2 0 0 0
 Operations Research

1-4 1 0 6 1 7 6 0 0
2-6 4 2 7 6 11 5 0 0
3-7 5 2 3 7 8 1 0 0
3-5 8 2 2 10 10 0 0 0
4-5 3 1 7 4 10 6 6 0
5-9 5 10 10 15 15 0 0 0
6-8 1 6 11 7 12 5 4 0
7-8 4 7 8 11 12 1 0 0
8-9 3 11 12 14 15 1 1 0

Example 2:
Draw the network diagram for the fol owing project and find the critical path and maximum time for completion
of the project.

Activity A B C D E F G H I J K L
Preceded by - A A B B C C F D G, E I
H
Duration 10 9 7 6 12 6 8 8 4 11 5 7
(weeks)

Solution :
Network diagram of the above problem is shown in Fig. 20.2.

Visit Our Website - www.InstantAgriculture.com 125 Download Our App - Instant Agriculture
 Operations Research

Fig. 20.2 Network diagram

To determine the critical path, compute the earliest start Ei and latest finish time Lj for each activity (i,j).The
calculations are given below

Forward Pass Computation (For earliest time event): As shown in fig 20.2 assuming E1=0 and
E2=E1+D12=0+10=10, E3=E2+D23=10+9=19, E4=E2+D24=10+7=17, E5=E3+D35=19+6=25,
E6=E3+D36=19+12=31, E7=E4+D47=17+6=23,

E9=E5+D59=25+4=29,

From this computation it can be inferred that this project wi

wil be completed in 42 weeks.

Backward Pass computations (For latest allowable time) in backward computation method assign the latest
al owable time determined in forward pass computation method i.e. put L10=42

L9=L10-D109=42-7=35,
7=35, L6=L10-D106=42-5=37,
5=37, L8=L10-D
D108=42-11=31,
11=31,

L5=L9-D
D95=35-4=31,
4=31, L7=L8-D
D87=31-8=23
8=23

=17

L1=L2-D
D21=10-10=0
10=0

The path 1-2-4-77-8-10 is the critical path which shown in Fig. 20.2 with double lines joining al those events
where Ei=Lj The total duration of project is equal to 10+7+6+8+11=42 weeks

Activity Duration Start Finish Float

(i-j) Dij Earliest Latest (4) Earliest (5) Latest Total (7) Free (8)= Independent (9)=
(1) (2) (3) (6) =(4) - (3) (5) -(3)
(3) - (8) - [(3)
[(3)-(2)]
 Operations Research
(2)

= (6)− (2) =(3)+ (2)

A(1-2) 10 0 0 10 10 0 0 0
B(2-3) 9 10 16 19 25 6 0 0
C(2-4) 7 10 10 17 17 0 0 0
D(3-5) 6 19 25 25 31 6 0 0
E(3-6) 12 19 25 31 37 6 0 0
F(4-7) 6 17 17 23 23 0 0 0
G(4-8) 8 17 23 25 31 6 6 0
H(7-8) 8 23 23 31 31 0 0 0
I(5-9) 4 25 31 29 35 6 2 0
J(8-10) 11 31 31 42 42 0 0 0
K(6-10) 5 31 37 36 42 6 6 0
L(9-10) 7 29 35 36 42 6 6 0

Critical Path Method (CPM)

Critical Path Method (CPM) , was developed by M.R.Walker and J. E. Kel y. They came up with arrow
diagram as the most logical representation of the interrelationships between the jobs in a project to be executed
in a wel defined sequence. The arrow diagram designed by them, as wel as the method of calculating the critical
path are the same as in PERT network, except that they used the single time estimate and did not enter the
problem of uncertainty of the duration of time for the individual jobs.
CPM emphasizes the relationship between applying more men or other resources to shorten the duration of
given jobs in a project and the increased cost of these additional resources. With CPM the amount of time
needed to complete various parts of the project is assumed to be known with certainty. Moreover, the relation
between the amount of resources employed and the time needed to complete the project is also assumed to be
known. The interactive procedure of determining the critical path involves the folowing steps:

i) Break down the project into various activities systematical y. Label al activities. Arrange al the activities
in logical sequence. Construct the arrow diagram.

i) ) Number al the nodes (events) and activities. Find the time for each activity considering it to be
deterministic. Indicate the activity times on the arrow diagram.

ii) ) Calculate earliest start time, earliest finish time, latest start time and latest finish time. Tabulate activity
normal times, earliest time and latest time.

iv) Determine the total float for each activity by taking dif erence between the earliest time and the latest time
for each node.

v) Identify the critical activities (the activities with zero float) and connect them with the beginning node and
the ending node in the network diagram by double line arrow. This gives the criticalpath.

vi) Calculate the total project duration.

Visit Our Website - www.InstantAgriculture.com 127 Download Our App - Instant Agriculture
 Operations Research

Module 9. Project planning and network analysis

Lesson 21
PROJECT EVALUATION AND REVIEW TECHNIQUE (PERT)

Introduction

The network method discussed so far may be termed as deterministic, since estimated activity times are
assumed to be known with certainty. While this assumption holds for the CPM analysis. In most of the
projects, these activity times are random variables. A new technique known as Project Evaluation and
Review Technique (PERT) was devised in 1958 for the POLARIS missile program by the Program
Evaluation Branch of the Special Projects of ice of the U. S. Navy, helped by the Lockheed Missile Systems
division and the Consultant firm of Booz-Al en & Hamilton.

Project Evaluation & Review Technique (PERT)

In research project of designing a new machine or development of a new dairy product, various activities to
be performed are based on judgment. A reliable time estimate is dif icult to get because the technology is
changing rapidly. Time values are subjected to variation. The main objective of the analysis through PERT is
to find out the completion for a particular event within specified date. What are the chances of completing the
job? This approach takes into account uncertainties. In this approach three time values are estimated with
each activity: Optimistic time, most likely time and Pessimistic time. The three time values provide a measure
of uncertainty associated with that activity.

Optimistic time

It is the shortest possible time in which the activity can be finished and assumes that everything goes very wel .
In other words, it is the estimate of the minimum possible time, which an activity takes to complete under ideal
conditions i.e. no provision are made for breakdown, delays etc. They are general y denoted by (t0) or (a).

Most likely time

This is estimate of the normal time the activity would take. This assumes normal delays. It is denoted by (tm)
or (m). If a graph is plotted between the time of completion and frequency of completion in that period, the
highest frequency of occurrence is denoted by most likely time as shown in Fig. 21.1.

Visit Our Website - www.InstantAgriculture.com 128 Download Our App - Instant Agriculture
 Operations Research

Fig. 21.1 Time distribution curve

Pessimistic time

The longest time, the activity could take if everything goes wrong. In other words, it is the longest time the
activity can conceivably take. This is general
genera y denoted by (tp) or (b).
The three time values are shown in Fig. 21.1.
The PERT technique makes the fo
fol owing assumptions:
 Activity times are statistical
statistica y independent and usual y associated with ‘beta’ distribution.
 There are enough activities involved in the network and totals of activity times based on their means
and variances wil be normal y distributed.
 The three estimates of the activity duration can be obtained for each activity.
In PERT calculation, al values are used to obtain the expected value.

Estimated time

This is the average time an activity wi

wil take if it is to be repeated large number of times and is based on the
assumption that the activity time fol ows Beta distribution.

Variance

Variance of each activity is given by formula:

where to (a) , tp (b) and tm(m) are optimistic , pessimistic and mostly likely times respectively.

Once the expected times of the activities are obtained, the critical path of the project network is determined
using three time estimates. Having found the critical path, the PERT methodology assumes that the
 Operations Research

aggregation of the mean times and the summation of the variances of critical jobs would yield the expected
project duration and its variance. PERT uses the variance of critical path activity to help in determining the
variance of the overal
overa project. Project variance is computed by summing variance of just critical activities.

PERT Algorithm

The various steps involved in the PERT network for analyzing any project are summarized below:

i) Develop a list of activities involved in the project including the immediate predecessors.

i ) Draw the network diagram using the rules and conventions as discussed before.

ii)
ii ) Number the events in ascending order from left to right

iv)
iv From the three time estimates compute the expected time (te) for each activity using the formula

Using the expected activity time estimates, determine the earliest start time and earliest finish time for
each activity .

v)
v Compute the latest start time and latest finish time and the float associated with each activity. Find the
activities with zero total float which are known as critical activities. From these critical activities find
the critical path.

vi Using the value for tp (b) and to(a) , calculate the variance (σ2) by using the formula
vi)

vii ) Use the variability in the activity times to estimate the variability of the project completion date; using
this estimate compute the probability of meeting a specified date by using the standard normal
equation

where Z is a standard normal variate

Illustrative Examples on PERT

The computation procedure used for PERT is described in the examples 1and 2

Example 1:

A smal
sma project is composed of nine activities whose time estimates are listed in the fo
fol owing table:
 Operations Research

Activity t0 tp tm
1-2 5 10 8
1-3 18 22 20
1-4 26 40 33
2-5 16 20 18
2-6 15 25 20
3-6 6 12 9
4-7 7 12 10
5-7 7 9 8
6-7 3 5 4

a)
a Find the expected task time and their variance.
b)
b Earliest and latest expected time of each node.
c)
c Critical path
d)
d Probability that project wil complete in 41.5 weeks and 44weeks.

Solution:

The expected task time and variances of di

dif erent activities are computed by the fo
fol owing formulae:

and

and these values are given in fo

fol owing table

Activity t0 tp tm

1-2 5 10 8 47/6 25/36

1-3 18 22 20 20 16/36
1-4 26 40 33 33 196/36
2-5 16 20 18 18 16/36
2-6 15 25 20 20 100/36
3-6 6 12 9 9 36/36
4-7 7 12 10 59/6 25/36
5-7 7 9 8 8 4/36
6-7 3 5 4 4 4/36

Construct the network diagram showing earliest and latest expected time of each node.
 Operations Research

Rule to find out variance of Events:-

Events:

We take initial value of variance

, ,

NOTE: At merge point the variance is computed along the longest path in case of two path having the same
length the larger of the two variance of that event.

Critical path is 1--4-7 and total duration is 257/36 = 42.833 weeks with variance = 221/36

Now probability that project wil complete in 41.5 weeks is

 Operations Research

Hence, there are 29.81% chances that project wil be completed before 41.5 weeks.
Now probability that project wil complete in 44 weeks is

Hence, there are 99.81% chances that project wil be completed before 44 weeks.

Example 2

The fol
fo owing table gives the estimates of optimistic time (t0), most likely time (tm) and pessimistic time (tp) of
dif erent activities of a project.

Activity t0 tm tp
1-2 4 8 12
2-3 1 4 7
3-4 8 12 16
3-5 3 5 7
4-5 0 0 0
4-6 3 6 9
5-7 3 6 9
5-8 4 8 6
6-10 4 6 8
7-9 4 8 12
8-9 2 5 8
9-10 4 10 16

i) Construct the network diagram when it is given that scheduled completion is 40 days.

i ) Calculate the probability of finishing the project

a)
a within the scheduled time
b)
b less than 45days
c)
c less than 38 days.

Solution :

The expected task time and their variances are given in fo

foll owing table
 Operations Research

Activity t0 tm tp te σ2
1-2
2 4 8 12 8 64/36
2-3
3 1 4 7 4 36/36
3-4
4 8 12 16 12 64/36
3-5
5 3 5 7 5 16/36
4-5
5 0 0 0 0 0
4-6
6 3 6 9 6 36/36
5-7
7 3 6 9 6 36/36
5-8
8 4 8 6 7 4/36
6-10
10 4 6 8 6 16/36
7-9
9 4 8 12 8 64/36
8-9
9 2 5 8 5 36/36
9-10
10 4 10 16 10 144/36

From the table it is clear that the activity 4

4-5
5 is dummy activity.

The network diagram with earliest time and latest times are given below

Rule to find out variance of Events:

We take initial value of variance

, ,
 Operations Research

The critical path is 1-2-3-5--7-9-10 with expected time for completion of the project is 41 days with variance
=10.
a)
a Probability that project wil complete in given time i.e. 40 days is

Hence, there are 37.83% chances that project wi

wil be completed in the stipulated time of 40 days.
b)
b Probability that project wil complete in 45 days is

Hence , there are 89.62% chances that project wi

wil be completed before 45 days.
c)
c Probability that project wil complete in 38 days is

Hence ,there are only 17.36% chances that project wi

wil be completed before 38 days.
 Operations Research

Comparison between PERT and CPM

As stated earlier both PERT and CPM techniques were developed independently with dif erent set of
objectives. However, the basic dif erences between the two are given below:

PERT CPM
1. It is probabilistic model with uncertainty in A deterministic model with wel known activity
activity duration. The duration of each activity is (single) time based upon the past experience. It
normal y computed from multiple time estimates. does not deal with uncertainty with time.
2. PERT is said to be an event oriented as the It is an activity oriented as its results are
result of analysis are expressed in terms of calculated on the basis of activities.
events.
3. It uses dummy activities to represent project It does not make use of dummy activities to
sequencing of the activities. represent the project sequencing.
4. PERT is usual y used for those projects where This is commonly used for those projects which
time required to complete various activities is not are repetitive in nature and here one has prior
known a priori. experience of handling similar projects.
5. PERT is general y applied for planning and CPM is general y used for construction and
scheduling research program and developing business problems.
projects.
6. PERT analysis usual y does not consider cost. CPM deals with the cost of project schedules
and their minimization.
7. PERT is an important control device as it assists It is dif icult to use CPM as control ing device
the management in control ing a project by because it requires repetition of the entire
constant review of such delays in the activities. evaluation of project each time the changes are
introduced in the network.
8. PERT helps the manager to schedule and CPM plans dual emphasis on time cost and
coordinate various activities so that project can evaluates the tradeof between project cost and
be completed on scheduled time. time.
9. It makes use of the statistical devices in the It does not make use of the statistical devices in
determination of time estimates. the determination of time estimates

Visit Our Website - www.InstantAgriculture.com 136 Download Our App - Instant Agriculture
 ☺******
******
This Book Download From e-course of
ICAR Visit for Other Agriculture books,
News, Recruitment, Information, and
Events at www.InstantAgriculture.com

Disclaimer:
The information on this website does not
warrant or assume any legal liability or
responsibility for the accuracy,
completeness or usefulness of the
courseware contents.
The contents are provided free for
noncommercial purpose such as teaching,
training, research, extension and self
learning.
******☺******
.