1 - M - Com - (F&C) - 335 21 - Operations Research PDF

ALAGAPPA UNIVERSITY
[Accredited with ‘A+’ Grade by NAAC (CGPA:3.64) in the Third Cycle

and Graded as Category–I University by MHRD-UGC]
(A State University Established by the Government of Tamil Nadu)
KARAIKUDI – 630 003
M.Com. [Finance & Control]

II - Semester
335 21
OPERATIONS RESEARCH
Authors
CR Kothari, Ex-Associate Professor, Depatment of Economic Administration and Fin. Management, University of Rajasthan
Units: (1.0-1.1, 1.3-1.11, 2.0-2.2.3)
S Kalavathy, Dean, Department of Science and Humanities, K Ramakrishnan College of Technology, Samayapuram, Trichy
Units: (1.2, 2.2.4-2.7, 3-14)
"The copyright shall be vested with Alagappa University"
All rights reserved. No part of this publication which is material protected by this copyright notice
may be reproduced or transmitted or utilized or stored in any form or by any means now known or
hereinafter invented, electronic, digital or mechanical, including photocopying, scanning, recording
or by any information storage or retrieval system, without prior written permission from the Alagappa
University, Karaikudi, Tamil Nadu.
Information contained in this book has been published by VIKAS® Publishing House Pvt. Ltd. and has
been obtained by its Authors from sources believed to be reliable and are correct to the best of their
knowledge. However, the Alagappa University, Publisher and its Authors shall in no event be liable for
any errors, omissions or damages arising out of use of this information and specifically disclaim any
implied warranties or merchantability or fitness for any particular use.
Vikas® is the registered trademark of Vikas® Publishing House Pvt. Ltd.

VIKAS® PUBLISHING HOUSE PVT. LTD.
E-28, Sector-8, Noida - 201301 (UP)
Phone: 0120-4078900  Fax: 0120-4078999
Regd. Office: 7361, Ravindra Mansion, Ram Nagar, New Delhi 110 055
 Website: www.vikaspublishing.com  Email: [email protected]
Work Order No. AU/DDE/DE1-291/Preparation and Printing of Course Materials/2018 Dated 19.11.2018 Copies - 500
SYLLABI-BOOK MAPPING TABLE
Operations Research
BLOCK - I: OPERATIONS RESEARCH AND LINEAR PROGRAMMING

PROBLEM CONCEPT
Unit 1: Operations Research: Historical Background, Scope of Operations Unit 1: Operations Research
Research, Features of Operations Research, Phases of Operations (Pages 1-32)
Research, Types of Operations Research Models, Operations Research
Techniques and Tools, Structure of the Mathematical Model, Limitations
of Operations Research.
Unit 2: Linear Programming: Linear Programming Problem, Requirements Unit 2: Linear Programming
of LPP, Mathematical Formulation of LPP, Case Studies of LPP, Graphical (Pages 33-48)
Methods to Solve Linear Programming Problems, Applications,
Advantages, Limitations.
Unit 3: Graphical Analysis of Linear Programming Problems: Graphical Unit 3: Graphical Analysis
Analysis, Some Basic Definitions, Graphical Methods to Solve LPP, Some of LPP
Exceptional Cases, Important Geometric Properties of LPP. (Pages 49-62)
Unit 4: Simplex Method: Standard Form of LPP, Fundamental Theorem of Unit 4: Simplex Method
LPP, Solution of LPP - Simplex Method, The Simplex Algorithm, Penalty (Pages 63-92)
Cost Method or Big M-method, Two Phase Method, Solved Problems on
Minimisation.
Unit 5: Duality in Linear Programming Problem: Importance of Duality Unit 5: Duality in Linear
Concepts, Formulation of Dual Problem, Economic Interpretation of Duality, Programming Problem
Sensitivity Analysis. (Pages 93-106)
Unit 6: Transportation Problem: Formulation of Transportation Problem Unit 6: Transportation Problem

(TP), Transportation Algorithm (MODI Method), the Initial Basic Feasible (Pages 107-125)
Solution, Moving Towards Optimality.
Unit 7: Assignment Problem: Mathematical Formulation of the Problem, Unit 7: Assignment Problem
Hungarian Method Algorithm, Routing Problem, Travelling Salesman (Pages 126-150)
Problem.
BLOCK - II: INTEGER PROGRAMMING AND QUEUING CONCEPTS

Unit 8: Integer Programming Problem: Types of Integer Programming Unit 8: Integer Programming Problem
Problems, Gomory’s All-IPP Method, All IPP Algorithm, Branch and Bound (Pages 151-181)
Technique.
Unit 9: Infinite Queuing Models: Queuing Theory, Operating Characteristics Unit 9: Infinite Queuing Models
of a Queuing System, Constituents of a Queuing System, Service Facility, (Pages 182-189)
Queue Discipline.
Unit 10: Mathematical Analysis of Queuing Theory: Mathematical Analysis Unit 10: Mathematical Analysis
of Queuing Process, Properties of Queuing System, Notations, Service of Queuing Theory
System, Single Channel Models, Multiple Service Channels, Erlang Family (Pages 190-210)
of Distribution of Service Times, Applications of Queuing Theory,
Limitations of Queuing Theory.
Unit 11: Finite Queuing Models: Finite Queuing Models - Simulation: Unit 11: Finite Queuing Models
Methodology of Simulation, Basic Concepts, Simulation Procedure, (Pages 211-218)
Application of Simulation.
BLOCK - III: SIMULATION GAME THEORY CONCEPTS

Unit 12: Simulation Monte-Carlo Method: Monte-Carlo Simulation, Unit 12: Simulation: Monte Carlo
Applications of Simulation, Advantages of Simulation, Limitations of Method
Simulation. (Pages 219-232)
Unit 13: Project Scheduling and PERT - CPM: Basic Difference between Unit 13: Project Scheduling and
PERT and CPM, PERT / CPM Network Components and Precedence PERT-CPM
Relationship, Project Management - PERT. (Pages 233-266)
Unit 14: Game Theory: Competitive Situations, Characteristics of Unit 14: Game Theory:
Competitive Games, Maximin - Minimax Principle, Dominance. Competitive Situations
(Pages 267-288)
CONTENTS
INTRODUCTION
BLOCK I: OPERATIONS RESEARCH AND LINEAR PROGRAMMING
PROBLEM CONCEPT
UNIT 1 OPERATIONS RESEARCH 1-32
1.0 Introduction
1.1 Objectives
1.2 Historical Background
1.2.1 Nature and Scope of OR
1.2.2 Phases of Operations Research
1.2.3 Features of Operations Research
1.3 Scientific Methods in Operations Research
1.4 Problem Formulation
1.5 Types of Models in Operations Research
1.5.1 Structure of the Mathematical Model
1.6 Application Areas of Operations Research in Management
1.6.1 Limitations of OR
1.6.2 Techniques Tools of Operations Research
1.7 Answers to Check Your Progress Questions
1.8 Summary
1.9 Key Words
1.10 Self Assessment Questions and Exercises
1.11 Further Readings
UNIT 2 LINEAR PROGRAMMING 33-48
2.0 Introduction
2.1 Objectives
2.2 Formulation of Linear Programming
2.2.1 Meaning of Linear Programming
2.2.2 Fields Where Linear Programming Can Be Used
2.2.3 Components of a Linear Programming Problem
2.2.4 Mathematical Formulation of a Linear Programming Problem
2.2.5 Case Studies
2.2.6 LPP: Advantages/Applications and Limitations
2.4 Summary
2.5 Key Words
UNIT 3 GRAPHICAL ANALYSIS OF LPP 49-62
3.0 Introduction
3.1 Objectives
3.2 Procedure for Solving LPP by Graphical Method: Some Exceptional Cases
3.2.1 Important Geometric Properties of LPP
3.3 Canonical or Standard forms of LPP
3.5 Summary
3.6 Key Words
UNIT 4 SIMPLEX METHOD 63-92
4.0 Introduction
4.1 Objectives
4.2 Solution of LPP – Simplex Method
4.2.1 Solved Problems on Minimisation
4.3 The Simplex Algorithm
4.3.1 Penalty Cost Method or Big M-Method
4.3.2 The Two-Phase Simplex Method
4.5 Summary
4.6 Key Words
UNIT 5 DUALITY IN LINEAR PROGRAMMING PROBLEM 93-106
5.0 Introduction
5.1 Objectives
5.2 Concept of Duality and Sensitivity Analysis
5.2.1 Importance of Duality Concept
5.2.2 Formulation of a Dual Problem
5.2.3 Economic Interpretation of Duality
5.2.4 Sensitivity Analysis
5.4 Summary
5.5 Key Words
UNIT 6 TRANSPORTATION PROBLEM 107-125
6.0 Introduction
6.1 Objectives
6.2 Transportation Problem
6.2.1 Formulation of Transportation Problem (TP)
6.2.2 Initial Basic Feasible Solution
6.2.3 Moving Towards Optimality
6.2.4 Transportation Algorithm (MODI) Method
6.4 Summary
6.5 Key Words
UNIT 7 ASSIGNMENT PROBLEM 126-150
7.0 Introduction
7.1 Objectives
7.2 Assignment Problem
7.2.1 Mathematical Formulation of an Assignment Problem
7.2.2 Hungarian Method Algorithm
7.2.3 Routing Problem: The Travelling Salesman Problem
7.4 Summary
7.5 Key Words
BLOCK II: INTEGER PROGRAMMING AND QUEUING CONCEPTS

UNIT 8 INTEGER PROGRAMMING PROBLEM 151-181
8.0 Introduction
8.1 Objectives
8.2 Importance of Integer Programming Problems
8.2.1 Types of Integer Programming Problem
8.2.2 Gomory’s All-IPP Method
8.2.3 Gomory’s Fractional Cut Algorithm or Cutting Plane Method for Pure (All) IPP
8.2.4 Mixed Integer Programming Problem
8.2.5 Branch and Bound Technique
8.4 Summary
8.5 Key Words
UNIT 9 INFINITE QUEUING MODELS 182-189
9.0 Introduction
9.1 Objectives
9.2 Queuing Theory and Models
9.2.1 Classification of Queueing Models into Finite and Infinite Models
9.3 Operating Characteristics and Constituents of a Queueing system
9.3.1 The Input (Arrival Pattern)
9.3.2 The Service Facility
9.3.3 The Queue Discipline
9.3.4 Customer’s Behaviour
9.3.5 How to Manage Queues?
9.5 Summary
9.6 Key Words
UNIT 10 MATHEMATICAL ANALYSIS OF QUEUING THEORY 190-210
10.0 Introduction
10.1 Objectives
10.2 Model I: (M/M/I) (/FCFS) (Birth and Death Model)
10.3 Model III: (Multiservice Model) (M/M/S): (/FCFS)
10.4 Properties of Queuing System
10.5 Single and Multiple Channel Queuing Model
10.6 Erlang Family of Distribution of Service Times
10.6.1 Applications of Queuing Theory
10.6.2 Limitations of Queuing Theory
10.8 Summary
10.9 Key Words
UNIT 11 FINITE QUEUING MODELS 211-218
11.0 Introduction
11.1 Objectives
11.2 Model II (M/M/1): (N/FCFS)
11.3 Model IV (M/M/S): (M/FCFS)
11.4 Methodology to Simulation
11.4.1 Types of Simulation
11.6 Summary
11.7 Key Words
BLOCK III: SIMULATION GAME THEORY CONCEPTS

UNIT 12 SIMULATION: MONTE CARLO METHOD 219-232
12.0 Introduction
12.1 Objectives
12.2 Importance and Uses of Simulation Technique
12.2.1 Areas of Application of Monte Carlo Simulation
12.3 Monte Carlo Simulation
12.4 Application of Simulation
12.4.1 Advantages and Disadvantages of Simulation
12.6 Summary
12.7 Key Words
UNIT 13 PROJECT SCHEDULING AND PERT-CPM 233-266
13.0 Introduction
13.1 Objectives
13.2 PERT/CPM Network Components and Precedence Relationship
13.2.1 Basic Difference between PERT and CPM
13.2.2 Basic Terms
13.2.3 Common Errors
13.2.4 Rules of Network Construction
13.2.5 Numbering the Events (Fulkerson’s Rule)
13.2.6 Construction of Network
13.2.7 Time Analysis
13.2.8 Forward Pass Computations (For Earliest Event Time)
13.2.9 Backward Pass Computations (For Latest Allowable Time)
13.2.10 Determination of Floats and Slack Times
13.3 Critical Path Method (CPM)
13.4 Project Management-PERT
13.6 Summary
13.7 Key Words
UNIT 14 GAME THEORY: COMPETITIVE SITUATIONS 267-288
14.0 Introduction
14.1 Objectives
14.2 Basic Concepts of Game Theory
14.2.1 Types of Games and Their Characteristics
14.2.2 The Maximin-Minimax Principle
14.2.3 Games without Saddle Points (Mixed Strategies)
14.3 Dominance Property
14.5 Summary
14.6 Key Words
Introduction
INTRODUCTION
Operational Research, or simply OR, originated in the context of military operations,

but today it is widely accepted as a powerful tool for planning and decision-making, NOTES
especially in business and industry. The OR approach has provided a new tool for
managing conventional management problems. In fact, operational research
techniques do constitute a scientific methodology of analysing the problems of the
business world. They provide an improved basis for taking management decisions.
The practice of OR helps in tackling intricate and complex problems, such as that
of resource allocation, product mix, inventory management, sequencing and
scheduling, replacement and a host of similar problems of modern business and
industry. With IT facilities becoming widely available, the significance and scope
of OR has grown, and is still growing. Hence, OR is now an integral part of
courses of computer science, economics, business management, public
administration and several other disciplines.
Operations research (OR) is an interdisciplinary branch of applied
mathematics and formal science that uses mathematical methods, such as
mathematical modelling, statistics and algorithms to arrive at optimal or near optimal
solutions to complex problems. Basically, it is concerned with optimizing the maxima
(profit, assembly line performance, bandwidth, etc.) or minima (loss, risk, etc.) of
some objective function. It also helps management achieve its goals using scientific
methods. The field of operations research is closely related to Industrial Engineering
and Industrial Engineers consider operations research techniques as their major
toolset. Some of the primary tools used by operations researchers are statistics,
optimization, probability theory, queuing theory, game theory, graph theory, decision
analysis and simulation. Because of the computational nature of these fields, OR is
linked to computer science and OR professionals use specific custom-written
software for computation of data and decision-making.
The uniqueness of OR prompted industries to use its formal tools, such as
operations analysis, system analysis, management science, decision science, etc.
Commercial industries such as airlines, automobiles, communications, electronics,
transportation, chemicals and mining use OR techniques to optimally utilize their
limited resources and thereby maximize profits. Hence, OR is the application of
the methods of science to complex problems arising in the direction and management
of large systems of men, machines, materials and money in industry, business,
government and defence. The distinctive approach is to develop a scientific model
of the system, incorporating measurement of factors, such as chance and risk,
with which to predict and compare the outcomes of alternative decision strategies
and controls.
Self-Instructional
Material 11
Introduction This book, Operations Research, has been carefully and painstakingly
planned to prepare the students of distance learning programmes to become
successful managers and practitioners. The material has been presented in the
self-instructional mode or the SIM format. In this approach, we have begun with
NOTES an Introduction of the topic of the unit; then, we have outlined the Objectives;
which is followed by the details of the contents in a simple and easy-to-learn
format. At the end of each unit, we have highlighted a Summary and ‘Key Words’
for quick recollection. Finally, we have carefully posed Self Assessment Questions
and Exercises along with ‘Check Your Progress’ questions to increase your
comprehension of the subject.
Self-Instructional
12 Material
Operations Research
BLOCK - I
OPERATIONS RESEARCH AND LINEAR
PROGRAMMING PROBLEM CONCEPT
NOTES
UNIT 1 OPERATIONS RESEARCH

Structure
1.0 Introduction
1.1 Objectives
1.2 Historical Background
1.3 Scientific Methods in Operations Research
1.4 Problem Formulation
1.5 Types of Models in Operations Research
1.6 Application Areas of Operations Research in Management
1.8 Summary
1.9 Key Words
1.0 INTRODUCTION
Modern technological progress is accompanied by growth in scientific techniques.

While existing methods have been improved to meet the challenges arising from
the development of commerce and industry, a large number of new techniques or
so-called sophisticated tools of analysis have been and are even now being devised
to enlarge the application of scientific knowledge to an unlimited extent. Such
techniques have brought about a virtual revolution and can be reckoned as the
controlling forces of different walks of life. Operations Research, popularly known
as OR, is a recent addition to a long list of scientific tools that provide a new
outlook to many conventional management problems. OR adds greater
sophistication to solving management problems. It seeks to determine the best
(optimum) course of action of a decision problem in the face of limited resources.
Therefore, OR has become a versatile tool in the field of management and its
potential for future use is indeed substantial.
With the advent of automation, the centralization of organizational
management has become disintegrated. In a single unit of industry, different
Self-Instructional
Material 1
Operations Research departments of production, sales and inventory are governed by specialized
agencies. Their targets and goals often diverge, even as they serve the common
goals of the organization. The policy decisions required to coordinate these
conflicting directives may be taken quite effectively if an optimal solution can be
NOTES determined from all the available alternative compromise formulae. OR provides
an effective scientific technique to solve such decision-making problems of modern
business and industry.
1.1 OBJECTIVES
After going through this unit, you will be able to:
 Discuss the nature and scope of operations research
 Analyse the significance of operations research
 Explain the scientific methods used in operations research
 Interpret problem formulation
 Describe the models in operations research
 Mention the techniques of operations research
 Assess the application areas of operations research
1.2 HISTORICAL BACKGROUND
The term, ‘Operations Research’ was first coined by well-known authors J. F.

McCloskey and F.N. Trefethen in 1940. This new science came into existence in
a military context. During World War II, military management called on scientists
from various disciplines and organized them into teams to assist in solving strategic
and tactical problems, relating to air and land defence of the country. Their mission
was to formulate specific proposals and plans for aiding the military commands to
arrive at decisions on optimal utilization of scarce military resources and efforts
and also to implement the decisions effectively. This new approach to the systematic
and scientific study of the operations of the system was called Operations Research
(OR). Hence, OR can be referred to as ‘an art of winning the war without actually
fighting it’.
Operations Research (OR) has been defined so far in various ways and it is perhaps
still too early to be defined in some authoritative way. It is not possible to give
uniformly-acceptable definitions of OR. The following definitions are proposed
by various specialists in the field of OR. These have been changed according to
the development of the subject.
OR is a scientific method of providing executive departments with a
quantitative basis for decisions regarding the operations under their
Self-Instructional control. P. M. Morse, GE and Kimball
2 Material
OR is the scientific method of providing the executive with an analytical Operations Research
and objective basis for decisions.
P. M. S. Blackett
OR is the art of giving bad answers to problems to which otherwise
worse answers are given. NOTES
T. L. Saaty
OR is a systematic method oriented study of the basic structures,
characteristics, functions and relationships of an organization to provide
the executive with a sound, scientific and quantitative basis for decision-
making.
E. L. Arnoff and M. J. Netzorg
OR is a scientific approach to problem solving for executive
management.
H. M. Wagner
OR is an aid for the executive in making his decisions by providing
him with the quantitative information based on the scientific method
of analysis.
C. Kittee
OR is the scientific knowledge through interdisciplinary team effort
for the purpose of determining the best utilization of limited resources.
H. A. Taha
The various definitions given here, bring out the following essential
characteristics of operations research:
 Systems orientation
 Use of interdisciplinary terms
 Application of scientific methods
 Uncovering new problems
 Quantitative solutions
 Human factors
Scope of Operations Research
There is great scope for economists, statisticians, administrators and technicians
working as a team to solve problems of defence by using the OR approach. Besides
this, OR is useful in various other important fields, such as the following:
 Agriculture
 Finance
 Industry
 Marketing
 Personnel management
Self-Instructional
Material 3
Operations Research  Production management
 Research and development
NOTES
The procedure to be followed in the study of OR generally involves the following
major phases:
 Formulating the problem
 Constructing a mathematical model
 Deriving a solution from the model
 Testing the model and its solution (updating the model)
 Controlling the solution
 Implementation
OR has gained increasing importance since World War II in the technology of
business and industry administration. It greatly helps in tackling the intricate and
complex problems of modern business and industry. OR techniques are, in fact,
examples of the use of scientific method of management. The features of OR can
be well understood under the following heads:
 OR provides a tool for scientific analysis: OR provides the executives
with a more precise description of the cause-and-effect relationship and
risks underlying the business operations in measurable terms. This eliminates
the conventional intuitive and subjective basis on which managements used
to formulate their decisions decades ago. In fact, OR replaces the intuitive
and subjective approach of decision-making by an analytical and objective
approach. The use of OR has transformed the conventional techniques of
operational and investment problems in business and industry. As such, OR
encourages and enforces disciplined thinking about organizational problems.
 OR provides solution for various business problems: The OR
techniques are being used in the field of production, procurement, marketing,
finance and other allied fields. Problems like the following, and similar other
problems, can be solved with the help of OR techniques: How best can the
managers and executives allocate the available resources to various products
so that in a given time, the profits are maximum or the cost is minimum?
Is it possible for an industrial enterprise to arrange the time and quantity of
orders of its stocks such that the overall profit with given resources is
maximum?
How far is it within the competence of a business manager to determine the
number of men and machines to be employed and used in such a manner
Self-Instructional
4 Material
that neither remains idle and at the same time, the customer or the public Operations Research
has not to wait unduly long for service?. Similarly, we might have a complex
of industries—steel, machine tools and others—all employed in the
production of one item, say steel. At any particular time, we have a number
NOTES
of choices of allocating resources, such as money, steel and tools for
producing autos, building steel factories or tool factories. What should be
the policy which optimizes the total number of autos produced over a given
period? OR techniques are capable of providing an answer in such a situation.
Planning decisions in business and industry are largely governed by the picture
of anticipated demands. The potential long-range profits of the business
may vary in accordance with different possible demand patterns. The OR
techniques serve to develop a scientific basis for coping with the uncertainties
of future demands. Thus, in dealing with the problem of uncertainty over
future sales and demands, OR can be used to generate ‘a least risk’ plan.
At times, there may be a problem in finding an acceptable definition of long-
range company objectives. Management may be confronted with different
viewpoints—some may stress the desirability of maximizing net profit whereas
others may focus attention primarily on the minimization of costs. OR
techniques (especially that of mathematical programming, such as linear
programming) can help resolve such dilemmas by permitting systematic
evaluation of the best strategies for attaining different objectives. These
techniques can also be used for estimating the worth of technical innovations
as also of potential profits associated with the possible changes in rules and
policies.
How much changes can be there in the data on which a planning formulation
is based without undermining the soundness of the plan itself? How accurately
must managements know the cost coefficients, production performance
figures and other factors before it can make planning decisions with
confidence? Many of the basic data required for the development of long-
range plans are uncertain. Such uncertainties though cannot be avoided, but
through various OR techniques, the management can know how critical
such uncertainties are and this in itself is a great help to business planners.
 OR enables proper deployment of resources: OR renders valuable help
in proper deployment of resources. For instance, Programme Evaluation and
Review Technique (PERT) enables us to determine the earliest and the latest
times for each of the events and activities and, thereby, helps in the identification
of the critical path. All these help in the deployment of resources from one
activity to another to enable the project completion on time. Thus, this technique
provides for determining the probability of completing an event or project
itself by a specified date.
Self-Instructional
Material 5
Operations Research  OR helps in minimizing waiting and servicing costs: The waiting line
or queuing theory helps the management in minimizing the total waiting and
servicing costs. This technique also analyses the feasibility of adding facilities
and, thereby, helps the business people in taking a correct and profitable
NOTES decision.
 OR enables the management to decide when to buy and how much to
buy: The main objective of inventory planning is to achieve balance between
the cost of holding stocks and the benefits from stock holding. Hence, the
technique of inventory planning enables the management to decide when to
buy and how much to buy.
 OR assists in choosing an optimum strategy: Game theory is especially
used to determine the optimum strategy in a competitive situation and enables
the businessmen to maximize profits or minimize losses by adopting the optimum
strategy.
 OR renders great help in optimum resource allocation: The linear
programming technique is used to allocate scarce resources in an optimum
manner in problems of scheduling, product mix, and so on. This technique is
popularly used by modern managements in resource allocation and in affecting
optimal assignments.
 OR facilitates the process of decision-making: Decision theory enables
the businessmen to select the best course of action when information is
given in the probabilistic form. Through decision tree (a network showing
the logical relationship between the different parts of a complex decision
and the alternative courses of action in any phase of a decision situation)
technique, executive’s judgement can systematically be brought into the
analysis of the problems. Simulation is another important technique used to
imitate an operation or process prior to actual performance. The significance
of simulation lies in the fact that it enables in finding out the effect of alternative
courses of action in a situation involving uncertainty where mathematical
formulation is not possible. Even complex groups of variables can be handled
through this technique.
 Through OR management, you can know the reactions of the
integrated business systems: The Integrated Production Models
technique is used to minimize cost with respect to workforce, production
and inventory. This technique is quite complex and is usually used by
companies having detailed information concerning their sales and costs
statistics over a long period. Besides, various other OR techniques also
help management people in taking decisions concerning various problems
of business and industry. The techniques are designed to investigate how
the integrated business system would react to variations in its component
elements and/or external factors.
Self-Instructional
6 Material
Operations Research
Check Your Progress

1. State the essential characteristics of operations research.
2. How is operations research used as a tool for scientific analysis? NOTES
1.3 SCIENTIFIC METHODS IN OPERATIONS

RESEARCH
The methodology of operations research generally involves the following steps:

 Formulating the problem: The first step in an OR study is to formulate
the problem in an appropriate form. Formulating a problem consists of
identifying, defining and specifying the measures of the components of a
decision model. This means that all quantifiable factors which are pertinent
to the functioning of the system under consideration are defined in
mathematical language: variables (factors which are not controllable) and
parameters or coefficient, along with the constraints on the variables and
the determination of suitable measures of effectiveness.
 Constructing the model: The second step comprises constructing the
model, by which we mean that appropriate mathematical expressions are
formulated that describe interrelations of all variables and parameters. In
addition, one or more equations or inequalities are required to express the
fact that some or all of the controlled variables can only be manipulated
within limits. Such equations or inequalities are termed as constraints or the
restrictions. The model must also include an objective function, which defines
the measure of effectiveness of the system. The objective function and the
constraints, together constitute a model of the problem that we want to
solve. This model describes the technology and the economics of the system
under consideration through a set of simultaneous equations and inequalities.
 Deriving the solution: Once the model is constructed, the next step in
an OR study is that of obtaining the solution to the model, that is, finding
the optimal values of the controlled variables—values that produce the
best performance of the system for specified values of the uncontrolled
variables. In other words, an optimum solution is determined on the basis
of the various equations of the model satisfying the given constraints and
interrelations of the system, and at the same time maximizing profit or
minimizing cost or coming as close as possible to some other goal or
criterion. How the solution can be derived depends on the nature of the
model. In general, there are three methods available for the purpose,
namely the analytical methods, the numerical methods and the simulation
methods. Analytical methods involve expressions of the model by
mathematical computations and the kind of mathematics required depends
Self-Instructional
Material 7
Operations Research upon the nature of the model under consideration. This sort of mathematical
analysis can be conducted only in some cases without any knowledge of
the values of the variables, but in others the values of the variables must
be known concretely or numerically. In latter cases, we use the numerical
NOTES methods, which are concerned with iterative procedures through the use
of numerical computations at each step. The algorithm (or the set of
computational rules) is started with a trial or initial solution and continued
with a set of rules for improving it towards optimality. The initial solution
is then replaced by the improved one and the process is repeated until
no further improvement is possible. However, in those cases where the
analytical as well as the numerical methods cannot be used for deriving
the solution, then simulation methods are used, that is, experiments are
conducted on the model in which values of the uncontrolled variables are
selected with the relative frequencies dictated by their probability
distributions. The simulation methods involve the use of probability and
sampling concepts, and are generally used with the help of computers.
Whichever method is used, our objective is to find an optimal or near
optimal solution, that is, a solution which optimizes the measure of
effectiveness in a model.
 Testing the validity: The solution values of the model, obtained as
stated in the previous step, are then tested against actual observations. In
other words, effort is made to test the validity of the model used. A
model is supposed to be valid if it can give a reliable prediction of the
performance of the system represented through the model. If necessary,
the model may be modified in the light of actual observations and the
whole process is repeated till a satisfactory model is attained. The
operational researcher quite often realizes that his model must be a good
representation of the system and must correspond to reality, which in turn
requires this step of testing the validity of the model in an OR study. In
effect, performance of the model must be compared with the policy or
procedure that it is meant to replace.
 Controlling the solution: This step of an OR study establishes control
over the solution by proper feedback of information on variables, which
might have deviated significantly. As such, the significant changes in the
system and its environment must be detected and the solution must
accordingly be adjusted. This is particularly true when solutions are rules
for repetitive decisions or decisions that extend over time.
 Implementing the results: Implementing the results constitutes the last
step of an OR study. Since the objective of OR is not merely to produce
reports but to improve the performance of systems, the results of the research
must be implemented, if they are accepted, by the decision-makers. It is
through this step that the ultimate test and evaluation of the research is
Self-Instructional
8 Material
made and it is in this phase of the study where the researcher has the greatest Operations Research
opportunity for learning.

Thus, the procedure for an OR study generally involves some major steps,
namely formulating the problem, constructing the mathematical model to represent NOTES
the system under study, deriving a solution from the model, testing the model and
the solution so derived, establishing controls over the solution, and lastly, putting
the solution to work-implementation. (Although the said phases and the steps are
usually initiated in the order listed in an OR study, it should always be kept in mind
that they are likely to overlap in time and interact, that is, each phase usually
continues until the study is completed.)
The OR approach can as well be illustrated by the following flow chart shown
is Figure 1.1.
Facts, opinions Define the

and symptoms problem
of the problem
Technical, Determine
financial factors affecting
and economic the problem
information (variables,
from all sources constraints and Model Tools
assumptions) development of trade
Develop objective Maximize OR

and alternative or Tech-
solutions Minimize niques
function
Detailed Analyse
information alternatives
on factors and deriving
optimum solution
Computer
service
Determination (may be used)
of validity
recommend action
control and
implementation
Fig. 1.1 Flow Chart Showing OR Approach
Check Your Progress

3. State the first step in the scientific method of operations research.
4. How are the solution values of a decision model (obtained) validated?
Self-Instructional
Material 9
Operations Research
1.4 PROBLEM FORMULATION
To formulate a problem correctly, one must know what a problem is and must
understand the nature of the problem. A problem is said to exist when we have the
NOTES
following conditions satisfied:
 There must be an individual or a group, which has some difficulty or problem.
 There must be some objectives to be attained. If one wants nothing, one
cannot have a problem.
 There must be alternative means (or the courses of action) for attaining the
objective(s) one wishes to attain. This, in other words, means that there
must be atleast two means available. If there is no choice of means, one
cannot have a problem.
 There must remain doubt with regard to selection of alternatives, that is, the
relative efficiency of the alternative means must differ.
 There must be some environment(s) to which the difficulty pertains.
The nature of the problem must be understood in context of the above
points before its formulation takes place. It is only after this that the problem must
be stated along with the objective(s) and the bounds within which it is to be studied.
In fact, this involves the task of laying down boundaries within which we shall
study the problem with a predetermined objective in view. In brief, the task of
formulating the problem consists in making various components of the problem
explicit.
Formulating the Objectives
Generally, the objectives of an organization may either be retentive or acquisitive.
Retentive objectives are concerned with preserving either resources of value, such
as money, time, energy, or states such as comfort, safety, stability, and the like. In
other words, retentive objectives are concerned with what is consumed by courses
of action and as such can easily be translated as minimizing objectives. On the
other hand, we may have acquisitive objectives which are concerned with acquiring
resources or attaining states and can easily be translated as maximizing objectives.
Essentially then, we may have either minimizing objectives or maximizing objectives.
Accordingly, profits, sales, revenues, output and the like are always to be maximized,
and cost, time and the like are to be minimized.
Formulation of the problem and the objective often requires a detailed
knowledge of how the concerning system operates, which can be learnt through
performing a system analysis. This is a process analogous to the physical
examination that a doctor performs on his patient after initial discussion of symptoms.
Such an analysis can provide the background information required in formulating
the problem and the model required to solve it.
Self-Instructional
10 Material
Measure of Performance Operations Research
After understanding the nature of the problem and after identifying the courses of
action, the uncontrolled variables and the objectives, it is necessary to construct a
measure of performance that can be used to determine which alternative is best NOTES
and what function of this measure (usually described as the objective function)
should be used as the criterion of best solution of the concerning problem. Selecting
a criterion of the best solution often requires a good knowledge of decision theory.
The type of appropriate criterion depends on the state of knowledge of outcomes
that we assume to be available. Usually, we may have one of the three situations
with regard to the state of outcomes, namely, deterministic situation or state of
certainty, stochastic situation or risk situation, and finally the situation of uncertainty.
They are discussed as follows:
 Deterministic situations: These are situations in which each course of
action results in only one outcome. In such situations, the probability of
each outcome is known or believed to be either equal to unity or zero. Due
to this, the criterion or the measure of performance used in deterministic
situations is that of the maximization of utility.
 Stochastic situations: These are situations in which each course of action
can result in alternative outcomes; the probabilities of which are known or
can be estimated. The criterion that is generally accepted as the most
appropriate in such situations is that of maximizing the expected utility and is
often termed as either the Expected Monetary Value (EMV) criterion or
the Expected Opportunity Loss (EOL) criterion. To select the best
alternative, we select that alternative of which EMV is highest possible (or
EOL is the lowest possible).
 Uncertainty situations: These are situations in which each course of action
can result in alternative outcomes, the probabilities of which cannot be
assigned, as the decision-maker does not know which outcomes can or will
occur. This means that the decision-maker has to act with imperfect
information in such a situation. Consequently, there is no single best criterion
for selecting a strategy to deal with such a situation. Accordingly, various
criteria have been developed for selecting a course of action in uncertainty
situations. The following criteria deserves mention in this context:
(i) Maximin decision rule
(ii) Maximax decision rule
(iii) Minimax decision rule (or regret rule)
(iv) Hurwicz decision rule
(v) Laplace decision rule
Nothing can be said as to which one of these criteria is the best criterion in any
given situation involving uncertainty. The choice for the selection must be left to the
decision-maker who must resort to his own judgement, skill and experience.
Self-Instructional
Material 11
Operations Research Thus, problem formulation, which is considered as the first phase of an OR
study, requires a statement of the problem’s components that include the
controllable or the decision variables, the uncontrollable parameters or the
environment, the constraints on the variables and also the objective along with an
NOTES appropriate measure of performance.
Check Your Progress

5. What are stochastic situations?
6. What are retentive objectives?
1.5 TYPES OF MODELS IN OPERATIONS

RESEARCH
After problem formulation, modelling plays an important part in OR. In fact, it is

considered as the essence of operations research approach. Modelling permits
the operational researcher to test rigorously the implications of new plans, new
schemes and new ideas. The term ‘modelling’ generally refers to the idea of
constructing a model of the given problem and then deriving the solution from it for
the problem under consideration. Modelling is a device used to arrive at a well-
structured view of reality. By analysing or experimenting on models, we can usually
determine how changes in the relevant system will affect its performance.
What is a model?
A model is a physical or symbolic representation of the relevant aspects of the
reality or system with which one is concerned. In other words, a model is a means
of portraying the system or reality of concern to the decision-maker. As such, the
concept of a model generally implies a series of connected and identifiable
relationships that essentially demonstrate the proposition: if this action, then this
result. Thus, ‘the model, being an abstraction of the assumed real system, identifies
the pertinent relationships of the system in the form of an objective and a set of
constraints.’
Types of models
Models can be of several types, but models commonly referred are as follows:
 Iconic models: Iconic model is a pictorial or visual representation of certain
aspects of a system. In iconic models, the relevant properties of the real
thing are represented by the properties themselves, usually with a change of
scale. Such models generally look like what they represent, but differ in
size. An iconic model is said to be ‘scaled down’ or ‘scaled up’ as the
dimensions of the model are smaller or bigger than those of the real thing.
For instance, iconic models of the sun housed in planetoriums, the model of
a huge building or of an airplane or of an automobile for exhibition purposes
Self-Instructional
12 Material
are scaled down; whereas models of the atom or of a cell used for, say Operations Research
teaching purposes are scaled up. Similarly, maps, photographs, drawings

also constitute examples of iconic models. Iconic models, such as models
of aircraft, machine, engine or of any tangible thing may also be termed as
physical models. Iconic models are generally specific, concrete and easy to NOTES
observe, but are difficult to manipulate for experimental purposes and are
not very useful for the purpose of prediction. Moreover, such models
represent a static event only.
 Analogue models: Analogue models use one set of properties to represent
another set of properties. They are more abstract than iconic models.
However, such models are easier to manipulate and can represent dynamic
situations. Graphs representing time series, stock market changes or other
phenomenons, flow charts, demand curves and frequency graphs are
examples of analogue models for they use geometrical magnitudes and
location to represent a wide variety of variables and the relationships between
them. Similarly, a hydraulic system can be used as an analogue model of
electrical, traffic and economic systems. Maps in different colours, each
colour in a map representing a particular characteristic, such as blue colour
representing water, brown representing land, and so on, are also termed as
analogue models. Such models are generally more useful than iconic models
because of their relatively greater capacity to represent the characteristics
of the real system.
 Symbolic or mathematical models: In symbolic models, the components
of what is represented and their interrelationships are given by symbols.
These models use letters, numbers and other types of symbols to represent
variables and the relationships between them. Such models generally assume
the form of equations or inequalities depicting the relationships amongst the
variables of the system which they represent. For instance, a mathematical
equation representing a relationship between constants and variables may
be considered a symbolic model. Thus, a symbolic model means a
mathematical description of an activity, which expresses the relationships
among the various elements with sufficient accuracy that it can be used to
predict the actual outcome under any expected set of circumstances.
Symbolic models are most general and precise, and also are most abstract
in nature. They are usually the easiest to manipulate experimentally.
A great advantage of such models is that they lend themselves easily for
manipulation on computers. In most OR applications, it is assumed that the
objective and constraints can be expressed mathematically as functions of
decision variables. In such a case, we say that we are dealing with a symbolic
or a mathematical model. Important symbolic or mathematical models often
used in operations research are as follows:
(i) Allocation models: They are used in finding a solution of optimizing a
given objective such as profit maximization under certain restrictions.
Self-Instructional
Material 13
Operations Research (ii) Scheduling models: They are used in determining an optimum sequence
of performing certain operations of jobs with a view to minimize overall
time and/or cost. PERT and other network techniques fall under this
category.
NOTES
(iii) Queuing models: They represent the random arrival of customers at
a service station where the facility is limited and the object of such
models happens to be to minimize the costs of both servicing and
waiting.
(iv) Inventory models: These models relate mainly to optimizing inventory
levels. The main emphasis in such models is on minimizing the total
costs associated with inventories.
(v) Replacement models: These models are concerned with replacement
strategies.
(vi) Competitive models: These models are concerned with the
determination of optimum strategies under competitive conditions.
Game theory models are examples of competitive theory.
These wide variety of models used in OR give rise to a corresponding
number of solution techniques, such as linear, integer, dynamic and non-linear
programming that represent algorithms for solving the models.
 Simulation models: At times, however, it may not be possible to depict a
real system in mathematical formulation, that is, it may just happen that a
mathematical model of a given problem cannot be constructed, and as such,
we may adopt simulation models.
Simulation models differ from mathematical models in that the relationships
between input and output are not explicitly stated. A simultation model breaks
down the modelled system into basic modules that are then linked to one another
by a well-defined logical relationship.
 Functional, operational and accounting models: A functional model is
a model without values specified for the constants and as such the variables
can only be identified in abstract mathematical notation. A functional model
specifies the form of the relationship, but not the exact value of the relationship.
However, when the values which must be associated with some measurable
quantity in the environment being represented have been identified, we have
an operating model. For instance, Yi= ai+ biXi is an example of a functional
model, but when we identify the values of the two constants, such as a1 =
4 and b1 = 3 or a2 = 5 and b2= 4, then we can write the said model either
as Y1 = 4 + 3X1 or Y2 = 5 + 4X2, and in this form, it is known as an
operating model. Certain equations are essentially tautologies because they
merely represent addition of values to give a total value; thus, in a sense,
Self-Instructional
14 Material
they are related to accounts. Suppose, we are given Y1 and Y2. The total Operations Research
production, Yi is equal to Y1 + Y2. This is nothing but an accounting statement.

Does it constitute a model? In the business world, we assume it to be a
model (and consider it an accounting model) because it does appear to
represent symbolically relevant aspects of the reality with which we are NOTES
often concerned.
 Prediction, improvement and optimization models: Mostly this
classification is stated in the context of managerial analysis. Optimization
models are the result of quantitative analysis and allow the measurable
criterion to assume the highest or the lowest possible value, given the model
structure and constraints. Improvement models are the result of system
analysis and consequently the highest value of the criterion is chosen from
among the alternatives investigated. Either of these models depends upon
inputs which predict outcomes of relationship; thus, they depend upon a
prediction model.
 Static, dynamic, linear and non-linear models: According to Joseph
Wright Forrester (a pioneer in operations research), all models possess
differing degrees of various characteristics, namely staticness, linearity,
stability and openness. In view of this, models can as well be classified as
static model or dynamic model, linear model or non-linear model. A static
model is one in which time plays no role, that is, the relationships do not
vary with time. As against this, a dynamic model contains time-varying
interactions. A linear model is one in which all of the equations or inequalities
are linear, but a model which contains one or more non-linear equations is
known as a non-linear model. In case of non-linear models, the degree of
non-linearity may vary greatly from model to model.
 Stable and unstable models: In the words of Forrester, ‘A stable system
is one that tends to return to its initial condition after being disturbed. It may
overshoot and oscillate like a single pendulum that is set in motion, but the
disturbances decline and die out. In an unstable system that starts at rest, an
initial disturbance is amplified, leading to growth or oscillations whose
amplitude increases.’Another sub-element of stability is concerned with the
presence of steady state or transient models. Using the words of Forrester
again, we can say that a steady state pattern is one that is repetitive with
time in which the behaviour in one period is of the same nature as any other
period. On the other hand, the transient behaviour refers to those changes
where the character of the system changes with time.
 Open and close models: On the basis of the characteristic of openness, a
model can be classified either as an open model or as a closed model. A
closed model is one that internally generates the values of variables through
time by the interaction of variables one on another, whereas the open model
is one in which some variables are supplied from the external environment.
Self-Instructional
Material 15
Operations Research Properties of a good model
We can list down the properties of a good model as follows:
 A good model must represent the reality or the system with which it is
NOTES concerned as accurately as possible and must not compromise with any of
the critical elements of the system. However, this does not mean that a
model should be as close as possible to reality in every respect, for then, it
would require an infinite length of time to construct such a model, which at
times may even be beyond human comprehension. Thus, a good model
must be economical of time and also be readily workable. Usable
approximate models are much better than more exact models that cannot
be used.
 Since a model happens to be a simplified representation of an empirical
situation, it must possess the characteristic of predicting outcomes reasonably
well and must be consistent with effective action.
 A good model must be simple enough so as to be readily understood and,
if necessary, must be capable of being modified quickly and effectively.
 A good model must identify the factors that can influence the course of a
particular decision and must also determine how these factors interact to
produce departures from the expected or desired outcome.
 A model must make certain assumptions about the structure of the problem
under study for no system of equations can produce meaningful results if it
consists entirely of unknown, but the number of assumptions should be as
small as possible in a good model.
 A good model is one that is capable of answering in case when small
perturbations take place in the data. This sensitivity of the model, that is, its
capacity to answer in context of small changes constitutes an important
virtue of a good model.
Advantages of modelling
There are several advantages that result from modelling. Important ones are as
follows:
 The main advantage of modelling is that the model formulated under it can
be manipulated in a variety of ways until the best solution of the problem
under consideration is found. The actual system or the organization can
then be operated in an optimum fashion with a minimum of costly experiments
and adjustments. For instance, models of aircraft are always tested in wind
tunnels before production is started.
 Models can be used on a ‘what if’ basis to explore the possible future
consequences of decisions that might be made today. This enables the
creative manager to test the implications of new plans, new schemes and
Self-Instructional
16 Material
new ideas that are most critical to the success or failure of a particular Operations Research
project. Modelling, thus, helps in finding avenues for research and

improvements.
 Modelling provides an easy and systematic approach for studying the
NOTES
problem concerned. Through this technique, the problem under study
becomes controllable. Had the models been as complex and difficult to
control as reality, then there would have been no advantage in their use. In
fact, modelling strips a natural phenomenon of its bewildering complexity
and duplicates the essential behaviour of the natural phenomenon with a
few variables, simply related.
 Modelling provides an opportunity to explore a broad range of solutions
and to estimate the sensitivity of the solution to changes in individual elements
or in combination of these elements. This, in turn, will enable the decision-
maker to move away safely from the precise optimum point by a given
margin in order to accommodate other factors. Thus, modelling indicates
the limitations and scope of an activity.
 Model building and analysis does not replace intuition and judgement; it
rather supports them with tools for handling complexity and uncertainty
with which the human intuition cannot cope unaided. In a sense, the use of
a model frees the intuition and permits it to concentrate on those problems
to which it is particularly suited.
 Modelling enables quick and economical experimentation for finding an
optimum solution for a given problem. The results of such experiments can
be evaluated to determine which strategy is most consistent with a desired
objective. Besides, the validity of the model (that is, whether the model
results in a reliable prediction of the system’s performance) can be checked,
and in case of even little doubt, the model can accordingly be modified
quickly and effectively.
Deriving Solutions from Models
(a) Abstraction
Abstraction happens to be the first step in modelling and consists in selecting the
critical factors or variables from the empirical situation. There are usually an
uncountable number of ‘facts’ in any empirical situation and the decision-maker
must intelligently abstract those factors which he considers to be most relevant to
the problem he faces. Relevance has two charactreristics: (i) There may be variables
which affect the environment, but whose impact is negligible and as such they are
not considered relevant. (ii) There are some variables which do not affect the
environment and thus, need not be even considered tentatively as a part of the
model. Essentially, the model must deal with the relevant aspects of the reality with
which we are concerned. For instance, one can classify automobiles in several
ways: on the basis of car’s origin one can differentiate between domestic and
Self-Instructional
Material 17
Operations Research foreign cars; they can as well be classified on the basis of age, cost or manufacturer.
One can as well classify them on the basis of body style, colour, size of engine or
on the basis of the other factor. The point is that all these factors are valid in a
particular situation. However, if someone is interested in conducting the study of
NOTES the effect of the weather on automobiles, the paint colour would be very useful in
such a study. If one is, however, concerned with automobile accidents, then the
classification on the basis of colour would be useless and the classification on the
basis of engine type and size would be considered as appropriate and valuable.
(b) Model building
The next step in modelling is that of model building. The relevant factors or variables
selected as stated in (a) are put in some logical manner, so that they form a model
of the given problem. In model building, it is generally considered desirable to
simplify reality but only to the point where there is no significant loss of accuracy.
In fact, the decision-maker tries his level best to construct the simplest possible
model that predicts outcomes reasonably well and is amenable to quick modification
in case there is need for the same models to be represented in a variety of ways.
However, in the relatively recent past, emphasis is laid on the development of
symbolic models particularly with regard to business problems. As such, one often
talks about symbolic models in the context of model building. The question then is:
‘How to construct a mathematical model for a given problem?’ The answer depends
on that the variables and parameters which are relevant and are also capable of
quantitative measurement of some form which must be put in logical manner
indicating the necessary relationships along with constraints and the objective
function. Such an arrangement is often described as a model of the given problem.
It is not possible to prepare a manual of constructions for model building as much
depends upon the imagination, insight and experience of the decision-maker himself.
However, past experience of model building proves a good guide in developing
an appropriate model. Accordingly, we give the following types of illustrative
symbolic models:
 Case 1 (Total Expense model): Symbolic model can just be presented in
the equation form. A very simple model of total expense can be as follows:
T = a + bX
where, T = total costs; a = fixed costs; b = variable costs; X = number of
units.
This model can be used as a predictor of the total costs associated with a
given level of output, of course, with certain assumptions. This model implies
that total costs are dependent upon the number of units produced. Thus, T
is the dependent variable and X is the independent variable in the given
context. On the other hand, a and b are constants (or the parameters) of
the given model. The model, thus, represents the relationship between the
Self-Instructional
18 Material
total cost on one hand and number of units produced along with given Operations Research
constants on the other. If the object happens to have the total costs at its
minimum level, then we have to find that value of X corresponding to which
T happens to be minimum possible.
NOTES
 Case 2 (Milkman’s model): Consider the situation of a milkman who must
decide how many cases of milk to order to maximize his expected profits. He
buys a certain number of milk cases each day and sells some or all of them.
He makes a profit on each one he sells. However, the unsold case brings him
a loss. The demand for milk cases in the market varies from day to day, but
the probability that any specified number of milk cases will be sold on a
particular day can be determined on the basis of past data. We can develop
the model for this type of problem by identifying the relevant variables and
assigning symbols to them as follows:
n =The number of milk cases ordered per day
p =The profit made on each milk case sold
l =The loss incurred on each milk case unsold
d =The demand, that is, the number of milk cases that could be sold
per day if n  d
pr (d) = The probability that the demand will equal supply on a randomly
selected day
P = Net profit per day (negative P represents a loss)
If the demand on a particular day exceeds the number ordered, that is, If
d > n, his profit would be:
P(d > n) – n.p
On the other hand, if demand does not exceed the number ordered, his
profit would be as follows:
P(n  d) = d.p – (n – d).1
Then the exacted net profit per day ( P ) can be expressed as follows:
n 
P  P  d   dp –  n – d  l    P  d  n. p
d 1 d  n 1
This is a decision model under risk situation, wherein P is the measure of
performance, n is the controlled variable, d is an uncontrolled variable, and
p and l are uncontrolled constants. To solve the problem represented through
this model, we should find the value of n that maximizes P .
 Case 3 (Linear Programming model): A simple linear programming model
can be stated as follows:
n
Maximize: Z = c j X j ( j = 1,2,...... n) (Objective Function)
j 1
Self-Instructional
Material 19
Operations Research Subject to the constraints:
n
 aij Xj  bi (i = 1,2,...... m) (Constraints concerning resources)
j 1
NOTES and Xj  0 (Non-negativity condition)
The linear programming model pertains to maximization case. It assumes
prior knowledge of all the coefficients, that is, it operates only under
conditions of certainty. It has three components, namely the objective
function, constraints and the non-linearity condition. The various assumptions
are the linearity of the objective function and that of the constraints, additivity
of resources and divisibility of decision variables. Non-negativity condition
implies that no X can be negative. The model requires the determination of
n production activities pursued at level X subject to a limited amount of m
resources being available. Each unit of the jth activity yields a return of c,
and uses an amount aij of the ith resource. Z denotes the optimal value of
the objective function for a given system.
 Case 4 (Simple inventory model concerning economic lot size): After
the model has been built, certain conclusions may be derived about its
behaviour by means of logical analysis. Some mathematical calculations are
often required for obtaining the solution to the model. The solution to the
model means those values of the controllable variables that optimize the
given objective. If the logic adopted in deriving the conclusions from the
model is correct, then the solution to the model will also serve as the effective
solution for the empirical problem represented by the said model.
(c) The question of error
It should always be remembered that the solution derived from the model always
contains some degree of error because of the abstraction process under which the
decision-maker may select the wrong variables, or not enough variables for
constructing his model. If the error involved is not big enough to disturb the validity
of the concerned model, then we generally do not mind the error and accordingly
accept the solution provided by the model. However, when the error becomes
significant, then we have to rethink about the model. The question of when the
error becomes so large that the conclusions must be modified before it can be
adopted as a solution, is one of judgement of the decision-maker depending upon
his experience and knowledge of the given situation.
(d) Updating the model
A good model is one which gives valid results and permits reliable prediction of the
system’s performance. As such the model once built up to represent a system must
be continuously updated to take account of the past, present and future specifics of
the problem. This, in other words, means that the model should be modified as soon
as one or more of the controlled variables change significantly increasing the size of
the error involved.
Self-Instructional
20 Material
A word of caution in modelling Operations Research
The usefulness of modelling has already been stated, but one must note that the
same follows only when the model happens to be a reasonable approximation of
reality. If the model is a good representation of the problem, the solution of the NOTES
model will be a good solution of the problem. However, the solution of a poor
model, even though it may be an exact one, will not be a good solution of the
problem. Hence, the construction of the model and the translation of the results to
the actual problem must be done with utmost care. The key to successful modelling
is to represent the decision situation as accurately as possible and in particular, not
to compromise the critical elements of the problem. For this purpose, the problem
must be well defined before constructing the model along with a clear and precise
specification of all relevant factors and the interrelationships among them.
It should also be noted that though many problems can be easily represented
in the form of models, but there may be problems in respect of which we may not
construct reasonably good models. Accordingly, we must construct an appropriate
model of a problem if we can, otherwise we must use an intuitive or heuristic
approach to solve the given problem. ‘Finally, if a problem is so complex that it
cannot be modelled adequately for solution by one of the available methods or it
cannot be solved adequately with a heuristic method, then the problem solver
usually resorts to simulation. Of course, simulation is not the answer to all problems;
nevertheless, it does have a great deal of merit in studying large, complex systems
where the components are highly interrelated.’
Check Your Progress

7. Mention the properties of a good model.
8. What is quadratic programming?
1.6 APPLICATION AREAS OF OPERATIONS

RESEARCH IN MANAGEMENT
Operation research techniques, having their origin with war operations analysis during
the World War II, have since then assisted defence management to a larger extent.
Defence management being vitally concerned with problems of logistics, such as
provisioning, distributing, search and forecasting, OR has given valuable help in decision
formation. Although OR originated in context of military operations, its impact
nowadays can be observed in many other areas. It has successfully entered into
different research areas relating to modern business and industry. In fact, it is being
reckoned as an effective scientific technique to solve several managerial decision-
making problems, such as inventory management problems, resource allocation
problems, problems relating to replacement of both men as well as machines,
sequencing and scheduling problems, queuing problems, competitive problems, search
problems, and the others.
Self-Instructional
Material 21
Operations Research Big and Small Organizations
The OR approach, though being practised by big organized business and industrial
units, is equally applicable to small organizations. For instance, whenever a
NOTES departmental store faces the problem, such as employing additional sales girls and
purchasing an additional van, one can apply some techniques of the operations
research to minimize cost and maximize benefit for such decision. As such, it is
often said that wherever there is a problem, there is scope for applying operations
research techniques.
Military, business and research
In recent years of organized development, operations research has successfully
entered, apart from its military and business applications, into several other areas of
research. For instance, the basic problem in most of the developing countries in Asia
and Africa is to remove poverty as quickly as possible. There is a considerable
scope to solve this problem by applying OR techniques. Similarly, with the explosion
of population and consequent shortage of food, every country is facing the problem
of optimum allocation of land for various crops in accordance with the climatic
conditions and available facilities. The problem of optimal distribution of water from
a source, like a canal for irrigation purposes, is faced by each developing country.
Hence, good amount of scientific work related to the OR can be done in this direction.
Thus, there is considerable scope of applying OR techniques for solving problems
relating to national planning. National planning can be largely improved if optimum
coordination can be arrived at through central planning agency by resorting to
operations research models and techniques. In India, there is a promising field for
application of operations research techniques in this direction.
Finance
In the field of finance, the progress of operations research has been rather slow.
Financial institutions have many situations similar to those prevailing in other
commercial organizations. Investment policy must maximize the return on it, keeping
the factor of risk below a specified level. Long-range corporate objectives of an
institution are always studied along with functional objectives of individual
departments, consistency between the two being essential for corporate planning.
In particular, banking institutions require precise and accurate forecasting on cash
management and capital budgeting. Linear programming models for many problems
and quadratic programming formulation for some complex problems may prove
the usefulness of operations research techniques to credit institutions.
Hospital management, health planning and transportation system
The application of operations research techniques can as well be noticed in the
context of hospital management, health planning programmes, transportation system
and in several other sectors. For instance, hospital management quite often faces
the problem of allotting its limited resources, of its multiphased activities. Optimum
Self-Instructional
22 Material
allocation of resources, so as to ensure a certain desirable level of service to Operations Research
patients, can be reached through OR. Similarly, for operating a transportation

service profitably, an optimum schedule for vehicles and crew members becomes
necessary. Punctuality, waiting time, total travel time, speed of the vehicles and
agreement with trade unions of crew members are some of the variables to be NOTES
taken into account while preparing optimum schedules, which can be worked out
utilizing OR techniques.
As stated earlier, operations research is generally concerned with problems
that are tactical rather than strategic in nature, that is, its use to long-range
organizational planning problems is very much limited. However, one can hope
that with further developments taking place in the field, operations research will be
able to deal with organizations in their entirety, rather than with slices through
them.
This description brings home the point that during the last four decades, the
scope of OR has extensively been widened and hopefully shall attain new strides
by the end of the present century. The art of systems analysis, so well developed
in the military context, will spread to other contexts—notably such civil government
branches as criminal justice, urban problems, housing, health, care, education and
social services. There is growing awareness amongst the people that unless they
make themselves familiar with OR techniques, they would not be able to understand
and appreciate the problems of modern business units. With computer facilities
becoming widespread the significance and scope of OR is likely to grow in the
coming years. In fact, the future holds great promise for the growth of OR in
power, scope and practical importance and utility. OR will continue its currently
vigorous efforts to reach out to new arenas of exploration and application.
OR and Modern Business Management
From what has just been stated, we can say that OR renders valuable service in
the field of business management. It ensures improvement in the quality of managerial
decisions in all functional areas of management. The role of OR in business
management can be mentioned as follows:
 Helps the directing authority: OR techniques help the directing authority
in optimum allocation of various limited resources, namely men, machines,
money, material, time, and so on, to different competing opportunities on an
objective basis for achieving effectively the goal of a business unit. They
help the chief of executive in broadening management vision and perspectives
in the choice of alternative strategies to the decision problems, such as
forecasting manpower, production capacities, capital requirements and plans
for their acquisition.
 Useful to the production management: OR is useful to the production
management in (i) selecting the building site for a plant, scheduling and
controlling its development and designing its layout; (ii) locating within the
Self-Instructional
Material 23
Operations Research plant and controlling the movements of required production materials and
finished goods inventories; (iii) scheduling and sequencing production by
adequate preventive maintenance with optimum number of operatives by
proper allocation of machines; and (iv) calculating the optimum product-
NOTES mix.
 Useful in personnel management: OR is useful to the personnel
management to find out (i) optimum manpower planning, (ii) the number of
persons to be maintained on the permanent or full- time roll, (iii) the number
of persons to be kept in a work pool intended for meeting the absenteeism,
(iv) the optimum manner of sequencing and routing of personnel to a variety
of jobs, and (v) in studying personnel recruiting procedures, accident rates
and labour turnover.
 Help in marketing management: OR techniques equally help the
marketing management to determine (i) where distribution points and
warehousing should be located, their size, quantity to be stocked and the
choice of customers; (ii) the optimum allocation of sales budget to direct
selling and promotional expenses; (iii) the choice of different media of
advertising and bidding strategies; and (iv) the consumer preferences relating
to size, colour, packaging, and so forth, for various products as well as to
outbid and outwit competitors.
 Useful in financial management: OR is also very useful to the financial
management in (i) finding long-range capital requirements as well as how to
generate these requirements, (ii) determining optimum replacement policies,
(iii) working out a plan for the firm, (iv) developing capital investments plans,
and (v) estimating credit and investment risks.
In addition to all this, OR provides the business executives an understanding
of the business operations which gives them new insights and capability to determine
better solutions for several decision-making problems with great speed, competence
and confidence. When applied on the level of management where policies are
formulated, OR assists the executives in an advisory capacity. However, on the
operational level where production, personnel, purchasing, inventory and
administrative decisions are made, it provides management with a means for
handling and processing information. Thus, in brief, OR can be considered as
scientific method of providing executive departments with a quantitative basis for
taking decisions regarding operations under their control.
OR though is a great aid to management, as outlined, but still it cannot be a
substitute for decision-making. The choice of a criterion as to what is actually best
for a business enterprise is still that of an executive who has to fall back upon his
experience and judgement. This is so because of the several limitations of OR.
Important limitations are given as follows:
Self-Instructional
24 Material
 The inherent limitations concerning mathematical expressions: OR Operations Research
involves the use of mathematical models, equations and similar other

mathematical expressions. Assumptions are always incorporated in the
derivation of an equation or model and such an equation or model may be
correctly used for the solution of the business problems. This happens when NOTES
the underlying assumptions and variables in the model are present in the
concerning problem. If this caution is not given due care, then there always
remains the possibility of wrong application of OR techniques. Quite often
the operations researchers have been accused of having many solutions
without being able to find problems that fit.
 High costs are involved in the use of OR techniques: OR techniques
usually prove very expensive. Services of specialized persons are invariably
called for (and along with it the use of computer) while using OR techniques.
As such only big concerns can think of using such techniques. Even in big
business organizations we can expect that OR techniques will continue to
be of limited use simply because they are not in many cases worth their
cost. As opposed to this, a typical manager, exercising intuition and
judgement, may be able to make a decision very inexpensively. Thus, the
use of OR is a costlier affair and this constitutes an important limitation of
operations research.
 OR does not take into consideration the intangible factors, that is,
non-measurable human factors: OR makes no allowance for intangible
factors, such as skill, attitude, vigour of the management people in taking
decisions, but in many instances success or failure hinges upon the
consideration of such non-measurable intangible factors. There cannot be
any magic formula for getting an answer to management problems; much
depends upon proper managerial attitudes and policies.
 OR is only a tool of analysis and not the complete decision-making
process: It should always be kept in mind that OR alone cannot make the
final decision. It is just a tool and simply suggests best alternatives, but in the
final analysis many business decisions will involve human element. Thus,
OR is at best a supplement rather than a substitute for management; subjective
judgement is likely to remain a principal approach to decision-making.
 Other limitations: Among other limitations of OR, the following deserve
to be mentioned:
(i) Bias: The operational researchers must be unbiased. An attempt to
shoehorn results into a confirmation of management’s prior preferences
can greatly increase the likelihood of failure.
(ii) Inadequate objective functions: The use of a single objective
function is often an insufficient basis for decisions. Laws, regulations,
public relations, market strategies, and so on, may all serve to overrule
a choice arrived at in this way.
Self-Instructional
Material 25
Operations Research (iii) Internal resistance: The implementation of an optimal decision may
also confront internal obstacles, such as trade unions or individual
managers with strong preferences for other ways of doing the job.
(iv) Competence: Competent OR analysis calls for the careful
NOTES
specification of alternatives, a full comprehension of the underlying
mathematical relationships and a huge mass of data. Formulation of
an industrial problem to an OR set programme is quite often a difficult
task.
(v) Reliability of the prepared solution: At times, a non-linear
relationship is changed to linear for fitting the problem to linear
programming pattern. This may disturb the solution.
Mathematical models have been constructed for the categorized OR problems
and methods for solving the models available in many cases. Such methods are
usually termed as OR techniques. Some of the important OR techniques often
used by decision-makers in modern times in business and industry are as follows:
 Linear programming: This technique is used in finding a solution for
optimizing a given objective, such as profit maximization or cost minimization
under certain constraints. This technique is primarily concerned with the
optimal allocation of limited resources for optimizing a given function. The
name linear programming is because of the fact that the model in such cases
consists of linear equations indicating linear relationship between the different
variables of the system. Linear programming technique solves product-mix
and distribution problems of business and industry. It is a technique used to
allocate scarce resources in an optimum manner in problems of scheduling,
product-mix, and so on. Key factors under this technique include an
objective function, choice among several alternatives, limits or constraints
stated in symbols and variables assumed to be linear.
 Waiting line or queuing theory: This theory deals with mathematical study
of queues. The queues are formed whenever the current demand for service
exceeds the current capacity to provide that service. Waiting line technique
concerns itself with the random arrival of customers at a service station
where the facility is limited. Providing too much of capacity will mean idle
time for servers and will lead to waste of money. On the other hand, if the
queue becomes long, there will be a cost due to waiting of units in the
queue. Waiting line theory, therefore, aims at minimizing the costs of both
servicing and waiting. In other words, this technique is used to analyse the
feasibility of adding facilities and to assess the amount and cost of waiting
time. With its help, we can find the optimal capacity to be installed, which
will lead to a sort of an economic balance between cost of service and cost
of waiting.
Self-Instructional
26 Material
 Inventory control/planning: It aims at optimizing inventory levels. Operations Research
Inventory may be defined as a useful idle resource which has economic
value, for example, raw materials, spare parts, finished products, and so
forth. Inventory planning, in fact, answers the two questions, namely: how
much to buy and when to buy? Under this technique, the main emphasis is NOTES
on minimizing costs associated with holding of inventories, procurement of
inventories and the shortage of inventories.
 Game theory: This theory is used to determine the optimum strategy in a
competitive situation. The simplest possible competitive situation is that of
two persons playing zero-sum game, that is, a situation in which two persons
are involved and one person wins exactly what the other loses. More
complex competitive situations of real life can as well be imagined where
game theory can be used to determine the optimum strategy.
 Decision theory: This theory concerns with making sound decisions under
conditions of certainty, risk and uncertainty. As a matter of fact, there are
three different kinds of under which decisions are made, namely deterministic,
stochastic and uncertainty. The decision theory explains how to select a
suitable strategy to achieve some object or goal under each of these three
states.
 Network analysis: It involes the determination of an optimum sequence
of performing certain operations concerning some jobs in order to minimize
overall time and/or cost. Programme Evaluation and Review Technique
(PERT), Critical Path Method (CPM) and other network techniques, such
as Gantt Chart comes under Network Analysis. Key concepts under this
technique are network of events and activities, resource allocation, time
and cost considerations, and network paths and critical paths.
 Simulation: It is a technique of testing a model that resembles a real-life
situation. This technique is used to imitate an operation prior to actual
performance. Two methods of simulation are there—Monte Carlo method
of simulation and System simulation method. The former one, using random
numbers, is used to solve problems which involve conditions of uncertainty
and the mathematical formulation is impossible. However, in case of System
simulation, there is a reproduction of the operating environment and the
system allows for analysing the response from the environment to alternative
management actions. This method draws samples from a real population
instead of drawing samples from a table of random numbers.
 Integrated production models: This technique aims at minimizing cost
with respect to workforce, production and inventory. This technique is a
highly complex one and is used only by big business and industrial units.
This technique can be used only when sales and costs statistics for a
considerable long period are available.
Self-Instructional
Material 27
Operations Research  Some other OR techniques: In addition, there are several other
techniques, such as non-linear programming, dynamic programming, search
theory, the theory of replacement, and so on. A brief mention of some of
these is as follows:
NOTES
(i) Non-linear programming: It is that form of programming in which
some or all of the variables are curvilinear. In other words, this means
that either the objective function or constraints or both are not in linear
form. In most of the practical situations, we encounter non-linear
programming problems, but for computation purpose, we approximate
them as linear programming problems. Even then, there may remain
some non-linear programming problems which may not be fully solved
by presently known methods.
(ii) Dynamic programming: It refers to a systematic search for optimal
solutions to problems that involve many highly complex interrelations
that are, moreover, sensitive to multistage effects such as successive
time phases.
(iii) Heuristic programming: It is also known as discovery method and
refers to step-by-step search towards an optimum when a problem
cannot be expressed in mathematical programming form. The search
procedure examines successively a series of combinations that lead to
stepwise improvements in the solution and the search stops when a
near optimum has been found.
(iv) Integer programming: It is a special form of linear programming in
which the solution is required in terms of integral numbers (that is,
whole numbers) only.
(v) Algorithmic programming: It is just the opposite of heuristic
programming. It may also be termed as near mathematical
programming. This programming refers to a thorough and exhaustive
mathematical approach to investigate all aspects of the given variables
in order to obtain optimal solution.
(vi) Quadratic programming: It refers to a modification of linear
programming in which the objective equations appear in quadratic
form, that is, they contain squared terms.
(vii) Parametric programming: It is the name given to linear programming
when the latter is modified for the purpose of inclusion of several
objective equations with varying degrees of priority. The sensitivity of
the solution to these variations is then studied.
(viii) Probabilistic programming: It also known as stochastic
programming and refers to linear programming that include an
evaluation of relative risks and uncertainties in various alternatives of
Self-Instructional choice for management decisions.
28 Material
(ix) Search theory: It concerns itself with search problems. A search Operations Research
problem is characterized by the need for designing a procedure to

collect information on the basis of which one or more decisions are
made. This theory is useful in places in which some events are known
NOTES
to occur, but the exact location is not known. The first search model
was developed during World War II to solve decision problems
connected with air patrols and their search for submarines. Advertising
agencies’ search for customers, personnel departments’ search for
good executives, and so on, are some of the examples of search
theory’s application in business.
(x) The theory of replacement: It is concerned with the prediction of
replacement costs and determination of the most economic
replacement policy. There are two types of replacement models—
one type of model deals in replacing equipment that deteriorate with
time and the other type of model helps in establishing replacement
policy for those equipment which fail completely and instantaneously.
All these techniques are not simple, but involve higher mathematics. The tendency
today is to combine several of these techniques and put them together into more
sophisticated and advanced programming models.
Check Your Progress

9. Mention the limitations of operations research.
10. How is operations research applied in the field of hospital management?
1.7 ANSWERS TO CHECK YOUR PROGRESS

QUESTIONS
1. The essential characteristics of operations research are as follows:

 Systems orientation
 Use of interdisciplinary terms
 Application of scientific methods
 Uncovering new problems
 Quantitative solutions
 Human factors
2. Operations research provides the executives with a more precise description
of the cause-and-effect relationship and risks underling the business
operations in measurable terms. This eliminates the conventional intuitive
and subjective basis on which managements used to formulate their decisions
Self-Instructional
decades ago. Material 29
Operations Research 3. The first step in the scientific method of operations research is to formulate
the problem in an appropriate form. Formulating a problem consists of
identifying, defining and specifying the measures of the components of a
decision model.
NOTES
4. The solution values of a decision model (obtained) are tested against actual
observations. In other words, effort is made to test the validity of the model
used. A model is supposed to be valid if it can give a reliable prediction of
the performance of the system represented through the model.
5. Stochastic situations are those in which each course of action can result in
alternative outcomes; the probabilities of which are known or can be
estimated.
6. Retentive objectives are concerned with preserving either resources of value,
such as money, time, energy or states, such as comfort, safety, stability and
others.
7. The properties of a good model are as follows:
 A good model must represent the reality or the system with which it is
concerned as accurately as possible and must not compromise with any
of the critical elements of the system.
 A good model must be simple enough so as to be readily understood
and, if necessary, must be capable of being modified quickly and
effectively.
 A good model is one that is capable of answering in case when small
perturbations take place in the data.
8. Quadratic programming refers to a modification of linear programming in
which the objective equations appear in quadratic forms, that is, they contain
squared terms.
9. The important limitations of operations research are as follows:
 The inherent limitations concerning mathematical expressions.
 High costs are involved in the use of operations research techniques.
 Operations research does not taken into consideration the intangible
factors.
 Operations research is only a tool of analysis and not the complete
decision-making process.
10. Operations research has been applied in the field of hospital management.
For example, hospital management quite often faces the problem of allotting
its limited resources, of its multiphased activities. Optimum allocation of
resources, so as to ensure a certain desirable level of service to patients,
can be reached through operations research.
Self-Instructional
30 Material
Operations Research
1.8 SUMMARY
 There have been various definitions for operations research, such as applied
decision-making, quantitative common sense and making of economic NOTES
decisions.
 An OR study generally involves three phases, namely the judgment phase,
the research phase and the action phase. Of these three, the research phase
is the longest and the largest, but the remaining two phases are very important
since they provide, respectively, the basis for and implementation of the
research phase.
 Mathematical models have been constructed for the categorized OR
problems and methods for solving the models available in many cases. Such
methods are usually termed as OR techniques.
 OR has gained increasing importance since World War II in the technology
of business and industry administration.
 OR, though a great aid to management, cannot be a substitute for decision-
making. The choice of a criterion as to what is actually best for a business
enterprise is still that of an executive who has to fall back upon his experience
and judgement.
 OR techniques, having their origin with war operations analysis during World
War II, have since then assisted defence management to a larger extent.
 To formulate a problem correctly, one must know what a problem is and
must understand the nature of the problem.
 Generally, the objectives of an organization may either be retentive or
acquisitive. Retentive objectives are concerned with preserving either
resources of value, such as money, time and energy, or states, such as
comfort, safety, stability and others.
 Modelling plays an important part in operational research. In fact, it is
considered as the essence of operations research approach.
1.9 KEY WORDS
 Gantt chart: It is a type of bar chart, adapted by Karol Adamiecki in 1896

and independently by Henry Gantt in the 1910s, which illustrates a project
schedule.
 Simulation: It is a technique of testing a model that resembles a real-life
situation.
 Linear programing: This technique is used in finding a solution for optimizing
a given objective, such as profit maximization or cost minimization under
certain constraints.
Self-Instructional
Material 31
Operations Research  Model: It is a physical or symbolic representation of the relevant aspects
of the reality or system with which one is concerned.
 Programme evaluation and review technique (PERT): It is a project
management tool used to schedule, organize and coordinate tasks within a
NOTES
project.
1.10 SELF ASSESSMENT QUESTIONS AND

EXERCISES
Short-Answer Questions
1. Write a short note on the important operations research techniques used in
modern business and industrial units.
2. What is the role of operations research techniques in business and industry?
3. Write a brief note on the various phases in solving an operations research
problem.
4. Write a short note on problem formulation in operational research.
5. What are the properties of a good model?
6. Outline the important types of operations research models.
7. What are the important aspects pertaining to modelling?
Long-Answer Questions
1. Define operations research. Explain some of its features.
2. Discuss the limitations of operations research techniques.
3. Is it possible to make all business decisions with the assistance of operations
research? Give detailed reasons to support your answer.
4. Enumerate the advantages of modelling in OR.
5. ‘Modelling enables quick and economical experimentation for finding an
optimum solution to a given problem.’ Do you agree? Give reasons for
your answer.
1.11 FURTHER READINGS
Sharma, A. 2009. Operations Research. New Delhi: Himalaya Publishing House.

Singh, J. K. 2009. Business Mathematics. New Delhi: Himalaya Publishing
House.
Kalavathy, S. 2010. Operations Research with C Programmes, 3rd edition.
New Delhi: Vikas Publishing House Pvt. Ltd.
Kothari, C. R. 1992. An Introduction to Operational Research. New Delhi:
Self-Instructional Vikas Publishing House Pvt. Ltd.
32 Material
Linear Programming
UNIT 2 LINEAR PROGRAMMING

Structure NOTES
2.0 Introduction
2.1 Objectives
2.2 Formulation of Linear Programming
2.2.2 Fields Where Linear Programming Can Be Used
2.2.5 Case Studies
2.4 Summary
2.5 Key Words
2.0 INTRODUCTION
In a mathematical model, linear programming is one method to achieve the best
outcome whose requirements are represented by linear relationships. Basically,
linear programming is a technique for the optimization of a linear objective function.
In this unit, you will study about formulation and meaning of linear programming,
the fields where linear programming can be used. You will also learn about the
components and presentation of a Linear Programming Problem (LPP).
2.1 OBJECTIVES
 Understand the meaning and formulation of linear programming
 Discuss the components of linear programming problem
 Describe the general form of the linear programming model
2.2 FORMULATION OF LINEAR PROGRAMMING

Decision-making has always been very important in the business and industrial
world, particularly with regard to the problems concerning production of
commodities.
The main questions that come up before a production manager are as follows:
(i) Which commodity(s) should be produced?
(ii) In what quantities and by which process or processes should they be
produced? Self-Instructional
Material 33
Linear Programming Alfred Marshall, an influential British economist, in this connection points out that
the businessman always studies his production function and input prices, and
substitutes one input for another till his costs become the minimum possible. This
sort of substitution, in the opinion of Marshall, is done by the businessman’s trained
NOTES instinct rather than with formal calculations. However, now there does exist a
method of formal calculations often termed as Linear Programming. This method
was first formulated by a Russian mathematician L.V. Kantorovich, but it was
developed later in 1947 by an American mathematical scientist George B. Dantzig
‘for the purpose of scheduling the complicated procurement activities of the United
States Air Force’. Today, this method is being used in solving a wide range of
practical business problems. The advent of electronic computers has further
increased its applications to solve many other problems in industry. It is being
considered as one of the most versatile management techniques.
Linear Programming (LP) is a major innovation since World War II in the field of
business decision-making, particularly under conditions of certainty. The word
‘Linear’ means that the relationships are represented by straight lines, that is, the
relationships are of the form y = a + bx, and the word ‘Programming’ means
taking decisions systematically. Thus, LP is a decision-making technique under
given constraints on the assumption that the relationships amongst the variables
representing different phenomena happen to be linear. In fact, Dantzig originally
called it ‘programming of interdependent activities in a linear structure’, but later
on shortened it to ‘Linear Programming’.
LP is generally used in solving maximization (sales or profit maximization)
or minimization (cost minimization) problems subject to certain assumptions. Putting
in a formal way, ‘Linear Programming is the maximization (or minimization) of a
linear function of variables subject to a constraint of linear inequalities.’ Hence, LP
is a mathematical technique designed to assist the organization in optimally allocating
its available resources under conditions of certainty in problems of scheduling,
product mix, and so on.
2.2.2 Fields where Linear Programming can be Used
The problem for which LP provides a solution may be stated as, maximize or
minimize for some dependent variable which is a function of several independent
variables when the independent variables are subject to various restrictions. The
dependent variable is usually some economic objective such as profits, production,
costs, workweeks, tonnage to be shipped, and so on. More profits are generally
preferred to less profits and lower costs are preferred to higher costs. Hence, it is
appropriate to represent either maximization or minimization of the dependent
variable as one of the firm’s objective. LP is usually concerned with such objectives
under given constraints with linearity assumptions. In fact, it is powerful to take in
its stride a wide range of business applications. The applications of LP are numerous
Self-Instructional
34 Material
and are increasing everyday. LP is extensively used in solving resource allocation Linear Programming
problems. Production planning and scheduling, transportation, sales and advertising,

financial planning, portfolio analysis, corporate planning, and so on, are some of
its most fertile application areas. More specifically, LP has been successfully applied
in the following fields: NOTES
 Agriculture: LP can be applied in farm management problems as it relates
to the allocation of resources, such as acreage, labour, water supply or
working capital in such a way that it maximizes net revenue.
 Contract awards: Evaluation of tenders by recourse to LP guarantees that
the awards are made in the cheapest way.
 Industries: Applications of LP in business and industry are of the most
diverse type. Transportation problems concerning cost minimization can be
solved by this technique. The technique can also be adopted in solving
problems of production (product mix) and inventory control.
Thus, LP is the most widely used technique of decision-making in business and
industry in modern times in various fields.
There are certain basic concepts and notations to be first understood for easy
adoption of the LP technique. A brief mention of such concepts is as follows:
 Linearity: The term ‘linearity’ implies straight line or proportional
relationships among the relevant variables. Linearity in economic theory is
known as constant returns, which means that if the amount of the input
doubles, the corresponding outputs and profits are also doubled. Linearity
assumption, thus, implies that two machines and two workers can produce
twice as much as one machine and one worker; four machines and four
workers twice as much as two machines and two workers, and so on.
 Process and its level: Process means the combination of particular inputs
to produce a particular output. In a process, factors of production are used
in fixed ratios, of course, depending upon technology and as such no
substitution is possible with a process. There may be many processes open
to a firm for producing a commodity and one process can be substituted for
another. There is, thus, no interference of one process with another when
two or more processes are used simultaneously. If a product can be produced
in two different ways, then there are two different processes (or activities
or decision variables) for the purpose of a linear programme.
 Criterion function: This is also known as the objective function which
states whether the determinants of the quantity should be maximized or
minimized. For example, revenue or profit is such a function when it is to be
maximized or cost is such a function when the problem is to minimize it. An
objective function should include all the possible activities with the revenue
(profit) or cost coefficients per unit of production or acquisition. The goal
Self-Instructional
Material 35
Linear Programming may be either to maximize this function or to minimize this function. In
symbolic form, let ZX denote the value of the objective function at the X
level of the activities included in it. This is the total sum of individual activities
produced at a specified level. The activities are denoted as j =1, 2,..., n.
NOTES The revenue or cost coefficient of the jth activity is represented by Cj.
Thus, 2X1, implies that X units of activity j = 1 yields a profit (or loss) of
C1 = 2.
 Constraints or inequalities: These are the limitations under which one has
to plan and decide, that is, restrictions imposed upon decision variables. For
example, a certain machine requires one worker to operate; another machine
requires at least four workers (that is, > 4); there are at most 20 machine
hours (that is, < 20) available; the weight of the product should be, say 10
lbs., and so on, are all examples of constraints or what are known as inequalities.
Inequalities like X > C (reads X is greater than C) or X < C (reads X is less
than C) are termed as strict inequalities. The constraints may be in the form of
weak inequalities like X  C (reads X is less than or equal to C) or X  C
(reads C is greater than or equal to C). Constraints may be in the form of
strict equalities like X = C (reads X is equal to C).
Let bi denote the quantity b of resource i available for use in various
production processes. The coefficient attached to resource i is the quantity
of resource i required for the production of one unit of product j.
 Feasible solutions: These are all those possible solutions which can be
worked upon under given constraints. The region comprising all feasible
solutions is referred to as Feasible Region.
 Optimum solution: Optimum solution is the best of all the feasible solutions.
General form of the linear programming model
Linear programming problem mathematically can be stated as follows:
Choose the quantities,
Xj > 0 (j = 1,..., n) ...(2.1)
This is also known as the non-negativity condition and in simple terms means
that no X can be negative.
To maximize,
n
Z  C j X j ...(2.2)
j 1
Subject to the constraints,

n
aij X j bi (i = 1,...,m) ...(2.3)
j 1
Self-Instructional
36 Material
This is the usual structure of a linear programming model in the simplest possible Linear Programming
form. This model can be interpreted as a profit maximization situation, where n

production activities are pursued at level Xj which have to be decided upon, subject
to a limited amount of m resources being available. Each unit of the jth activity
yields a return C and uses an amount aij of the ith resource. Z denotes the optimal NOTES
value of the objective function for a given system.
Assumptions or conditions to be fulfilled
LP model is based on the assumptions of proportionality, certainty, additivity,
continuity and finite choices.
Proportionality is assumed in the objective function and the constraint
inequalities. In economic terminology, this means that there are constant returns to
scale, that is, if one unit of a product contributes 5 toward profit, then 2 units will
contribute 10, 4 units 20, and so on.
Certainty assumption means the prior knowledge of all the coefficients in
the objective function, the coefficients of the constraints and the resource values.
LP model operates only under conditions of certainty.
Additivity assumption means that the total of all the activities is given by
the sum of each activity conducted separately. For example, the total profit in the
objective function is equal to the sum of the profit contributed by each of the
products separately.
Continuity assumption means that the decision variables are contiunous.
Accordingly, the combinations of output with fractional values, in case of product
mix problems, are possible and obtained frequently.
Finite choices assumption implies that finite number of choices are available
to a decision-maker and the decision variables do not assume negative values.
The procedure for mathematical formulation of an LPP consists of the following
steps:
Step 1: The decision variables of the problem are noted.
Step 2: The objective function to be optimized (maximized or minimized) as a
linear function of the decision variables is formulated.
Step 3: The other conditions of the problem, such as resource limitation, market
constraints, interrelations between variables, and so on, are formulated as linear
inequations or equations in terms of the decision variables.
Step 4: The non-negativity constraint from the considerations is added so that the
negative values of the decision variables do not have any valid physical interpretation.
The objective function, the set of constraints and the non-negative constraint
together form a linear programming problem.
Self-Instructional
Material 37
Linear Programming General formulation of a linear programming problem
The general formulation of the LPP can be stated as follows:
In order to find the values of n decision variables x1 x2 ... xn to maximize or
NOTES minimize the objective function,
Z  C1 x1  C2 x2    Cn xn ... (2.4)
a11 x1  a12 x2    a1n xn (, , )b1 

a21 x1  a22 x2    a2 n xn (, , )b2 
: 
 ... (2.5)
ai1 x1  ai 2 x2    ain xn (, , )bi 
: 

am1 x1  am 2 x2    amn xn ()bm 
Here, the constraints can be inequality  or  or even in the form of an

equation (=) and finally satisfy the non-negative restrictions:
x1 0, x2 0 xn 0 ... (2.6)
Matrix form of a linear programming problem
The LPP can be expressed in the matrix form as follows:
Maximize or minimize Z = Cx  Objective function
Subject to Ax () b  Constant equation
b > 0, x  0  Non-negativity restrictions
Where, x = ( x1 , x2  xn )
C = (C1 , C2 Cn )
a11a12  a1n
b1
a21a22  a2 n
b = b2 A
:
bm
am1am 2  amn
2.2.5 Case Studies

Example 2.1: A manufacturer produces two types of models, M1 and M2. Each
model of the type M1 requires 4 hours of grinding and 2 hours of polishing, whereas
each model of the type M2 requires 2 hours of grinding and 5 hours of polishing.
The manufacturers have 2 grinders and 3 polishers. Each grinder works 40 hours
Self-Instructional
38 Material
a week and each polisher works for 60 hours a week. The profit on M1 model is Linear Programming
3.00 and on model M2 is 4.00. Whatever is produced in a week is sold in the

market. How should the manufacturer allocate his production capacity to the two
types of models, so that he may make the maximum profit in a week?
NOTES
Solution:
Decision variables: Let X1 and X2 be the number of units of M1 and M2.
Objective function: Since the profit on both the models are given, we
have to maximize the profit, namely:
Max Z = 3X1 + 4X2
Constraints: There are two constraints: one for grinding and the other for
polishing.
The number of hours available on each grinder for one week is 40 hours.
There are 2 grinders. Hence, the manufacturer does not have more than 2 × 40 =
80 hours for grinding. M1 requires 4 hours of grinding and M2 requires 2 hours of
grinding.
The grinding constraint is given by,
4 X 1  2 X 2  80
Since there are 3 polishers, the available time for polishing in a week is
given by 3 × 60 = 180. M1 requires 2 hours of polishing and M2 requires 5 hours
of polishing. Hence, we have 2X1 + 5X2  180
Thus, we have,
Max Z = 3X1 + 4X2
Subject to 4 X 1  2 X 2  80
2 X 1  5 X 2  180
X1, X 2  0
Example 2.2: A firm produces two different types of products, Product M and
Product N. The firm uses the same machinery for manufacturing both the products.
One unit of Product M requires 10 minutes while one unit of Product N requires 2
minutes. The maximum hours the machine can function optimally for a week is 35
hours. The raw material requirement for Product M is 1 kg, while that of Product
N is 0.5 kg. Also, the market constraint on product M is 600 kg, while that of
Product N is 800 units per week. The cost of manufacturing Product M is 5 per
unit and it is sold at 10, while the cost of Product N is 6 per unit and sold at
8 per unit. Calculate the total number of units of Product M and Product N that
should be produced per week, so as to derive maximum profit.
Self-Instructional
Material 39
Linear Programming Solution:
Decision variables: Let X1 and X2 be the number of products of A and B.
Objective function: Cost of product A per unit is 5 and is sold at 10 per
NOTES unit.
Profit on one unit of product A = 10 – 5 = 5.
X1 units of product A contribute a profit of 5X1 from one unit of product.
Similarly, profit on one unit of B = 8 – 6 = 2.
X2 units of product B contribute a profit of 2X2.
The objective function is given by,
Max Z  5 X 1  2 X 2
Constraints: Time requirement constraint is given by,
10 X 1  2 X 2  (35  60)
10 X 1  2 X 2  2100
Raw material constraint is given by,
X 1  0.5 X 2  600
Market demand on product B is 800 units every week.
X2  800
The complete LPP is,
Max Z  5 X 1  2 X 2
Subject to, 10 X 1  2 X 2  2100

X 1  0.5 X 2  600
X 2  800
X1, X 2  0
Example 2.3: A person requires 10, 12 and 12 units of chemicals A, B and C,

respectively, for his garden. A liquid product contains 5, 2 and 1 units of A, B and
C respectively per jar. A dry product contains 1, 2 and 4 units of A, B, C per
carton. If the liquid product sells for 3 per jar and the dry product sells for 2
per carton, what should be the number of jars that needs to be purchased, in
order to bring down the cost and meet the requirements?
Solution:
Decision variables: Let X1 and X2 be the number of units of liquid and dry
products.
Self-Instructional
40 Material
Objective function: Since the cost for the products is given, we have to Linear Programming
minimize the cost.

Min Z = 3X1 + 2X2
Constraints: As there are three chemicals and their requirements are given, NOTES
we have three constraints for these three chemicals.
5 X 1  X 2  10
2 X 1  2 X 2  12
X 1  4 X 2  12
Hence, the complete LPP is,

Min Z = 3X1 + 2X2
Subject to,
5 X 1  X 2  10
2 X 1  2 X 2  12
X 1  4 X 2  12
X1, X 2  0
Example 2.4: A paper mill produces two grades of paper, X and Y. Because of
raw material restrictions, it cannot produce more than 400 tonnes of grade X and
300 tonnes of grade Y in a week. There are 160 production hours in a week. It
requires 0.2 and 0.4 hours to produce a tonne of products X and Y respectively
with corresponding profits of 200 and 500 per tonne. Formulate this as an
LPP to maximize profit and find the optimum product mix.
Solution:
Decision variables: Let X1 and X2 be the number of units of the two
grades of paper, X and Y.
Objective function: Since the profit for the two grades of paper X and Y
are given, the objective function is to maximize the profit.
Max Z = 200X1 + 500X2
Constraints: There are two constraints, one with reference to raw material,
and the other with reference to production hours.
Max Z = 200X1 + 500X2
Subject to,
X 1  400
X 2  300
0.2 X 1  0.4 X 2  160
Non-negative restriction X1, X2  0

Self-Instructional
Material 41
Linear Programming Example 2.5: A company manufactures two products, A and B. Each unit of B
takes twice as long to produce as one unit of A and if the company were to produce
only A, it would have time to produce 2000 units per day. The availability of the raw
material is enough to produce 1500 units per day of both A and B together. Product
NOTES B requiring a special ingredient, only 600 units of it can be made per day. If A fetches
a profit of 2 per unit and B a profit of 4 per unit, find the optimum product mix by
the graphical method.
Solution: Let X1 and X2 be the number of units of products A and B, respectively.
The profit after selling these two products is given by the objective function,
Max Z = 2X1 + 4X2
Since the company can produce at the most 2000 units of the product in a
day and Product B requires twice as much time as that of Product A, production
restriction is given by,
X 1  2 X 2  2000
Since the raw material is sufficient to produce 1500 units per day of both
A and B, we have X 1  X 2  1500.
There are special ingredients for Product B; hence, we have X2  600.
Also, since the company cannot produce negative quantities X1  0 and
X2  0.
Hence, the problem can be finally put in the form:
Find X1 and X2 such that the profits, Z = 2X1 + 4X2 is maximum.
Subject to, X 1  2 X 2  2000
X 1  X 2  1500
X 2  600
X1, X 2  0
Example 2.6: A firm manufacturers 3 products A, B and C. The profits are 3,
2 and 4, respectively. The firm has 2 machines and the following is the required
processing time in minutes for each machine on each product.
Product
A B C
Machines C 4 3 5
D 3 2 4
Machines C and D have 2000 and 2500 machine minutes respectively. The
firm must manufacture 100 units of A, 200 units of B and 50 units of C, but not
more than 150 units of A. Formulate an LP problem to maximize the profit.
Self-Instructional
42 Material
Solution: Let X1, X2, X3 be the number of units of the product A, B, C, respectively. Linear Programming
Since the profits are 3, 2 and 4, respectively, the total profit gained by
the firm after selling these three products is given by,
Z  3 X1  2 X 2  4 X 3 NOTES
The total number of minutes required in producing these three products at
Machine C is given by 4X1 + 3X2 + 5X3 and at Machine D is given by 3X1 + 2X2
+ 4X3.
The restrictions on machines C and D are given by 2000 minutes and 2500
minutes.
4 X 1  3 X 2  5 X 3  2000
3 X 1  2 X 2  4 X 3  2500
Also, since the firm manufactures 100 units of A, 200 units of B and 50 units
of C, but not more than 150 units of A, the further restriction becomes,
100  X 1  150
200  X 2  0
50  X 3  0
Hence, the allocation problem of the firm can be finally put in the form:
Find the value of X1, X2, X3 so as to maximize,
Z = 3X1 + 2X2 + 4X3
4 X 1  3 X 2  5 X 3  2000
3 X 1  2 X 2  4 X 3  2500
100  X 1  150, 200  X 2  0,50  X 3  0

Example 2.7: A peasant has 100 acres of farm. He can sell all potatoes, cabbage
or brinjals and can increase the cost to get 1.00 per kg for potatoes, 0.75
head for cabbage and 2.00 per kg for brinjals. The average yield per acre is
2000 kg of potatoes, 3000 heads of cabbage and 1000 kg of brinjals. Fertilizers
can be bought at 0.50 per kg and the amount needed per acre is 100 kg each for
potatoes and cabbage and 50 kg for brinjals. The manpower required for sowing,
cultivating and harvesting per acre is 5 man-days for potatoes and brinjals and 6
man-days for cabbage. A total of 400 man-days of labour is available at
20 per man-day. Solve this example as a linear programming model to increase
the peasant’s profit.
Solution: Let X1, X2, X3 be the area of his farm to grow potatoes, cabbage and
brinjals respectively. The peasant produces 2000X1 kg of potatoes, 3000X2 heads
of cabbage and 1000X3 kg of brinjals.
Self-Instructional
Material 43
Linear Programming The total sales of the peasant will be,
= (2000X1 + 0.75 × 3000X2 + 2 × 1000X3)
Fertilizer expenditure will be,
NOTES = 20 (5X1 + 6X2 + 5X3)
Peasant’s profit will be,
Z = Sale (in ) – Total expenditure (in )
= (2000X1 + 0.75 × 3000X2 + 2 × 1000X3) – 0.5 × [100(X1 + X2) +
50X2]
–20 × (5X1 + 6X2 + 5X3)
Z = 1850X1 + 2080X2 + 1875X3
Since the total area of the farm is restricted to 100 acres,
X1 + X2 + X3  100
Also, the total man-days manpower is restricted to 400 man-days.
5 X 1  6 X 2  5 X 3  400
Hence, the peasant’s allocation problem can be finally put in the form:
Find the value of X1, X2 and X3 so as to maximize,
Z = 1850X1 + 2080X2 + 1875X3
Subject to,
X 1  X 2  X 3  100
5 X 1  6 X 2  5 X 3  400
X1, X 2 , X 3  0
Example 2.8: ABC Company produces two products: juicers and washing
machines. Production happens in two different departments, I and II. Juicers are
made in Department I and washing machines in Department II. These two items
are sold weekly. The weekly production should not cross 25 juicers and 35 washing
machines. The organization always employs a total of 60 employees in the two
departments. A juicer requires two man-weeks’ labour, while a washing machine
needs one man-week’s labour. A juicer makes a profit of 60 and a washing
machine contributes a profit of 40. How many units of juicers and washing
machines should the organization make to achieve the maximum profit? Formulate
this as an LPP.
Solution: Let X1 and X2 be the number of units of juicers and washing machines
to be produced.
Each juicer and washing machine contributes a profit of 60 and 40.
Hence, the objective function is to maximize Z = 60X1 + 40X2.
Self-Instructional
44 Material
There are two constraints which are imposed: weekly production and labour. Linear Programming
Since the weekly production cannot exceed 25 juicers and 35 washing

machines,
X 1  25 NOTES
X 2  35
A juicer needs two man-weeks of hard work and a washing machine needs
one man-week of hard work and the total number of workers is 60.
2X1 + X2  60
Non-negativity restrictions: Since the number of juicers and washing machines
produced cannot be negative, we have X1  0 and X2  0.
Hence, the production of juicers and washing machines problem can be
finally put in the form of an LP model given as follows:
Find the value of X1 and X2 so as to maximize,
Z = 60X1 + 40X2
Subject to,
X 1  25
X 2  35
2 X 1  X 2  60
and X 1 , X 2  0
It is take noted that graphical methods to solve linear programming problem

have been discussed in detail in Unit 3.
Advantages/Applications
Linear programming helps to reduce the cost of production for maximum output.
In brief, linear programming models assist a decision-maker can most efficiently
and effectively employ his production factor and limited resources to get maximum
profit at minimum cost.
Limitations
(i) Linear programming can be applied only when the objective function and
all the restraints can be expressed in terms of linear equations.
(ii) Linear programming techniques provide solutions only when all the elements
related to a problem can be quantified.
(iii) Linear programming technique may give fractional valued answer which is
not desirable in some problems.
Self-Instructional
Material 45
Linear Programming
Check Your Progress

1. When and by whom linear programming method was formulated and
NOTES developed?
2. List the various fields where Linear Programming has been frequently
applied.
3. What do you mean by the term ‘linearity’?
4. What does the assumption ‘finite choices’ in an LP problem imply?
5. Mention the assumptions on which LP model is based.

QUESTIONS
1. The linear programming method was first formulated by a Russian

mathematician L.V. Kantorovich, but it was later developed by an American
mathematical scientist George B. Dantzig in 1947 ‘for the purpose of
scheduling the complicated procurement activities of the United States Air
Force’.
2. Linear programming has been successfully applied in the following fields:
 Agriculture: LP can be applied in farm management problems as it
relates to the allocation of resources, such as acreage, labour, water
supply or working capital in such a way that it maximizes net revenue.
 Contract awards: Evaluation of tenders by recourse to LP guarantees
that the awards are made in the cheapest way.
 Industries: Applications of LP in business and industry are of the most
diverse type. Transportation problems concerning cost minimization can
be solved by this technique. The technique can also be adopted in solving
problems of production (product mix) and inventory control.
3. The term ‘linearity’ implies straight line or proportional relationships among
the relevant variables. In economic theory, linearity is known as constant
returns, which means that if the amount of the input doubles, the
corresponding outputs and profits are also doubled.
4. Finite choices assumption implies that finite number of choices are available
to a decision-maker and the decision variables do not assume negative
values.
5. The LP model is based on a few assumptions or conditions such as
proportionality, certainty, additivity, continuity, and finite choices.
Self-Instructional
46 Material
Linear Programming
2.4 SUMMARY
 Linear programming (LP) is a decision-making technique under given

constraints on the assumption that the relationships among the variables NOTES
representing different phenomena happen to be linear.
 LP is the most widely used technique of decision-making in business and
industry. Thus, it finds application in production planning and scheduling,
transportation, sales and advertising, financial planning, portfolio analysis,
corporate planning, and many other fields.
 LP is extensively used in solving resource allocation problems. Production
planning and scheduling, transportation, sales and advertising, financial
planning, portfolio analysis, corporate planning, and so on, are some of its
most fertile application areas.
 The term ‘linearity’ implies straight line or proportional relationships among
the relevant variables. Linearity in economic theory is known as constant
returns, which means that if the amount of the input doubles, the
corresponding outputs and profits are also doubled.
 Criterion function, which is also known as objective function, is a component
of LP problems. It states whether the determinants of quantity should be
maximized or minimized.
 Constraints are the limitations under which one has to plan and decide, that
is, restrictions imposed upon decision variables.
 Feasible solutions are all those possible solutions which can be worked
upon under given constraints.
 LP model is based on the assumptions of proportionality, certainty, additivity,
continuity and finite choices.
2.5 KEY WORDS
 Linear Programming: It refers to a mathematical technique that is used to

determine the best possible outcome or solution from a given set of
parameters or list of requirements, which are represented in the form of
linear relationships.
 Criterion function: It refers to an objective function which states the
determinants of the quantity to be either maximized or minimized.
 Linearity: It refers to the property of a mathematical relationship or function
which means that it can be graphically represented as a straight line.
Self-Instructional
Material 47
Linear Programming
EXERCISES
NOTES Short Answer Questions

1. What do you mean by linear programming?
2. What is meant by proportionality in linear programming?
3. Write a brief note on certainty in linear programming.
4. What are the basic constituents of an LP model?
Long Answer Questions
1. Analyse the fields where linear programming can be used.
2. Discuss the components of a linear programming problem.
3. Describe the steps used in formulation of a Linear Programming problem
(LPP).

Singh, J. K. 2009. Business Mathematics. New Delhi: Himalaya Publishing House.
Vikas Publishing House Pvt. Ltd.
Self-Instructional
48 Material
Graphical Analysis
UNIT 3 GRAPHICAL ANALYSIS of LPP
OF LPP
NOTES
Structure
3.0 Introduction
3.1 Objectives
3.2 Procedure for Solving LPP by Graphical Method: Some Exceptional Cases
3.3 Canonical or Standard forms of LPP
3.5 Summary
3.6 Key Words
3.0 INTRODUCTION
The graphical method for linear programming is used to solve problems by finding
the highest or lowest point of intersection between the objective function line and
the feasible region on a graph. It is regarded as the best alternative for the
representation and solving linear programming models that have two decision
variables.
In this unit, you will study about the graphical analysis of LPP and will also
learn the procedure to solve LPP by graphical method. In addition to this, you will
also analyse the general formulation of LPP evaluate the matrix form of LPP.
3.1 OBJECTIVES

 Understand the concept of graphical analysis of LPP
 Examine the procedure for solving LPP by graphical method
 State the general formulation of LPP
 Assess the matrix form of LPP
Self-Instructional
Material 49
Graphical Analysis
of LPP 3.2 PROCEDURE FOR SOLVING LPP BY
GRAPHICAL METHOD: SOME
EXCEPTIONAL CASES
NOTES
The steps involved in graphical method are as follows:
Step 1 Consider each inequality constraint as an equation.
Step 2 Plot each equation on the graph, as each will geometrically represent a
straight line.
Step 3 Mark the region. If the inequality constraint corresponding to that line is
, then the region below the line lying in the first quadrant (due to non-negativity
of variables) is shaded. For the inequality constraint  sign, the region above the
line in the first quadrant is shaded. The points lying in the common region will
satisfy all the constraints simultaneously. The common region thus obtained is
called the ‘feasible region’.
Step 4 Assign an arbitrary value, say zero, to the objective function.
Step 5 Draw the straight line to represent the objective function with the arbitrary
value (i.e., a straight line through the origin).
Step 6 Stretch the objective function line till the extreme points of the feasible
region. In the maximization case, this line will stop farthest from the origin, passing
through at least one corner of the feasible region. In the minimization case, this
line will stop nearest to the origin, passing through at least one corner of the
feasible region.
Step 7 Find the co-ordinates of the extreme points selected in step 6 and find
the maximum or minimum value of Z.
Note: As the optimal values occur at the corner points of the feasible region,
it is enough to calculate the value of the objective function of the corner points
of the feasible region and select the one that gives the optimal solution. That is,
in the case of maximization problem, the optimal point corresponds to the corner
point at which the objective function has a maximum value, and in the case of
minimization, the optional solution is the corner point which gives the objective
function the minimum value for the objective function.
Example 3.1: Solve the following LPP by graphical method.
Minimize Z = 20X1 + 10X2
Subject to, X1 + 2X2  40
3X1 + X2  30
4X1 + 3X2  60
X1, X2  0.
Self-Instructional
50 Material
Solution: Replace all the inequalities of the constraints by equation Graphical Analysis
of LPP
X1 + 2X2 = 40 If X1 = 0 X2 = 20
If X2 = 0 X1 = 40
 X1 + 2X2 = 40 passes through (0, 20) (40, 0)
NOTES
3X1 + X2 = 30 passes through (0, 30) (10, 0)
4X1 + 3X2 = 60 passes through (0, 20) (15, 0)
Plot each equation on the graph.
The feasible region is ABCD.

C and D are points of intersection of lines.
C intersect X1 + 2X2 = 40, 3X1 + X2 = 30
and, D intersect 4X1 + 3X2 = 60, X1 + X2 = 30
C = (4, 18)
D = (6, 12)
Corner points Value of Z = 20X1 + 10X2
A (15, 0) 300
B (40, 0) 800
C (4, 18) 260
D (6, 12) 240 (Minimum value)
 The minimum value of Z occurs at D (6, 12). Hence, the optimal solution is
X1 = 6, X2 = 12.
Self-Instructional
Material 51
Graphical Analysis Example 3.2: Find the maximum value of Z = 5X1 + 7X2
of LPP
x1 + x2  4
NOTES 3x1 + 8x2  24
10x1 + 7x2  35
x1, x2  0
Solution: Replace all the inequalities of the constraints by forming equations
x1 + x2 = 4 passes through (0, 4) (4, 0)
3x1 + 8x2 = 24 passes through (0, 3) (8, 0)
10x1 + 7x2 = 35 passes through (0, 5) (3.5, 0)
Plot these lines on the graph and mark the region below the line as the
inequality of the constraint as  which is also lying in the first quadrant.
The feasible region is OABCD.

B and C are points of intersection of lines
B intersect x1 + x2 = 4, 10x1 + 7x2 = 35
and C intersect 3x1 + 8x2 = 24, x1 + x2 = 4.
On solving we get, B = (1.6, 2.3)
C = (1.6, 2.4)
Corner points Value of Z = 5x1 + 7x2
O (0, 0) 0
A (3.5, 0) 17.5
B (1.6, 2.3) 25.1
C (1.6, 2.4) 24.8 (Maximum value)
D (0, 3) 21
Self-Instructional
52 Material
 The maximum value of Z occurs at C (1.6, 2.4) and the optimal solution is x1 Graphical Analysis
of LPP
= 1.6, x2 = 2.4.
Example 3.3: A company produces 2 types of hats. Every hat A requires twice
as much labour time as the second hat B. If the company produces only hat B then
NOTES
it can produce a total of 500 hats a day. The market limits daily sales of hat A and
B to 150 and 250 respectively. The profits on hat A and B are Rs. 8 and Rs. 5
respectively. Solve graphically to get the optimal solution.
Solution: Let X1 and X2 be the number of units of type A and type B hats
respectively.
Max Z = 8X1 + 5X2
Subject to, 2X1 + 2X2  500
X1  150
X2  250
X1, X2  0.
First rewrite the inequality of the constraint into an equation and plot the
lines on the graph.
2X1 + X2 = 500 passes through (0, 500) (250, 0)
X1 = 150 passes through (150, 0)
X2 = 250 passes through (0, 250)
We mark the region below the lines lying in the first quadrant as the inequality
of the constraints are . The feasible region is OABCD. B and C are points of
intersection of lines
2X1 + X2 = 500,
X1 = 150 (B intersect) and X2 = 250 (C intersect)
On solving, we get B = (150, 200)
C = (125, 250)
Self-Instructional
Material 53
Graphical Analysis Corner points Value of Z = 8X1 + 5X2
of LPP
O (0, 0) 0
A (150, 0) 1, 200
NOTES B (150, 200) 2, 200
C (125, 250) 2, 250 (Maximum Z = 2,250)
D (0, 250) 1, 250
The maximum value of Z is attained at C (125, 250)
 The optimal solution is X1 = 125, X2 = 250.
i.e., The company should produce 125 hats of type A and 250 hats of type
B in order to get the maximum profit of Rs. 2, 250.
Example 3.4: By graphical method solve the following LPP.
Max Z = 3X1 + 4X2
3X1 + 5X2  150
5X1 + 4X2  100
8X1 + 4X2  80
and X1, X2  0.
Solution:
Replacing the inequality by equality
5x1 + 4x2 = 200 passes through (0, 50), (40, 0)
Self-Instructional
54 Material
Feasible region is given by OABCD. Graphical Analysis
of LPP
O (20, 0) 60
A (40, 0) 120 NOTES
B (30.8, 11.5) 138.4 (Maximum value)
C (0, 30) 120
D (0, 25) 100
 The maximum value of Z is attained at B (30.8, 11.5)
 The optimal solution is X1 = 30.8, X2 = 11.5.
Example 3.5: Use graphical method to solve the LPP.
Maximize Z = 6X1 + 4X2
Subject to, –2X1 + X2  2
X1 – X2  2
3X1 + 2X2  9
X1, X2  0.
Solution:
Replacing the inequality by equality
–2x1 + x2 = 2 passes through (0, 2), (–1, 0)
x1 – x2 = 2 passes through (0, –2), (2, 0)
3x1 + 2x2 = 9 passes through (0, 4.5), (3, 0)
Self-Instructional
Material 55
Graphical Analysis Feasible region is given by ABC.
of LPP
0 (0, 0) 0
NOTES A (2, 0) 12
78  12 90
B (13/5, 3/5)  = 18 (Maximum value)
5 5
5 24  26
C  , = = 18 (Maximum value)
7 7  7
D (0, 2) = 8
The maximum value of Z is attained at C (13/5, 3/5) or at D (5/7, 24/7)
 The optimal solution is X1 = 13/5, X2 = 3/5, or x1 = 5/7, x2 = 24/7.
Example 3.6: Use graphical method to solve the LPP.
Subject to, 5X1 + X2  10
X1 + X2  6
X1 + 4X2  12
X1, X2  0.
Solution:

A (0, 10) 20
B (1, 5) 13 (Minimum value)
C (4, 2) 16
D (12, 0) 36
Self-Instructional
56 Material
Since the minimum value is attained at B (1, 5) the optimum solution is Graphical Analysis
of LPP
X1 = 1, X2 = 5.
Note: In the above problem if the objective function is maximization, then
the solution is unbounded, as maximum value of Z occurs at infinity.
NOTES
There are some linear programming problems which may have,
(i) a unique optimal solution
(ii) an infinite number of optimal solutions
(iii) an unbounded solution
(iv) no solution.
The following examples will illustrate these cases.
Example 3.7: Solve the LPP by graphical method.
Subject to, 5X1 + 2X2  1,000
3X1 + 2X2  900
X1 + 2X2  500
and, X1, X2  0
Solution:
The solution space is given by the feasible region OABC.

O (0, 0) 0
A (200, 0) 20,000 (Max value of Z)
B (125, 187.5) 20,000
C (0, 250) 10,000
Self-Instructional
Material 57
Graphical Analysis  The maximum value of Z occurs at two vertices A and B.
of LPP
Since there are infinite number of points on the line joining A and B it gives
the same maximum value of Z.
NOTES Thus, there are infinite number of optimal solutions for the LPP.
Example 3.8: Solve the following LPP.
Max Z = 3X1 + 2X2
Subject to, X1 + X2  1
X1 + X2  3
X1, X2  0
Solution: The solution space is unbounded. The value of the objective function
at the vertices A and B are Z (A) = 6, Z (B) = 6. But there exist points in the
convex region for which the value of the objective function is more than 8.
In fact, the maximum value of Z occurs at infinity. Hence, the problem has
an unbounded solution.
No feasible solution
When there is no feasible region formed by the constraints in conjunction
with non-negativity conditions, then no solution to the LPP exists.
Example 3.9: Solve the following LPP.
Max Z = X1 + X2
Subject to the constraints, X1 + X2  1
–3X1 + X2  3
X1, X2  0
Solution: There being no point (X1, X2) common to both the shaded regions,
we cannot find a feasible region for this problem. So the problem cannot be
solved. Hence, the problem has no solution.
Self-Instructional
58 Material
Graphical Analysis
of LPP
NOTES
General formulation of LPP

The general formulation of the LPP can be stated as follows:
Maximize or Minimize
Z = C1X1 + C2X2 + ... + CnXn ...(1)
Subject to m constraints
 a11 X1  a12 X 2  ...  a1 j X j  ...  a1n X n (  ) b1 
 a X  a X  ...  a X  ...  a X (  ) b2 
 21 1 22 2 2j j 2n n 
 
 a X  a X  ...  a X  ...  a X (  ) bi 
...(2)
 i1 1 i 2 2 1j j in n
 
 am1 X1  am 2 X 2  ...  amj X j  ...  amn X n (  ) bm 

In order to find the values of n decision variables X1 X2… Xn to maximize
or minimize the objective function and the non-negativity restrictions
X1  0, X2  0... Xn  0 …(3)
Matrix form of LPP
The linear programming problem can be expressed in the matrix form as follows:
Maximize or Minimize Z = CX

 
Subject to AX  b

 
X  0.
 x1   b1 
   
where, x =  x2  , b=  b2  , C = (C1 C2 ... Cn)
     
   
 xn   bm 
 a11 a12  a1n 
 
and, A=  a21 a22  a2n  .
  
 
 am1 am 2  amn 
Self-Instructional
Material 59
Graphical Analysis
of LPP 3.3 CANONICAL OR STANDARD FORMS OF LPP
The general LPP can be classified as canonical or standard forms.

NOTES In standard form, irrespective of the objective function, namely, maximize
or minimize, all the constraints are expressed as equations. Moreover RHS of
each constraint and all variables are non-negative.
Characteristics of the Standard Form
(i) The objective function is of maximization type.
(ii) All constraints are expressed as equations.
(iii) Right hand side of each constraint is non-negative.
(iv) All variables are non-negative.
In canonical form, if the objective function is of maximization, all the
constraints other than non-negative conditions are ‘’ type. If the objective
function is of minimization, all the constraints other than non-negative condition
are ‘’ type.
Characteristics of the Canonical Form
(i) The objective function is of maximization type.
(ii) All constraints are of () type.
(iii) All variables Xi are non-negative.
Note:
(i) Minimization of a function Z is equivalent to maximization of the negative
expression of this function, i.e., Min Z = – Max (– Z)
(ii) An inequality in one direction can be converted into an inequality in the
opposite direction by multiplying both sides by (–1).
(iii) Suppose we have the constraint equation,
a11 x1 + a12 x2 + … + a1n xn = b1
This equation can be replaced by two weak inequalities in opposite
directions,
a11 x1 + a12 x2+ … + a1n xn  b1
and, a11 x1 + a12 x2 + … + a1n xn  b1
(iv) If a variable is unrestricted in sign, then it can be expressed as a difference
of two non-negative variables, i.e., if X1 is unrestricted in sign, then X1 =
X1 – X1, where X1, X1 are  0.
(v) In standard form, all the constraints are expressed in equation, which is
possible by introducing some additional variables called ‘slack variables’ and
‘surplus variables’ so that a system of simultaneous linear equations is
Self-Instructional obtained. The necessary transformation will be made to ensure that bi  0.
60 Material
Graphical Analysis
of LPP
Check Your Progress
1. Define feasible region and feasible solution.
2. What do you mean by optimal solution? NOTES
3. What is the difference between feasible solution and basic feasible solution?
4. What are the two forms of LPP?
5. What do you mean by standard form of LPP?

QUESTIONS
1. A region in which all the constraints are satisfied simultaneously is called a

feasible region. While any solution to a LPP which satisfies the non-negativity
restrictions of the LPP is called its feasible solution.
2. Any feasible solution which optimizes (minimizes or maximizes) the objective
function is called its optimal solution.
3. The solution of m basic variables when each of the (n-m) non-basic variable
is set to zero is called basic solution. A basic solution in which all the basic
variables are P 0 is called a basic feasible solution.
4. The two forms of LPP are (i) Standard form and (ii) Canonical form.
5. In standard form, irrespective of the objective function namely maximize or
minimize, all the constraints are expressed as equations, also right hand side
constants are non-negative, i.e., all the variables are non-negative.
3.5 SUMMARY
 Simple linear programming problems with two decision variables can be

easily solved by graphical method.
 As the optimal values occur at the corner points of the feasible region, it is
enough to calculate the value of the objective function of the corner points
of the feasible region and select the one that gives the optimal solution.
 In the case of maximization problem, the optimal point corresponds to the
corner point at which the objective function has a maximum value, and in
the case of minimization, the optional solution is the corner point which
gives the objective function the minimum value for the objective function.
 Any feasible solution, which optimizes (minimizes or maximizes) the objective
function of the LPP is called its optimum solution.
 Basic feasible solutions are of two types: non-degenerate and degenerate.
Self-Instructional
Material 61
Graphical Analysis  If the value of the objective function can be increased or decreased
of LPP
indefinitely, such solutions are called unbounded solutions.
NOTES
3.6 KEY WORDS
 Linear Programming Problem: It consists of a linear function to be

maximized or minimized subject to certain constraints in the form of linear
equations or inequalities.
 Optimal Solution: It refers to the feasible solution with the largest objective
function value (for a maximization problem).
 Unbounded Solution: It refers to a situation where objective function is
infinite. A LPP is said to have unbounded solution if its solution can be made
infinitely large without violating any of its constraints in the problem.

EXERCISES
Short Answer Questions

1. What do you mean by a general LPP?
2. What do you mean by canonical form of a LPP?
3. Define unbound solution.
4. Mention the characteristics of the standard form.
1. Discuss the procedure for solving LPP by graphical method.
2. Give a matrix form of representing a general LPP.
3. State the characteristics of canonical form and write the canonical form of
LPP in matrix form.

House.
Self-Instructional
62 Material
Simplex Method
UNIT 4 SIMPLEX METHOD

Structure NOTES
4.0 Introduction
4.1 Objectives
4.2 Solution of LPP – Simplex Method
4.3 The Simplex Algorithm
4.5 Summary
4.6 Key Words
4.0 INTRODUCTION
Simplex method is used to find the optimal solution to multivariable problems. The
method is basically an algorithm which is used by an examinee to examine corner
points in a methodical fashion until he/she arrive at the best solution. In this method,
slack variables are added to change the constraints into equations and write all
variables to the left of the equal sign and constants to the right.
In this unit, you will study in detail about the simplex method in LPP and you
will also get to know the steps involved in the simplex algorithm. In addition to this,
you will also learn about Charne’s BIG M method and get to know the way to
solve an LPP using the same. The two-phase simplex method used to solve a LPP
is also explained in the unit.
It is to be noted that the standard from of LPP has been discussed in detail
in Unit 3 and the fundamental theorem of LPP has been discussed in Unit 2.
4.1 OBJECTIVES

 Discuss the simplex method in LPP
 Comprehend the steps of simplex algorithm
 Find the solution of any LPP by using simplex algorithm
 Understand solving an LPP using the Charne’s BIG M method
 Know the two-phase simplex method to solve a given LPP
Self-Instructional
Material 63
Simplex Method
4.2 SOLUTION OF LPP – SIMPLEX METHOD
Simplex method is an iterative procedure for solving LPP in a finite number of

NOTES steps. It provides an algorithm, which consists of moving from one vertex of the
region of feasible solution to another in such a manner that the value of the
objective function at the succeeding vertex is less or more, as the case may be,
than at the previous vertex. This procedure is repeated and since the number of
vertices is finite, the method leads to an optimal vertex in a finite number of steps
or indicates the existence of unbounded solution.
Definition
(i) Let XB be a basic feasible solution to the LPP.
Max Z = CX
Subject to, AX = b
and, X 0. Such that it satisfies XB = B–1b
where B is the basis matrix formed by the column of basic variables.
The vector CBj = (CB1, CB2  CBm) where CBj are components of C
associated with the basic variables, called the cost vector associated with
the basic feasible solution XB.
(ii) Let XB be a basic feasible solution to the LPP.
Max Z = CX where,
AX = b and X0.
Let CB be the cost vector corresponding to XB. For each column vector aj in
A1, which is not a column vector of B, let
m
aj =  aij bi
i 1
m
Then the number Zj = CBi aij
i 1
is called the evaluation corresponding to aj and the number (Zj – Cj) is

called the net evaluation corresponding to j.
Simplex algorithm
The simplex algorithm is the classical method to solve the optimization problem of
linear programming.
For the solution of any LPP by simplex algorithm, the existence of an initial basic
feasible solution is always assumed. The steps for the computation of an optimum
solution are as follows:
Step 1 Check whether the objective function of the given LPP is to be maximized
or minimized. If it is to be minimized then we convert it into a problem of
maximization by
Self-Instructional Min Z = –Max (–Z)
64 Material
Step 2 Check whether all bi (i = 1, 2  m) are positive. If any bi is negative Simplex Method
then multiply the inequation of the constraint by –1 so as to get all bi to be positive.

Step 3 Express the problem in standard form by introducing slack/surplus variables
to convert the inequality constraints into equations.
NOTES
Step 4 Obtain an initial basic feasible solution to the problem in the form XB =
B–1b and put it in the first column of the simplex table. Form the initial simplex
table as given below:
Step 5 Compute the net evaluations Zj – Cj by using the relation Zj – Cj = CB

(aj – cj)
Examine the sign of Zj – Cj
(i) If all Zj – Cj  0, then the initial basic feasible solution XB is an optimum
basic feasible solution.
(ii) If at least one Zj – Cj < 0, then proceed to next step as the solution is not
optimal.
Step 6 (To find the entering variable, i.e., key column)
If there are more than one negative Zj – Cj, choose the most negative of them.
Let it be Zr – Cr for some j = r. This gives the entering variable Xr and is
indicated by an arrow at the bottom of the rth column. If there are more than one
variables having the same most negative Zj – Cj then, any one of them can be selected
arbitrarily as the entering variable.
(i) If all air  0 (i = 1, 2 ... m) then there is an unbounded solution to the
given problem.
(ii) If at least one air > 0 (i = 1, 2 ... m) then the corresponding vector Xr
enters the basis.
Step 7 (To find the leaving variable or key row)
Compute the ratio (XBi /air, air > 0)
If the minimum of these ratios be XBi /akr, then choose the variable Xk to leave
the basis called the key row and the element at the intersection of key row and
key column is called the key element.
Step 8 Form a new basis by dropping the leaving variable and introducing the
entering variable along with the associated value under CB column. Convert the
leading element to unity by dividing the key equation by the key element and all
other elements in its column to zero by using Gauss Elimination on the formula
 Product of elements in key row and column 
New element = old element –   Self-Instructional
 key element 
Material 65
Simplex Method Step 9 Go to step (5) and repeat the procedure until either an optimum solution
is obtained or there is an indication of unbounded solution.
Example 4.1 Use simplex method to solve the LPP.
NOTES Max Z = 3X1 + 2X2
Subject to, X1 + X2  4
X1 – X2  2
X1, X2  0
Solution By introducing the slack variables S1, S2, convert the problem in
standard form.
Max Z = 3X1 + 2X2 + 0S1 + 0S2
Subject to, X1 + X2 + S1 = 4
X1 – X2 + S2 = 2
X1, X2, S1, S2  0
Writing is Martix form AX = 6
 X1 
 X1 X 2 S1 S 2   
 1 X  4
 1 1 0   2  =  
 S1   2 
 1 1 0 1  
 S2 
An initial basic feasible solution is given by
XB = B–1b,
where, B = I2, XB = (S1 S2).
i.e., (S1 S2) = I2 (4, 2) = (4, 2)
Initial simplex table
Zj = CB aj
 0  1
Z1 – c1 = CB a1 – c1 =      3 = – 3
 0  1
0
Z2 – c2 = CB a2 – c2 =   (1  1)  2 = – 2
0
0
Z3 – c3 = CB a3 – c3 =  0  (1 0)  0 = – 0
 
0
Z4 – c4 = CB a4 – c4 =   (0 1)  0 = – 0.
0
Self-Instructional
66 Material
Simplex Method
NOTES
Since there are some Zj – Cj < 0, the current basic feasible solution is not
optimum.
Since Z1 – C1= –3 is the most negative, the corresponding non-basic
variable X1 enters the basis.
The column corresponding to this X1 is called the key column.
X Bi
To find the ratio = Min , X ir 0
X ir
= Min  ,  = 2, which corresponds to S2.

4 2
 1
1
 The leaving variable is the basic variable S2. This row is called the key
row. Convert the leading element X21 to units and all other elements in its column
n i.e. (X1) to zero by using the formula or Gauss Elimination:
 Product of elements in key row and key column 
New element = Old element –  
 key element 
To apply this formula, first we find the ratio, namely
the element to be zero 1
= =1
key element 1
Apply this ratio for the number of elements that are converted in the key
row. Multiply this ratio by key row elements as shown below.
1×2
1×1
1 × –1
1×0
1×1
Now subtract this element from the old element. The element to be
converted into zero is called the old element row. Finally we have,
4–1×2 = 2
1–1×1 = 0
1–1×–1 = 2
1–1×0 = 1
0–1×1 = –1
Self-Instructional
Material 67
Simplex Method  The improved basic feasible solution is given in the following simplex table
First iteration
NOTES
Since Z2 – C2 is most negative, X2 enters the basis.

X 
To find Min  B , X i 2  0 
 X i2 
2 2 
Min  ,  = 1. ( negative or zero value are not considered)
 2 1 
This gives the outgoing variables. Convert the leading element into one. This is done
by dividing all the elements in the key row by 2. The remaining elements should be
made zero using the formula as shown below.
1
– is the common ratio. Put this ratio 5 times and multiply each ratio by key
2
row elements.
1
 2
2
1
 0
2
1
 2
2
1
 2
2
1
  1
2
Subtract this from the old element. All the row elements that are converted
into zero, are called the old element.
 1 
2    2 = 3
 2 
 1 
1     0 = 1
 2 
 1 
1     2  = 0
 2 
Self-Instructional
68 Material
Simplex Method
 1  1
0     1 =
 2  2
 1  1
1      1 =
 2  2
NOTES
Second iteration
Since all Zj – Cj0, the solution is optimum. The optimal solution is Max Z =
11, X1 = 3, and X2 = 1.
Example 4.2 Solve the LPP
Max Z = 3X1 + 2X2
4X1 + X2  8
4X1 – X2  8
X1, X2  0
Solution Convert the inequality of the constraint into an equation by adding
slack variables S1, S2, S3,.
Max Z = 3X1 + 2X2 + 0S1 + 0S2 + 0S3
Subject to, 4X1 + 3X2 + S1 = 12
4X1 + X2 + S2 = 8
4X1 – X2 + S3 = 8
X1, X2, S1, S2, S3  0
 X1 
 X1 X2 S1 S2 S3   
 4 X2 12 
 3 1 0 0     
 S1   8
 4 1 0 1 0    
  S2 8
 4 1 0 0 1    
 S3 
Self-Instructional
Material 69
Simplex Method Initial table
NOTES
xB
 Z1 – C1 is most negative, X1 enters the basis. And the min , xil 0 =
xil
min (3, 2, 2) = 2 gives S3 as the leaving variable.
Convert the leading element into 1, by dividing key row element by 4 and the
remaining elements into 0.
First iteration
4 4
8 8 =0 12  8 =4
4 4
4 4
4  4 =0 4  4 =0
4 4
4 4
1  1 = 2 3  1 = 4
4 4
4 4
0  0 =0 1  0 =1
4 4
4 4
1  0 =1 0  0 =0
4 4
11 4
0  1 = –1 0  1 = –1
4 4
Self-Instructional
70 Material
11 Simplex Method
Since Z2 – C2 = – is the most negative, x2 enters the basis.
4
 xB 
To find the outgoing variable, find Min  , xi 2  0 
 xi 2  NOTES
4 0 
Min  , ,   = 0
4 2 
First iteration
Therefore, S2 leaves the basis. Convert the leading element into 1 by dividing the
key row elements by 2 and make the remaining elements in that column as zero using
the formula.
 Product of elements in key row and key column 
New element = old element   
 key element 
Second iteration
Since Z5 – C5 = –5/8 is most negative, S3 enters the basis and
XB 4 2
Min , Si 3 = Min , 1 = 4.
S13 1 1/18
Therefore, S1 leaves the basis. Convert the leading element into one and remaining
elements as zero.
Third iteration
Since all Zj – Cj  0, the solution is optimum and it is given by X1 = 3/2, X2 =

2 and Max Z = 17/2.
Self-Instructional
Material 71
Simplex Method Example 4.3 Using simplex method solve the LPP.
Max Z = X1 + X2 + 3X3
Subject to, 3X1 + 2X2 + X3  3
NOTES 2X1 + X2 + 2X3  2
X1, X2, X3  0
Solution Rewrite the inequality of the constraints into an equation by adding
slack variables.
Max Z = X1 + X2 + 3X3 + 0S1 + 0S2
Subject to, 3X1 + 2X2 + X3 + S1  3
2X1 + X2 + 2X3 + S 2 2
Initial basic feasible solution is,
X1 = X2 = X3 = 0
S1 = 3, S2 = 2 and Z = 0
 X1 X2 X3 S1 S2 
 3 2 1 1 0 

 2 1 2 0 1 
 
 1 1 3 0 0 
Since Z3 – C3 = –3 is the most negative, the variable X3 enters the basis. The
column corresponding to X3 is called the key column.
  xB  3 2  
 To determine the key row or leaving variable, find Min  , xi 3  0  Min  = 3, = 1 = 1
 x
 i3   1 2  
Therefore, the leaving variable is the basic variable S2, the row is called the key row
and the intersection element 2 is called the key element.
Convert this element into one by dividing each element in the key row by 2 and
the remaining elements in that key column as zero using the formula
Product of elements in key row and key column
New element = old element
key element
Self-Instructional
72 Material
First iteration Simplex Method
NOTES
Since all Zj – Cj0, the solution is optimum and it is given by X1 = 0, X2

= 0, X3 = 1, Max Z = 3.
Example 4.4 Use simplex method to solve the LPP.
Min Z = X2 – 3X3 + 2X5
Subject to, 3X2 – X3 + 2X5  7
– 2X2 + 4X3  12
– 4X2 + 3X3 + 8X5  10
X2, X3, X5  0
Solution Since the given objective function is of minimization we shall convert
it into maximization using Min Z = –Max(–Z) = –Max Z*
Max Z* = – X2 + 3X3 – 2X5
Subject to, 3X2 – X3 + 2X5  7
– 2X2 + 4X3  12
– 4X2 + 3X3 + 8X5  10
We rewrite the inequality of the constraints into an equation by adding slack
variables S1, S2, S3 and the standard form of LPP becomes.
Max Z = – X2 + 3X3 – 2X5 + 0S1 + 0S2 + 0S3
Subject to 3X2 – X3 + 2X5 + S1  7
– 2X2 + 4X3 + S2  12
– 4X2 + 3X3 + 8X5 + S3  10
X2, X3, X5, S1, S2, S3  0
 The initial basic feasible solution is given by S1 = 7, S2 = 12, S3 = 10. (X2 =
X3 = X5 = 0)
Self-Instructional
Material 73
Simplex Method Initial table
NOTES
Since Z2 – C2 = –3 < 0, the solution is not optimum.

The incoming variable is X3 (key column) and the outgoing variable (key
row) is given by,
XB  12 10 
Min X i3 0 = Min  , ,  = 3.
X i3  4 3
Hence, S2 leaves the basis.

First iteration
Since Z1 – C1 < 0, the solution is not optimum. Improve the solution by

allowing the variable X2 to enter into the basis and the variable S1 to leave the
basis.
Second iteration
Self-Instructional
74 Material
Since all Zj – Cj0, the solution is optimum. Simplex Method
 The optimal solution is given by Max zx = 11

X2 = 4, X3 = 5, X5 = 0
 Min Z = – Max (–Z) = – 11 NOTES
 Min Z = –11, X2 = 4, X3 = 5, X5 = 0.
Example 4.5 Solve the following LPP using simplex method.
Max Z = 15X1 + 6X2 + 9X3 + 2X4
Subject to, 2X1 + X2 + 5X3 + 6X4  20
3X1 + X2 + 3X3 + 25X4  24
7X1 + X4  70
X1, X2, X3, X4  0
Solution Rewriting the inequality of the constraint into an equation by adding
slack variables S1, S2 and S3 the standard form of LPP becomes.
Max Z = 15X1 + 6X2 + 9X3 + 2X4 + 0S1 + 0S2 + 0S3
Subject to,2X1 + X2 + 5X3 + 6X4 + S1  20
3X1 + X2 + 3X3 + 25X4 + S2  24
7X1 + X4 + S3  70
X1, X2, X3, X4, S1, S2, S3  0
The initial basic feasible solution is S1 = 20, S2 = 24, S3 = 70 (X1 = X2
= X3 = X4 = 0 non-basic)
The initial simplex table is given by
 s some of Zj – Cj  0 the current basic feasible solution is not optimum.

Z1 – C1 = –15 is the most negative value and hence X1 enters the basis and the
variable S2 leaves the basis.
Self-Instructional
Material 75
Simplex Method First iteration
NOTES
Since Z2 – C2 = –1 < 0 the solution is not optimal therefore, X2 enters the basis
and the basic variable S1 leaves the basis.
Second iteration
Since all Zj – Cj  0, the solution is optimal and is given by,

Max Z =132, X1 = 4, X2 = 12, X3 = 0, X4 = 0.
Example 4.6
Max Z = 3X1 + 2X2 + 5X3
Subject to X1 + 2X2 + X3  430
3X1 + 2X3  260
X1 + 4X2  420
X1, X2, X3  0
Solution Rewrite the constraint into equation by adding slack variables S1,
S2, S3. The standard form of LPP becomes
Max Z = 3X1 + 2X2+ 5X3 + 0S1 + 0S2 + 0S3
Subject to, X1 + 2X2 + X3 + S1  430
3X1 + 2X3 + S2  460
Self-Instructional
76 Material
X1 + 4X2 + S3  420 Simplex Method
X1, X2, X3, S1, S2, S3  0.

The initial basic feasible solution is,
S1 = 430, S2 = 460, S3 = 420 (X1 = X2 = X3 = 0) NOTES
Initial table
Since some of Zj – Cj0, the current basic feasible solution is not optimum.
Since Z3 – C3 = –5 is the most negative, the variable X3 enters the basis. To
find the variable leaving the basis find,
XB 430 460
Min , X i3 0 = Min , , = 230.
X i3 1 2
 the variable S2 leaves the basis.
First iteration
Since Z2 – C2 = –2 is negative, the current basic feasible solution is not

optimum. Therefore, the variable X2 enters the basis and the variable S1 leaves
the basis.
Self-Instructional
Material 77
Simplex Method Second iteration
NOTES
Since all Zj – Cj0, the solution is optimum and is given by X1 = 0,

X2 = 100, X3 = 230 and Max Z = 1350.
4.3 THE SIMPLEX ALGORITHM

In this section, we introduce a new type of variable called the artificial variable.
These variables are fictitious and cannot have any physical meaning. The artificial
variable technique is merely a device to get the starting basic feasible solution,
so that simplex procedure may be adopted as usual until the optimal solution is
obtained. To solve such LPP there are two methods:
(i) The Charne’s Big M method
(ii) The Two-phase Simplex method
The following steps are involved in solving an LPP using the Big M method.
Step 1 Express the problem in the standard form.
Step 2 Add non-negative artificial variables to the left side of each of the equations
corresponding to constraints of the type  or =. However, addition of these
artificial variables causes violation of the corresponding constraints. Therefore,
we would like to get rid of these variables and not allow them to appear in the
final solution. This is achieved by assigning a very large penalty (–M for
maximization and M for minimization) in the objective function.
Step 3 Solve the modified LPP by simplex method, until any one of the three
cases may arise.
1. If no artificial variable appears in the basis and the optimality conditions
are satisfied, then the current solution is an optimal basic feasible solution.
2. If at least one artificial variable in the basis at zero level and the optimality
condition is satisfied, then the current solution is an optimal basic feasible
solution (though degenerated).
3. If at least one artificial variable appears in the basis at positive level and the
optimality condition is satisfied, then the original problem has no feasible
Self-Instructional
solution. The solution satisfies the constraints but does not optimize the
78 Material
objective function, since it contains a very large penalty M and is called Simplex Method
pseudo optimal solution.

Note: While applying simplex method, whenever an artificial variable happens to
leave the basis, we drop that artificial variable and omit all the entries
NOTES
corresponding to its column from the simplex table.
Example 4.6 Use penalty method to
Maximize Z = 3x1 + 2x2
2x1 + x2  2
3x1 + 4x2  12
x1, x2  0
Solution By introducing slack variable S1  0, surplus variable S2  0 and artificial
variable A1  0, the given LPP can be reformulated as:
Maximize Z = 3x1 + 2x2 + 0S1 + 0S2 – MA1
Subject to, 2x1 + x2 + S1 = 2
3x1 + 4x2 – S2 + A1 = 12
The starting feasible solution is S1 = 2, A1 = 12.
Initial table
Since some of the Zj – Cj  0, the current feasible solution is not optimum.

Choose the most negative Zj – Cj = – 4M – 2.
 x2 variable enters the basis, and the basic variable S1 leaves the basis.
First iteration
Since all Zj – Cj and an artificial variable appears in the basis at positive

level, the given LPP does not possess any feasible solution. But the LPP possesses
a pseudo optimal solution. Self-Instructional
Material 79
Simplex Method Example 4.7 Solve the LPP.
Minimize Z = 4x1 + x2
Subject to, 3x1 + x2 = 3
NOTES 4x1 + 3x2  6
x1 + 2x2  4
x1, x2  0
Solution Since the objective function is minimization, we covert it into
maximization using,
Min Z = – Max (– z) = –Max z* ( Z* = – Z)
Maximize z* = – 4x1 – x2
Subject to, 3x1 + x2 = 3
4x1 + 3x2  6
x1 + 2x2  4
x1, x2  0
Convert the given LPP into standard form by adding artificial variables A1,
A2, surplus variable S1 and slack variable S2 to get the initial basic feasible solution.
Minimize Z* = – 4x1 – x2 + 0S1 + 0S2 – MA1 – MA2
Subject to, 3x1 + x2 + A1 = 3
4x1 + 3x2 – S1 + A2  6
x1 + 2x2 + S2 = 4
x1x2, S1S2, A1A2  0
The starting feasible solution is A1 = 3, A2 = 6, S2 = 4.
Initial solution
Since some of the Zj – Cj  0, the current feasible solution is not optimum. As

Z1 – C1 is most negative, x1 enters the basis and the basic variable A1 leaves
the basis.
Self-Instructional
80 Material
NOTES
Since Z1 – C1 is negative, the current feasible solution is not optimum. Therefore,

x1 variable enters the basis and the artificial variable A2 leaves the basis.
Second iteration
Since Z4 – C4 is most negative, S1 enters the basis and the artificial variable A1
leaves the basis.
Third iteration
Since Z4 – C4 is most negative, S2 enters the basis and S1 leaves the basis.
Self-Instructional
Material 81
Simplex Method Fourth iteration
NOTES
Since all Zj – Cj  0, the solution is optimum and is given by x1 = 3/5, x2

= 6/5, and Max Z = – 18/5
 Min Z = – Max ( – Z) = 18/5.
Example 4.8 Solve by Big M method.
Maximize Z = x1 + 2x2 + 3x3 – x4
Subject to, x1 + 2x2 + 3x3 = 15
2x1 + x2 + 5x3 = 20
x1 + 2x2 + x3 + x4 = 10
Solution Since the constraints are equations, introduce artificial variables A1,
A2  0. The reformulated problem is given as follows.
Maximize Z = x1 + 2x2 + 3x3 – x4 – MA1 – MA2
Subject to,x1 + 2x2 + 3x3 + A1 = 15
2x1 + x2 + 5x3 + A2 = 20
x1 + 2x2 + x3 + x4 = 10
Initial solution is given by A1 = 15, A2 = 20 and x4 = 10.
Since Z3 – C3 is most negative, x3 enters the basis and the basic variable
A2 leaves the basis.
Self-Instructional
82 Material
NOTES
Since Z2 – C2 is most negative, x2 enters the basis and the basic variable
A1 leaves the basis.
Second iteration
Since Z1 – C1 = –6/7 is negative, the current feasible solution is not

optimum. herefore, x1 enters the basis and the basic variable x4 leaves the basis.
Third iteration
Self-Instructional
Material 83
Simplex Method Since all Zj – Cj  0, the solution is optimum and is given by x1 = x2 = x3
= 15/6 = 5/2, and Max Z = 15.
Example 4.9 Use penalty method to solve the following LPP:
NOTES Minimize Z = 5x + 3y
Subject to, 2x + 4y  12
2x + 2y = 10
5x + 2y  10
x, y  0
Solution First we convert the objective function minimization to maximization
using
Min Z = – Max(– Z).
Rewrite the given LPP into standard form by adding slack variable
S1  0, surplus variable S2  0 and the artificial variables A1, A2  0.
Maximize Z = – 5x – 3y + 0S1 + 0S2 – MA1 – MA2
Subject to, 2x + 4y + S1 = 12
2x + 2y + A1 = 10
5x + 2y – S2 + A2 = 10
x, y, S1, S2, A1, A2  0
Initial feasible solution is given by S1 = 12, A1 = 10 and A2 = 10
Since all Zj – Cj 0 (see Table 4.1) and no artificial variable is in the basis,
the solution is optimum and is given by:
x = 4, y = 1, Max Z = – 23
Min Z = – Max (– Z) = 23
Table 4.1
Self-Instructional
84 Material
Simplex Method
NOTES

The two-phase simplex method is another method to solve a given LPP involving
some artificial variables. The solution is obtained in two phases.
Phase I
In this phase, we construct an auxiliary LPP leading to a final simplex table
containing a basic feasible solution to the original problem.
Step 1 Assign a cost –1 to each artificial variable and a cost 0 to all other
variables and get a new objective function Z* = – A1 – A2 – A3... where Ai are
artificial variables.
Step 2 Write down the auxiliary LPP in which the new objective function is to
be maximized, subject to the given set of constraints.
Step 3 Solve the auxiliary LPP by simplex method until either of the following
three cases arise:
(i) Max Z* < 0 and at least one artificial variable appears in the optimum
basis at positive level.
(ii) Max Z* = 0 and at least one artificial variable appears in the optimum
basis at zero level.
(iii) Max Z* = 0 and no artificial variable appears in the optimum basis.
In case (i), given LPP does not possess any feasible solution, whereas in
cases (ii) and (iii) we go to phase II.
Self-Instructional
Material 85
Simplex Method Phase II
Use the optimum basic feasible solution of phase I as a starting solution for the
original LPP. Assign the actual costs to the variable in the objective function and
NOTES a zero cost to every artificial variable in the basis at zero level. Delete the artificial
variable column that is eliminated from the basis in phase I from the table. Apply
simplex method to the modified simplex table obtained at the end of phase I till
an optimum basic feasible solution is obtained or till there is an indication of
unbounded solution.
Example 4.10 Use two-phase simplex method to solve,
Subject to, 2x1 + x2  1
x1  4x2  6
x1, x2 0
Solution We convert the given problem into a standard form by adding slack,
surplus and artificial variables. We form the auxiliary LPP by assigning the cost
–1 to the artificial variable and 0 to all the other variables.
Phase I
Maximize Z* 0x1 + 0x2 + 0S1 + 0S2 – 1. A1
Subject to, 2x1 + x2 + S1 = 1
x1 + 4x2 – S2 + A1 = 6
Initial basic feasible solution is given by S11, A1 = 6.
Since all Zj – Cj  0, an optimum feasible solution to the auxiliary LPP is

obtained. But as M ax Z* < 0, and an artificial variable A1 is in the basis at a
positive level, the original LPP does not possess any feasible solution.
Self-Instructional
86 Material
Example 4.11 Use two-phase simplex method to solve, Simplex Method
Minimize Z = x1 – 2x2 – 3x3

Subject to,– 2x1 + x2 + 3x3 =2
2x1 + 3x2 + 4x3 = 1 NOTES
x1, x2, x3  0
Solution Convert the objective function into maximization using,
Min Z = – Max ( – Z)
Max Z = – x1 + 2x2 + 3x3
Introducing the artificial variables A1, A2  0, the constraints of the given
problem become,
– 2x1 + x2 + 3x3 + A1 = 2
2x1 + 3x2 + 4x3 + A2 = 1
x1, x2, x3, A1, A2  0
Phase I
Assigning a cost –1 to the artificial variables A1 and A2, and cost 0 to other
variables, the objective function of the auxiliary LPP is,
Maximize Z* = 0x1 + 0x2 + 0x3 – 1 . A1 – 1 . A2, subject to the given
constraints.
Since all Zj – Cj  0, an optimum basic feasible solution to the auxiliary

LPP has been attained. But since max Z* is negative and the artificial variable
A1 appears in the basis at positive level, the original problem does not possess
any feasible solution.
Self-Instructional
Material 87
Simplex Method Example 4.12
Maximize Z = 5x1 – 4x2 + 3x3
Subject to,2x1 + x2 – 6x3 = 20
NOTES 6x1 + 5x2 + 10x3  76
8x1 – 3x2 + 6x3  50
x1, x2, x3  0
Solution Introducing slack variables S1, S2  0 and an artificial variable A1  0 in
the constraints of the given LPP, the problem is reformulated in the standard form.
Initial basic feasible solution is given by A1 = 20, S1 = 76 and S2 = 50.
Phase I
Assigning a cost –1 to the artificial variable A1 and cost 0 to other variables, the
objective function of the auxiliary LPP is,
Maximize Z* = 0x1 + 0x2 + 0x3 + 0S1 + 0S2 – 1 . A1
Subject to,2x1 + x2 – 6x3 + A1 = 20
6x1 + 5x2 + 10x3 + S1 = 76
8x1 – 3x2 + 6x3 + S2 = 50
x1, x2, x3, S1, S2, A1  0
Self-Instructional
88 Material
Since all Zj Cj 0, an optimum solution to the auxiliary LPP has been Simplex Method
obtained. Also Max Z* 0 with no artificial variables in the basis. We go to

phase II.
Phase II NOTES
Consider the final simplex table of phase I. Consider the actual cost associated
with the original variables. Delete the artificial variable A1 column from the table
as it is eliminated in phase I.
Since all Zj – Cj 0, an optimum basic feasible solution has been reached.
Hence, an optimum feasible solution to the given LPP is x1 = 55/7, x2 = 30/7,
x3 = 0 and Max Z = 155/7.
Example 4.13 Solve by two-phase simplex method
Maximize Z = – 4x1 – 3x2 – 9x3
Subject to 2x1 + 4x2 + 6x3  15
6x1 + x2 + 6x3  12
x1, x2, x3  0
Solution Convert the given LPP into standard form by introducing surplus
variables S1, S2 and artificial variables A1, A2.The initial solution is given by
A1 = 15, A2 = 12.
Phase I
Construct an auxiliary LPP by assigning a cost 0 to all the variables and –1 to
each artificial variable subject to the given set of constraints, and it is given by
Max Z* = 0x1 + 0x2 + 0x3 + 0S1 + 0S2 – 1A1 – 1A2
Subject to, 2x1 + 4x2 + 6x3 + S1 + A1 = 15
6x1 + x2 + 6x3 – S2 + A2 = 12
Self-Instructional
Material 89
Simplex Method
NOTES
Since all Zj – Cj  0, the current basic feasible solution is optimal. Since Max
Z* = 0 and no artificial variable appears in the basis, we go to phase II.
Phase II
Consider the final simplex table of phase I; also consider the actual cost associated
with the original variables. Delete the artificial variables A1, A2 column from the
table as these variables are eliminated from the basis in phase I.
Since all Zj – Cj  0, the current basic feasible solution is optimal.

 The optimal solution is given by Max Z = – 15, x1 = 3/2, x2 = 3, x3 = 0.
Self-Instructional
90 Material
Simplex Method
Check Your Progress

1. What is simplex method?
2. Define simplex algorithm. NOTES
3. What is a two-phase simplex method?
4. What are the methods used to solve an LPP involving artificial variables?

QUESTIONS
1. Simplex method is an iterative procedure for solving LPP in a finite number

of steps. It provides an algorithm, which consists of moving from one vertex
of the region of feasible solution to another in such a manner that the value
of the objective function at the succeeding vertex is less or more, as the
case may be, than at the previous vertex.
2. The simplex algorithm is the classical method to solve the optimization problem
of linear programming.
3. The two-phase simplex method is another method to solve a given LPP
involving some artificial variables. The solution is obtained in two phases.
4. The two methods used to solve an LPP are:
(i) Big M method
(ii) Two-phase simplex method
4.5 SUMMARY
 Simplex method is an iterative procedure for solving LPP in a finite number

of steps. It provides an algorithm, which consists of moving from one vertex
of the region of feasible solution to another in such a manner that the value
of the objective function at the succeeding vertex is less or more, as the
case may be, than at the previous vertex.
 The simplex algorithm is the classical method to solve the optimization problem
of linear programming.
 There are two methods used to solve an LPP involving artificial variables,
namely:
(i) Big M method
(ii) Two-phase simplex method
 The two-phase simplex method is another method to solve a given LPP
involving some artificial variables.
Self-Instructional
Material 91
Simplex Method
4.6 KEY WORDS
 Simplex Method: It refers to an iterative procedure for solving LPP in a

NOTES finite number of steps.
 Charne’s Big M Method: It refers to surplus and artificial variables to
convert all inequalities into that form. The Big ‘M’ refers to a large number
associated with the artificial variables.
 Pseudo Optimal Solution: If any artificial variable is in the basis with
nonzero value at the optimal solution of the augmented problem, then the
original problem has no feasible solution. The solution satisfies the constraints
but does not optimize the objective function, since it contains a very large
penalty M and is called pseudo optimal solution.

EXERCISES

1. What is pseudo optimal solution?
2. What are the phases involved in two-phase simplex method?
1. Discuss the steps for the computation of an optimum solution using simplex
algorithm.
2. What are steps involved in solving an LPP using the Big M method? Discuss.

House.
Self-Instructional
92 Material
Duality in Linear
UNIT 5 DUALITY IN LINEAR Programming Problem
PROGRAMMING PROBLEM
NOTES
Structure
5.0 Introduction
5.1 Objectives
5.2 Concept of Duality and Sensitivity Analysis
5.2.2 Formulation of a Dual Problem
5.4 Summary
5.5 Key Words
5.0 INTRODUCTION
The concept of duality in linear programming states that every LPP (called the
primal) is associated with another LPP (called its dual). Either of the problem can
be considered as primal with the other as dual.
In this unit, you will study about the concept of duality, important results in
duality and sensitivity analysis.
5.1 OBJECTIVES
 Understand the concept of duality
 Discuss the important results in duality
 Examine sensitivity analysis as the optimal solution of LP
5.2 CONCEPT OF DUALITY AND SENSITIVITY

ANALYSIS

Two reasons are attributed to the importance of the concept of duality:
1. If a large number of constraints and a lesser number of variables constitute
the primal, then the procedure for computation can be minimized by
converting the primal into its dual and then finding its solution.
Self-Instructional
Material 93
Duality in Linear 2. While taking future decisions regarding the activities being programmed,
Programming Problem
the interpretation of the dual variables from the cost or economic point of
view proves extremely useful.
NOTES 5.2.2 Formulation of a Dual Problem

In the formulation of a dual problem, it should be first converted into its canonical
form. Formulation of a dual problem involves the following modifications:
(i) The objective function of minimization in the primal should be converted
into that of maximization in the dual and vice versa.
(ii) The number of constraints in the dual will be equal to the number of variables
in its primal and vice versa.
(iii) The right hand side constraints in the dual will be constituted by the cost
coefficients C1, C2 ... Cn in the objective function of the primal and vice
versa.
(iv) While forming the constraints for the dual, the transpose of the body matrix
of the primal problem should be considered.
(v) The variables in both problems should be positive, that is, there should be
no negative values.
(vi) If the variable in the primal is unrestricted in sign, then the corresponding
constraint in the dual will be an equation and vice versa.
Let the primal problem be:
Max Z = C1x1 + C2x2 + ... + Cnxn
Subject to: a11 x1 a12 x2  a1n xn b1
a21 x1 a22 x2  a2 n xn b2

am1 x1 am 2 x2  amn xn bm
x1 , x2  xn 0
Dual: The dual problem is defined as:
Min Z1 = b1w1 + b2w2 + ... + bmwm
Subject to: a11w1 a21w2  am1wm C1
a12 w1 a22 w2  am 2 wm C2

a1n w1 a2 n w2  amn wn Cn
w1 , w2  wm 0
Here, w1, w2, w3 ... wm are called dual variables.
Self-Instructional
94 Material
Example 5.1: Write the dual of the following primal LP problem. Duality in Linear
Programming Problem
Max Z = x1+ 2x2 + x3
Subject to: 2 x1 x2 x3 2
NOTES
2 x1 x2 5 x3 6
4 x1 x2 x3 6
x1 , x2 , x3 0
Solution: Since the problem is not in the canonical form, we interchange the
inequality of the second constraint.
Max Z = x1+ 2x2 + x3
Subject to: 2 x1 x2 x3 2
2 x1 x2 5 x3 6
4 x1 x2 x3 6
and x1 , x2 , x3 0
Dual: Let w1, w2, w3 be the dual variables.

Min Z1 = 2w1+ 6w2 + 6w3
Subject to: 2 w1 2 w2 4 w3 1
w1 w2 w3 2
w1 5w2 w3 1
w1 , w2 , w3 0.
Example 5.2: Find the dual of the following LPP.

Max Z = 3x1– x2 + x3
Subject to: 4 x1 x2 8
8 x1 x2 3x3 12
5 x1 6 x3 13
x1 , x2 , x3 0
Solution: Since the problem is not in the canonical form, we interchange the
inequality of the second constraint.
Max Z = 3x1– x2 + x3
Subject to: 4 x1 x2 0 x3 8
8 x1 x2 3 x3 12
5 x1 0 x2 6 x3 13
x1 , x2 , x3 0
Self-Instructional
Material 95
Duality in Linear Max Z = Cx
Programming Problem
Subject to: Ax B
x 0
NOTES
 x1   8 
   
C  (3  11)   x2  b   12 
x   13 
 3  
 4 1 0 
 
A   8 1 3 
 5 0 6 
 
Dual: Let w1, w2, w3 be the dual variables. The dual problem is
Min Z1 = bTW
Subject to: ATW  CT and W  0
 w1 
 
that is, Min Z1 = (8 12 13)  w2 
w 
 3
 4 8 5   w1   3 
Subject to:  1 1 0   w2    1
    
 0 3 6   w   1 
  3  
Min Z1 = 8w1 – 12w2 + 13w3
Subject to: 4 w1  8w2  5w3  3

 w1  w2  0w3  1
0 w1  3w2  6 w3  1
w1 , w2 , w3  0
Example 5.3: Write the dual of the following LPP.

Min Z = 2x2 + 5x3
Subject to: x1 x2 2
2 x1 x2 6 x3 6
x1 x2 3 x3 4
x1 , x2 , x3 0
Self-Instructional
96 Material
Solution: Since the given primal problem is not in the canonical form, we interchange Duality in Linear
Programming Problem
the inequality of the constraint. Also, the third constraint is an equation. This equation
can be converted into two inequations.
Min Z = 0x1 + 2x2 + 5x3
NOTES
Subject to: x1 x2 0 x3 2
2 x1 x2 6 x3 6
x1 x2 3x3 4
x1 x2 3x3 4
x1 , x2 , x3 0
Again on rearranging the constraint, we have

Min Z = 0x1 + 2x2 + 5x3
Subject to: x1 x2 0 x3 2
2 x1 x2 6 x3 6
x1 x2 3x3 4
x1 x2 3x3 4
x1 , x2 , x3 0
Dual: Since there are four constraints in the primal, we have four dual variables,
namely w1, w2, w3, w3
Max Z = 2w1 – 6w2 + 4w3 – 4w3
Subject to: w1 2w2 w3 w3 0

w1 w2 w3 w3 2
0 w1 6 w2 3w3 3w3 5
w1 , w2 , w3 , w3 0
Let w3 w3 w3
Max Z 2 w1 6w2 4( w3 w3 )
Subject to: w1 2 w2 ( w3 w3 ) 0
w1 w2 ( w3 w3 ) 2
Self-Instructional
Material 97
Duality in Linear
Programming Problem
Finally, we have, 0 w1 6w2 3( w3 w3 ) 5
Max Z 2 w1 6w2 4 w3
NOTES Subject to: w1 2 w2 w3 0

w1 w2 w3 2
0w1 6 w2 3w3 5
w1 , w2 0, w3 is unrestricted.
Example 5.4: Give the dual of the following problem.

Max Z = x + 2y
Subject to: 2 x 3 y 4
3x 4 y 5
x 0 and y unrestricted.
Solution: Since the variable y is unrestricted, it can be expressed as y y y ,

y,y 0 . On reformulating the given problem, we have
Max Z x 2( y y )
Subject to: 2x 3 ( y y ) 4
3x 4 ( y y ) 5
3x 4 ( y y ) 5
x, y , y 0
Since the problem is not in the canonical form we rearrange the constraints.
Max Z = x + 2y – 2y
Subject to: 2x 3 y 3y 4
3x 4 y 4y 5
3x 4 y 4y 5.
Dual: Since there are three variables and three constraints, in dual, we have
three variables, namely w1, w2, w2
Min Z 4 w1 5w2 5w2
Subject to: 2w1 3w2 3w2 1

3w1 4w2 4w2 2
3w1 4w2 4w2 2
w1 , w2 , w2 0
Self-Instructional
98 Material
Let w2 = w2 – w2, so that the dual variable w2 is unrestricted in sign. Finally, Duality in Linear
Programming Problem
the dual is:
Min Z 1 4 w1 5 ( w2 5w2 )
NOTES
2 w1 3 ( w2 w2 ) 1
Subject to: 3w1 4 ( w2 w ) 2
3w1 4 ( w2 w ) 2
that is, Min Z 1 = –4w1 + 5w2

Subject to: 2 w1 3w2 1
3w1 4 w2 2
3w1 4 w2 2
w1  0 and w2 is unrestricted.
Subject to: 2 w1 3w2 1

3w1 4 w2 2
3w1 4 w2 2

Subject to –2w1 + 3w2  1
–3w1 + 4w2  2, w1  0 and w2 is unrestricted.
Example 5.5: Write the dual of the following primal LPP.
Min Z = 4x1 + 5x2 – 3x3
Subject to: x1 x2 x3 22
3 x1 5 x2 2 x3 65
x1 7 x2 4 x3 120
x1 + x2  0 and x3 is unrestricted.
Solution: Since the variable x3 is unrestricted, x3 x3 x3 . Also, bring the
problem into the canonical form by rearranging the constraints.
Min Z = 4x1 + 5x2 – 3 ( x3 x3 )
Subject to: x1 x2 ( x3 x3 ) 22
x1 x2 x3 x3 22
3x1 5 x2 2 ( x3 x3 ) 65
x1 7 x2 4 ( x3 x3 ) 120
x1 , x2 , x3 x3 0 Self-Instructional
Material 99
Duality in Linear
Programming Problem
Min Z = 4x1 + 5x2 – 3 x3 3x3
Subject to: x1 x2 x3 x3 22
x1 x2 x3 x3 22
NOTES 3 x1 5 x2 2 x3 2 x3 65
x1 7 x2 4 x3 4 x3 120
x1 , x2 , x3 x3 0
Dual: Since there are four constraints in the primal problem, in dual, there are
four variables, namely w1 , w2 , w2 , w3 so that the dual is given by:
Max Z 22( w1 w1 ) 65w2 120 w3
Subject to: w1 w1 3w2 w3 4
w1 w1 5w2 7 w3 5
w1 w1 2w2 4w3 3
w1 w1 2w2 4w3 3
w1 w1 2w2 4w3 3
w1 , w1 , w2 , w3 0
Let w1 w1 w11 , i.e., the variable w1 is unrestricted.
Max Z 22( w1 w1 ) 65w2 120 w3
Subject to: w1 w1 3w2 w3 4
w1 w1 5w2 7 w3 5
( w1 w1 ) 2w2 4 w3 3
( w1 w1 ) 2w2 4 w3 3
that is, Max Z 22 w1 65w2 120 w3
Subject to: w1 3w2 w3 4
w1 5w2 7 w3 5
w1 2 w2 4 w3 3
w1 2 w2 4 w3 3
Finally, we have:
Min Z 22 w1 65w2 120 w3
Subject to: w1 3w2 4w3 4

w1 5w2 7 w2 5
w1 2 w2 4 w3 3
w2, w3  0 and w1 is unrestricted.
Self-Instructional
100 Material
Important Results in Duality Duality in Linear
Programming Problem
Some of the important results in duality are as follows:
 A primal will always constitute the dual of a dual.
 In a dual, if one is a minimization problem, then the other will be a NOTES
maximization problem.
 Both the LPP and its dual should have feasible solution, so as to arrive at an
optimal solution.
 The fundamental duality theorem states that in the LPP if the primal problem
has a finite optimal solution, then its dual also has a finite optimal solution.
Moreover, both the problems have the same optimal values of their objective
function, that is, Max Z = Min Z. The solution of the other problem can be
read from the Zj – Cj row below the columns of slack and surplus variables.
 The existence theorem states that there will be no feasible solution for a
problem if either problem in the LPP has an unbounded solution.
 The complementary slackness theorem states that:
(i) If a primal variable is positive, then its dual constraint will be an equation
at the optimum and vice versa.
(ii) If a primal constraint is an inequality, then the corresponding dual
variable is zero at the optimum and vice versa.
The optimal solution of a linear programming problem is formulated using various
methods. You have learned the use and importance of dual variables to solve the
LPP. Here, you will learn how sensitivity analysis helps to solve repeatedly the real
problem in a little different form. Generally, these scenarios crop up as an end
result of parameter changes due to the involvement of new advanced technologies
and the accessibility of well-organized latest information for key (input) parameters
or the ‘what-if’ questions. Thus, sensitivity analysis helps to produce optimal
solution of simple perturbations for the key parameters. For optimal solutions,
consider the simplex algorithm as a ‘black box’ which accepts the input key
parameters to solve LPP as shown below:
Simplex
Algorithm
Linear Optimal
Program Solution
Self-Instructional
Material 101
Duality in Linear Example 5.6: Illustrate sensitivity analysis using simplex method to solve the
Programming Problem
following LPP:
NOTES Subject to, x1  x2  3
3x1  x2  7
And x1, x2  0
Solution: Sensitivity analysis is done after making the initial and final tableau using
the simplex method. Add slack variables to convert it into equation form.
Maximize Z = 20x1 + 10x2 + 0S1 + 0S2
Subject to: x1  x2  S1  0 S2   3
3 x1  x2  0 S1  S2  7
Where x1, x2  0
To find basic feasible solution, we put x1 = 0 and x2 = 0. This gives Z = 0, S1
= 3 and S2 = 7. The initial table will be as follows:
Initial table
Cj 20 10 0 0
XB
CB B XB X1 X2 S1 S2 Min
Xi
0 S1 3 1 1 1 0 3/1 = 3
0 S2 7 3 1 0 1 7/3 = 2.33
Zj 0 0 0 0 0
Zj – Cj –20 –10 0 0
XB
Find X for each row and also find minimum for the second row. Here,
i
Zj – Cj is maximum negative (–20). Hence, x1 enters the basis and S2 leaves the
basis. It is shown with the help of arrows.
Key element is 3, key row is second row and key column is x1. Now
convert the key element into entering key by dividing each element of the key row
by key element using the following formula:
New element = Old element
 Product of elements in the key row and key column 

– Key element 
 
Self-Instructional
102 Material
The following is the first iteration table: Duality in Linear
Programming Problem
Cj 20 10 0 0
CB B XB X1 X2 S1 S2
NOTES
0 S1 2/3 0 2/3 1 –1/3 2 2
/ =1
3 3
20 X1 7/3 1 1/3 0 1 7 1
/ =7
3 3
Zj 140/3 20 20/3 0 20
Zj – Cj – 0 –10/3 0 20
Since Zj – Cj has one value less than zero, that is, negative value hence, this
is not yet the optima solution. Value –10/3 is negative; hence, x2 enters the basis
and S1 leaves the basis. Key row is upper row.
Cj 20 10 0 0
CB B XB X1 X2 S1 S2
 X2 1 0 1 3/2 –1/2
20 X1 4/3 1 0 0 4/3
Zj 110/3 20 10 0 25
Zj – Cj 0 0 0 25
4
Zj – Cj  0 for all; hence, optimal solution is reached, where x1  , x = 1,
3 2
80 110
Z + 10 = .
3 3
Check Your Progress

1. What are the modifications involved in the formulation of a dual problem?
2. State the important results in duality.
3. How does a sensitivity analysis help in producing optimal solution?

QUESTIONS
1. The modifications involved in the formulation of a dual problem are as follows:

 The objective function of minimization in the primal should be converted
into that of maximization in the dual and vice versa.
Self-Instructional
Material 103
Duality in Linear  The number of constraints in the dual will be equal to the number of
Programming Problem
variables in its primal and vice versa.
 If the variable in the primal is unrestricted in sign, then the corresponding
constraint in the dual will be an equation and vice versa.
NOTES
2. The important results in duality are as follows:
 A primal will always constitute the dual of a dual.
 In a dual, if one is a minimization problem, then the other will be a
 Both the LPP and its dual should have feasible solution, so as to arrive
at an optimal solution.
3. The sensitivity analysis helps to produce optimal solution of simple
perturbations for the key parameters. For optimal solutions, consider the
simplex algorithm as a ‘black box’ which accepts the input key parameters
to solve LPP as shown below:
Simplex
Algorithm
Linear Optimal
Program Solution
5.4 SUMMARY
 Every LPP (called the primal) is associated with another LPP (called its
dual). Either of the problem can be considered as primal with the other as
dual.
 In the formulation of a dual problem, it should be first converted into its
canonical form.
 The number of constraints in the dual will be equal to the number of variables
in its primal and vice versa.
 In a dual, if one is a minimization problem, then the other will be a
 The fundamental duality theorem states that in the LPP if the primal problem
has a finite optimal solution, then its dual also has a finite optimal solution.
Self-Instructional
104 Material
Moreover, both the problems have the same optimal values of their objective Duality in Linear
Programming Problem
function, that is, Max Z = Min Z. The solution of the other problem can be
read from the Zj – Cj row below the columns of slack and surplus variables.
 The complementary slackness theorem states that:
NOTES
(i) If a primal variable is positive, then its dual constraint will be an equation
at the optimum and vice versa.
(ii) If a primal constraint is an inequality, then the corresponding dual
variable is zero at the optimum and vice versa.
 The sensitivity analysis helps to produce optimal solution of simple
perturbations for the key parameters.
5.5 KEY WORDS
 Duality: It refers to the principle that optimization problems may be viewed

from either of two perspectives, the primal problem or the dual problem.
 Dual problem: It refers to an LP defined directly and systematically from
the primal LP model.
 Sensitivity Analysis: It refers to the study of how the uncertainty in the
output of a mathematical model or system can be divided and allocated to
different sources of uncertainty in its inputs.

EXERCISES

1. Write a brief note on the concept of duality.
2. Illustrate sensitivity analysis using simplex method to solve a hypothetical
LPP.
3. What do you mean by complementary slackness theorem?
1. What modifications are involved in the formulation of a dual problem?
Discuss.
2. What do you understand by primal problem? Discuss the concept of primal
problem with the help of suitable example.
Self-Instructional
Material 105
Duality in Linear
Programming Problem 5.7 FURTHER READINGS

NOTES Singh, J. K. 2009. Business Mathematics. New Delhi: Himalaya Publishing House.
Self-Instructional
106 Material
Transportation Problem
UNIT 6 TRANSPORTATION
PROBLEM
NOTES
Structure
6.0 Introduction
6.1 Objectives
6.2 Transportation Problem
6.2.3 Moving Towards Optimality
6.4 Summary
6.5 Key Words
6.0 INTRODUCTION
The transportation problem is a subclass of a linear programming problem (LPP).
Transportation problems deal with the objective of transporting various quantities
of a single homogeneous commodity initially stored at various origins, to different
destinations, in a way that keeps transportation cost at a minimum.
You will learn about the applications of the transportation problem and the
solutions and rules to solve such problems. The solution of any transportation
problem is obtained in two stages, namely initial solution and optimal solution.
There are three methods of obtaining an initial solution. These are: North West
Corner Rule, Least Cost Method and Vogel’s Approximation Method (VAM).
VAM is preferred since the solution obtained this way is very close to the optimal
solution. The optimal solution of any transportation problem is a feasible solution
that minimizes the total cost. An optimal solution is the second stage of a solution
obtained by improving the initial solution. Modified Distribution (MODI) method
is used to obtain optimal solutions and optimality tests.
6.1 OBJECTIVES
 Identify a transportation problem
 Discuss the various transportation models
 Explain the optimum solution using MODI method
 Explain the various special cases of a transportation problem
Self-Instructional
Material 107
6.2 TRANSPORTATION PROBLEM
The transportation problem (TP) is one of the subclasses of LPP (Linear

NOTES Programming Problem) in which the objective is to transport various quantities of
a single homogeneous commodity that are initially stored at various origins to different
destinations in such a way that the transportation cost is minimum. To achieve this
objective, we must know the amount and location of available supplies and the
quantities demanded. In addition, we must know the costs that result from
transporting one unit of commodity from various origins to various destinations.
Consider a transportation problem with m origins (rows) and n destinations
(columns). Let Cij be the cost of transporting one unit of the product from the ith
origin to jth destination, ai the quantity of commodity available at origin i, and bj
the quantity of commodity needed at destination j. Xij is the quantity transported
from ith origin to jth destination. This transportation problem can be stated in the
following tabular form.
X X X X
X X X X
X X X
X X X X
i=1 j=1
The linear programming model representing the transportation problem is given

by,
m n
Minimize Z   Cij X ij
i 1 j 1
n
 X ij  ai i  1, 2, ..., n
j 1
(Row Sum)
m
 X ij  b j j  1, 2, ..., n
i 1
(Column Sum)
X ij  0 For all i and j
Self-Instructional
108 Material
The given transportation problem is said to be balanced if, Transportation Problem
NOTES
that is, the total supply is equal to the total demand.
Feasible solution: Any set of non-negative allocations (Xij>0) which satisfies the
row and column sum (rim requirement) is called a feasible solution.
Basic feasible solution: A feasible solution is called a basic feasible solution if
the number of non-negative allocations is equal to m + n – 1, where m is the
number of rows and n the number of columns in a transportation table.
Non-degenerate basic feasible solution: Any feasible solution to a transportation
problem containing m origins and n destinations is said to be non-degenerate, if it
contains m + n – 1 occupied cells and each allocation is in independent positions.
The allocations are said to be in independent positions if it is impossible to
form a closed path. Closed path means by allowing horizontal and vertical lines
and when all the corner cells are occupied.
The allocations in the following tables are not in independent positions.
The allocations in the following tables are in independent positions.
Degenerate basic feasible solution: If a basic feasible solution contains

less than m + n – 1 non-negative allocations, it is said to be degenerate.
Self-Instructional
Material 109
Transportation Problem 6.2.3 Moving Towards Optimality
An optimal solution is a feasible solution (not necessarily basic) which minimizes
the total cost.
NOTES The solution of a transportation problem (TP) can be obtained in two stages,
namely initial solution and optimum solution.
Initial solution can be obtained by using any one of the three methods, namely:
 North West Corner Rule (NWCR)
 Least Cost Method or Matrix Minima Method
 Vogel’s Approximation Method (VAM)
VAM is preferred over the other two methods, since the initial basic feasible
solution obtained by this method is either optimal or very close to the optimal
solution.
The cells in the transportation table can be classified into occupied cells and
unoccupied cells. The allocated cells in the transportation table are called occupied
cells and the empty cells in the transportation table are called unoccupied cells.
The improved solution of the initial basic feasible solution is called optimal
solution which is the second stage of solution, that can be obtained by MODI.
North West Corner Rule
The following steps explain the North West Corner Rule:
Step 1: Starting with the cell at the upper left corner (North West) of the
transportation matrix, we allocate as much as possible so that either the capacity
of the first row is exhausted or the destination requirement of the first column is
satisfied, that is, X11 = Min (a1,b1).
Step 2: If b1>a1, we move down vertically to the second row and make the
second allocation of magnitude X22 = Min(a2 b1 – X11) in the cell (2, 1).
If b1<a1, move right horizontally to the second column and make the second
allocation of magnitude X12 = Min (a1, X11 – b1) in the cell (1, 2).
If b1 = a1, there is a tie for the second allocation. We make the second
allocations of magnitude,
X 12  Min (a1  a1 , b1 )  0 in the cell (1, 2)
or, X 21  Min (a2 , b1  b1 )  0 in the cell (2,1)
Step 3: Repeat steps 1 and 2 moving down towards the lower right corner of the
transportation table until all the rim requirements are satisfied.
Example 6.1: Obtain the initial basic feasible solution of a transportation problem
whose cost and rim requirement table is as follows:
Self-Instructional
110 Material
NOTES
Solution: Since ai = 34 = bj, there exists a feasible solution to the transportation
problem. We obtain the initial feasible solution as follows:
The first allocation is made in the cell (1, 1), the magnitude being
X11 = Min (5, 7) = 5.
The second allocation is made in the cell (2, 1) and the magnitude of the
allocation is given by X21 = Min (8, 7 – 5) = 2.
The third allocation is made in the cell (2, 2) the magnitude X22 = Min (8 – 2,
9) = 6.
The magnitude of the fourth allocation is made in the cell (3, 2) given by
X32 = Min (7, 9 – 6) = 3.
The fifth allocation is made in the cell (3, 3) with magnitude X33 = Min (7 – 3, 14)
= 4.
The final allocation is made in the cell (4, 3) with magnitude X43 = Min
(14, 18 – 4) = 14.
Hence, we get the initial basic feasible solution to the given TP and is given by,
X11 = 5; X21 = 2; X22 = 6; X32 = 3; X33 = 4; X43 = 14
Total Cost = 2 × 5 + 3 × 2 + 3 × 6 + 3 × 4 + 4 × 7 + 2 × 14
=10 + 6 + 18 + 12 + 28 + 28 = 102
Example 6.2: Determine an initial basic feasible solution to the following
transportation problem using North West Corner Rule.
Self-Instructional
Material 111
NOTES
Solution: The problem is a balanced TP, as the total supply is equal to the total
demand. Using the steps, we find the initial basic feasible solution as given in the
following table.
The solution is given by,

X11 = 6; X12 = 8; X22 = 2; X23 = 14; X33 = 1; X34 = 4
Total Cost = 6 × 6 + 4 × 8 + 2 × 9 + 2 × 14 +6 × 1 + 2 × 4
= 36 + 32 + 18 + 28 + 6 + 8 = 128
Least Cost or Matrix Minima Method
The following methods discuss the Least Cost or Matrix Minima Method:
Step 1 : Determine the smallest cost in the cost matrix of the transportation table.
Let it be Cij. Allocate Xij = Min (ai, bj) in the cell (i, j).
Step 2: If Xij = ai cross off the ith row of the transportation table and decrease
bj by ai. Then go to Step 3.
If Xij = bj cross off the jth column of the transportation table and decrease
ai by bj. Go to Step 3.
If Xij = ai = bj cross off either the ith row or the jth column, but not both.
Step 3: Repeat steps 1 and 2 for the resulting reduced transportation table until all
the rim requirements are satisfied. Whenever the minimum cost is not unique, make
an arbitrary choice among the minima.
Example 6.3: Obtain an initial feasible solution to the following TP using Matrix
Minima Method.
Self-Instructional
112 Material
Solution: Since ai = bj = 24, there exists a feasible solution to the TP using the Transportation Problem
steps in the least cost method, the first allocation is made in the cell (3, 1) the
magnitude being X31 = 4. This satisfies the demand at the destination D1 and we
delete this column from the table as it is exhausted.
NOTES
The second allocation is made in the cell (2, 4) with magnitude X24 = Min (6,
8) = 6. Since it satisfies the demand at the destination D4, it is deleted from the table.
From the reduced table, the third allocation is made in the cell (3, 3) with magnitude
X33 = Min (8, 6) = 6. The next allocation is made in the cell (2, 3) with magnitude X23
of Min (2, 2) = 2. Finally, the allocation is made in the cell (1, 2) with magnitude X12
= Min (6, 6) = 6. Now, all the requirements have been satisfied and hence, the initial
feasible solution is obtained.
X12 = 6; X23 = 2; X24 = 6; X31 = 4; X33 = 6
Since the total number of occupied cells = 5 < m + n – 1
We get a degenerate solution.
Total cost×××××
 28
Example 6.4: Determine an initial basic feasible solution for the following TP,
using the Least Cost Method.
Solution: Since ai = bj, there exists a basic feasible solution. Using the steps
in least cost method, we make the first allocation to the cell (1, 3) with magnitude
X13 = Min (14, 15) = 14 (as it is the cell having the least cost).
This allocation exhausts the first row supply. Hence, the first row is deleted.
From the reduced table, the next allocation is made in the next least cost cell (2, 3)
which is chosen arbitrarily with magnitude X23 = Min (1, 16) = 1. This exhausts
the third column destination.
Self-Instructional
Material 113
Transportation Problem From the reduced table, the next least cost cell is (3, 4) for which allocation
is made with magnitude Min (4, 5) = 4. This exhausts the destination D4 requirement.
Delete this fourth column from the table. The next allocation is made in the cell
(3, 2) with magnitude X32 = Min (1, 10) = 1 which exhausts the third origin capacity.
NOTES Hence, the third row is exhausted. From the reduced table, the next allocation is
given to the cell (2,1) with magnitude X21 = Min (6, 15) = 6. This exhausts the first
column requirement. Hence, it is deleted from the table.
Finally, the allocation is made to the cell (2, 2) with magnitude X22 = Min
(9, 9) = 9 which satisfies the rim requirement. These allocations are shown in the
following transportation table:
(I Allocation) (II Allocation)
(III Allocation) (IV Allocation)
(V, VI Allocation)
The following table gives the initial basic feasible solution.

X13=14; X21=6; X22= 9; X23= 1; X32= 1; X34= 4
Self-Instructional
114 Material
Transportation cost, Transportation Problem
= 14 × 1 + 6 × 8 + 9 × 9 + 1 × 2 + 3 × 1 + 4 × 2
= 14 + 48 + 81 + 2 + 3 + 8 = 156
Vogel’s Approximation Method (VAM) NOTES
The steps involved in Vogel’s Approximation Method (VAM) for finding the initial
solution are as follows:
Step 1: Find the penalty cost, namely the difference between the smallest and the
next smallest costs in each row and column.
Step 2: Among the penalties as found in Step (1), choose the maximum penalty. If
this maximum penalty is more than one (that is, if there is a tie), choose any one
arbitrarily.
Step 3: In the selected row or column as by Step (2), find out the cell having the
least cost. Allocate to this cell as much as possible depending on the capacity and
requirements.
Step 4: Delete the row or column which is fully exhausted. Again, compute
the column and row penalties for the reduced transportation table and then go to
Step (2). Repeat the procedure until all the rim requirements are satisfied.
Note: If the column is exhausted, then there is a change in row penalty and vice
versa.
Example 6.5: Find the initial basic feasible solution for the following transportation
problem using VAM.
Solution: Since ai = bj = 950, the problem is balanced and there exists a
feasible solution to the problem.
First, we find the row and column penalty PI as the difference between the
least and the next least cost. The maximum penalty is 5. Choose the first column
arbitrarily. In this column, choose the cell having the least cost name (1, 1). Allocate
to this cell with minimum magnitude (that is, Min (250, 200) = 200.) This exhausts
the first column. Delete this column. Since a column is deleted, there is a change
in row penalty PII while column penalty PII remains the same. Continuing in this
manner, we get the remaining allocations as given in the following table:
Self-Instructional
Material 115
Transportation Problem I Allocation II Allocation
NOTES
1
1
III Allocation IV Allocation
V Allocation VI Allocation
Finally, we arrive at the initial basic feasible solution which is shown in the
following table:
Self-Instructional
116 Material
There are six positive independent allocations which equals to m + n –1 = Transportation Problem
3 + 4 – 1. This ensures that the solution is a non-degenerate basic feasible solution.

Transportation cost
11 × 200 + 13 × 50 + 18 × 175 + 10 × 125 + 13 × 275 + 10 × 125 NOTES
= 12,075
Example 6.6: Find the initial solution to the following TP using VAM.
Solution: Since ai = bj, the problem is a balance TP. Hence, there exists a
feasible solution.
Finally, we have the initial basic feasible solution as given in the following
table.
There are six independent non-negative allocations equal to m + n – 1 = 3

+ 4 – 1 = 6. This ensures that the solution is non-degenerate basic feasible.
Self-Instructional
Material 117
Transportation Problem Transportation cost
= 3 × 45 + 4 × 30 + 1× 25 + 2 × 80 + 4 × 45 + 1 + 75
= 135 + 120 + 25 + 160 + 180 + 75
NOTES = 695
In this section, we will discuss the Optimum Solution Using MODI Method.
Optimality Test
Once the initial basic feasible solution has been computed, the next step in the
problem is to determine whether the solution obtained is optimum or not.
Optimality test can be conducted to any initial basic feasible solution of a
TP, provided such allocations has exactly m + n – 1 non-negative allocations,
where m is the number of origins and n is the number of destinations. Also, these
allocations must be in independent positions.
To perform this optimality test, we shall discuss the modified distribution
method (MODI). The various steps involved in the MODI method for performing
optimality test are as follows.
MODI Method
Step 1: Find the initial basic feasible solution of a TP by using any one of the three
methods.
Step 2: Find out a set of numbers ui and vj for each row and column satisfying
ui + vj = Cij for each occupied cell. To start with, we assign a number ‘0’ to any
row or column having the maximum number of allocations. If this maximum number
of allocations is more than 1, choose any one arbitrarily.
Step 3: For each empty (unoccupied) cell, we find the sum ui and vj written in the
bottom left corner of that cell.
Step 4: Find out for each empty cell the net evaluation value ij = Cij–(ui+vj) and
which is written at the bottom right corner of that cell. This step gives the optimality
conclusion as follows:
 If all ij > 0 (i.e., all the net evaluation values), the solution is optimum
and a unique solution exists.
 If ij  0, then the solution is optimum, but an alternate solution exists.
 If at least one ij < 0 , the solution is not optimum. In this case, we go to
the next step to improve the total transportation cost.
Step 5: Select the empty cell having the most negative value of ij. From this cell,
we draw a closed path by drawing horizontal and vertical lines with the corner
Self-Instructional
118 Material
cells occupied. Assign signs ‘+’ and ‘–’ alternately and find the minimum allocation Transportation Problem
from the cell having the negative sign. This allocation should be added to the
allocation having the positive sign and subtracted from the allocation having the
negative sign.
NOTES
Step 6: The previous step yields a better solution by making one (or more)
occupied cell as empty and one empty cell as occupied. For this new set of
basic feasible allocations, repeat from Step (2) till an optimum basic feasible
solution is obtained.
Example 6.7: Solve the following transportation problem.
Solution: We first find the initial basic feasible solution by using VAM. Since
ai= bj, the given TP is a balanced one. Therefore, there exists a feasible solution.
Finally, we have the initial basic feasible solution as given in the following
table.
Self-Instructional
Material 119
Transportation Problem From this table, we see that the number of non-negative independent
allocations is 6 = m+ n – 1 = 3 + 4 –1.
Hence, the solution is non-degenerate basic feasible.
NOTES Therefore, the initial transportation cost
= 11 × 13 + 3 × 14 + 4 × 23 + 6 × 17 + 17 × 10 + 18 × 9 = 711
To find the optimal solution: We apply the MODI method in order to determine
the optimum solution. We determine a set of numbers ui and vj for each row and
column, with ui+vj = Cij for each occupied cell. To start with, we give u2 = 0 as
the second row has the maximum number of allocation.
Now, we find the sum ui and vj for each empty cell and enter at the bottom
left corner of that cell.
C21 = u2 + v1 = 17 = 0 + v1 = 17  v1 = 17
C23 = u2 + v3 = 14 = 0 + v3 = 14  v3 = 14
C24 = u2 + v4 = 23 = 0 + v4 = 23  v4 = 23
C14 = u1 + v4 = 13 = u1 + 23 = 13  u1 = 10
C33 = u3 + v3 = 18 = u3 + 14 = 18  u3 = 4
Next, we find the net evaluations ji = Cij – (ui + vj) for each unoccupied
cell and enter at the bottom right corner of that cell.
Initial table
Since all ij > 0, the solution is optimal and unique. The optimum solution is
given by,
X14 = 11; X21 = 6; X23 = 3; X24 = 4; X32 = 10; X33 = 9
The minimum transportation cost
= 11 × 13 + 17 × 6 + 3 × 14 + 4 × 23 + 10 × 17 + 9 × 18 = 711
Example 6.8: Solve the following transportation problem starting with the initial
solution obtained by VAM.
Self-Instructional
120 Material
NOTES
Solution: Since ai = bj, the problem is a balanced TP. Therefore, there exists
a feasible solution.
Finally, the initial basic feasible solution is given as follows:
Since the number of occupied cells = 6 = m + n – 1 and are also independent,

there exists a non-degenerate basic feasible solution.
The initial transportation cost
=3×2+3×5+4×4+7×1+6×3+6×1= 68
To find the optimal solution, applying the MODI method, we determine a
set of numbers ui and vj for each row and column, such that ui + vj = Cij for each
occupied cell. Since the third now has the maximum number of allocations, we
give number u3= 0. The remaining numbers can be obtained as follows:
C31 = u3 + v1 = 7 = 0 + v1 =7  v1 = 7
C32 = u3 + v2 = 6 = 0 + v2 = 6  v2 = 6
C33 = u3 + v3 = 6 = 0 + v3 = 6  v3 = 6
Self-Instructional
Material 121
Transportation Problem C23 = u2 + v3 = 5 = u2 + 6 = 5  u2 = –1
C24 = u2 + v4 = 4 = –1 + v4 = 4  v4 = 5
C11 = u1 + v1 = 2 = u1 + 7 = 2  u1 = –5
NOTES We find the sum ui and vj for each empty cell and enter at the bottom left
corner of the cell. Next, we find the net evaluation ij given by ij = Cij – (ui + vj)
for each empty cell and enter at the bottom right corner of the cell.
Initial table
v1 v2 v3 v4
Since all ij > 0, the solution is optimum and unique. The solution is given
by,
X11 = 3; X23 = 3; X24 = 4, X31 = 1; X32 = 3; X33 = 1
The total transportation cost
=2×3+ 3×5+4×4+ 7×1+6 ×3+6×1= 68
Check Your Progress

1. What is a transportation problem?
2. Define feasible, basic, non-degenerate solutions of a transportation problem.
3. List the approaches used with transportation problems for determining the
starting solution.
4. State the optimal solution to a transportation problem.
5. What is the purpose of the MODI method?
6. State the two conditions necessary for an alternate solution.

QUESTIONS
1. A transportation problem deals with transportation of various quantities of

a single homogeneous commodity initially stored at various origins to different
destinations at the minimum cost.
Self-Instructional
122 Material
2. Definitions of feasible, basic, non-degenerate solutions of a transportation Transportation Problem
problem are as follows:

 Feasible solution: Any set of non-negative allocations which satisfies
the row and column sum (rim requirement) is called a feasible solution.
NOTES
 Basic feasible solution: A feasible solution is called a basic feasible
solution if the number of non-negative allocation is equal to m+n–1,
where m is the number of rows and n the number of columns in a
transportation table.
 Non-degenerate basic feasible solution: A feasible solution to a
transportation problem containing m origins and n destinations is said to
be non-degenerate, if it contains m+n–1 occupied cells and each
allocation is in independent positions.
3. The approaches used with transportation problems for determining the
starting solution are as follows:
 North West Corner Rule
 Least Cost Method (Matrix Minima)
 Vogel’s Approximation Method
4. An optimal solution to a transportation problem is one which minimizes the
total transportation cost.
5. The purpose of the MODI method is to get the optimal solution of a
transportation problem.
6. The two conditions necessary for an alternate solution are as follows:
(i) The slope of the objective function should be similar as constraint
which forms the edge of the feasible region.
(ii) The constraint must be an active constraint, that is, in the path of optimal
progress of the objective function.
6.4 SUMMARY
 The transportation problem (TP) is one of the subclasses of LPP (Linear

Programming Problem) in which the objective is to transport various quantities
of a single homogeneous commodity that are initially stored at various origins
to different destinations in such a way that the transportation cost is minimum.
 Any set of non-negative allocations (Xij > 0) which satisfies the row and
column sum (rim requirement) is called a feasible solution.
 A feasible solution is called a basic feasible solution if the number of non-
negative allocations is equal to m + n – 1, where m is the number of rows
and n the number of columns in a transportation table.
Self-Instructional
Material 123
Transportation Problem  Any feasible solution to a transportation problem containing m origins and
n destinations is said to be non-degenerate, if it contains m + n – 1 occupied
cells and each allocation is in independent positions.
 The allocations are said to be in independent positions if it is impossible to
NOTES
form a closed path. Closed path means by allowing horizontal and vertical
lines and when all the corner cells are occupied.
 If a basic feasible solution contains less than m + n – 1 non-negative
allocations, it is said to be degenerate basic feasible solution.
 An optimal solution is a feasible solution (not necessarily basic) which
minimizes the total cost.
 The solution of a transportation problem (TP) can be obtained in two stages,
namely initial solution and optimum solution.
 The cells in the transportation table can be classified into occupied cells and
unoccupied cells. The allocated cells in the transportation table are called
occupied cells and the empty cells in the transportation table are called
unoccupied cells.
 Optimality test can be conducted to any initial basic feasible solution of a
TP, provided such allocations has exactly m + n – 1 non-negative allocations,
where m is the number of origins and n is the number of destinations. Also,
these allocations must be in independent positions. To perform this optimality
test, modified distribution method (MODI) is used.
 MODI method is applied in order to determine the optimum solution. One
can determine a set of numbers ui and vj for each row and column, with
ui + vj = Cij for each occupied cell. To start with, we give u2 = 0 as the
second row has the maximum number of allocation.
6.5 KEY WORDS
 Unbalanced Transportation: The given TP is said to be unbalanced if the

total supply is not equal to the total demand.
 Vogel’s Approximation Method (VAM): It refers to iterative procedure
calculated to find out the initial feasible solution of the transportation problem.
 MODI Method: The Modified Distribution Method or MODI is an
efficient method of checking the optimality of the initial feasible solution.
Self-Instructional
124 Material
EXERCISES
Short Answer Questions NOTES
1. What are the numbers of non-basic variables for 4 rows and 5 columns?
2. While dealing with North West Corner rule, when does one move to the
next cell in next column?
3. What is the coefficient of Xij of constraints in a transportation problem?
4. When does degeneracy occur in an m × n transportation problem?
1. What do you understand by transportation model?
2. Explain the following with examples:
(i) North West Corner Rule
(ii) Least Cost Method
(iii) Vogel’s Approximation Method
3. Explain degeneracy in a transportation problem. Describe a method to resolve
it.
4. What do you mean by an unbalanced transportation problem? Explain the
process of converting an unbalanced transportation problem into a balanced
one.
5. Give the mathematical formulation of a transportation problem.
6. Write an algorithm to solve a transportation problem.

Self-Instructional
Material 125
Assignment Problem
UNIT 7 ASSIGNMENT PROBLEM

NOTES Structure
7.0 Introduction
7.1 Objectives
7.2 Assignment Problem
7.2.2 Hungarian Method Algorithm
7.2.3 Routing Problem: The Travelling Salesman Problem
7.4 Summary
7.5 Key Words
7.0 INTRODUCTION
This unit deals with a very interesting method called the ‘assignment technique’,
which is applicable to a class of very practical problems generally called ‘assignment
problems’.
The objective of assignment problems is to assign a number of origins (jobs)
to the equal number of destinations (persons) at a minimum cost or maximum
profit.
7.1 OBJECTIVES

 Define assignment problem and state mathematical formulation of the same
 Differentiate between transportation and assignment problems
 Derive the solution of an assignment problem using Hungarian Method
 Understand unbalanced assignment problem and modified matrix
 Assess maximization in assignment problem
7.2 ASSIGNMENT PROBLEM
Definition
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
degrees of efficiency. Let cij be the cost if the ith person is assigned to the jth job.
The problem is to find an assignment (which job should be assigned to which
Self-Instructional
126 Material
person, on a one to one basis) so that the total cost of performing all the jobs is Assignment Problem
minimum. Problems of this kind are known as assignment problems.

An assignment problem can be stated in the form of n × n cost matrix [cij] of
real numbers as given in the following table.
NOTES
Jobs
1 2 3 ... j ... n
c c c13 ... c1 j ... c1n 
1  11 12
c
2  21 22c c23 ... c2 j ... c2 n 

3  c31 c32 c33 ... c3 j ... c3n 
Persons  
i  ci1 ci 2 ci3 ... cij ... cin 
  
n  cn1 cn 2 cn3 ... cnj ... cnn 

Mathematically, an assignment problem can be stated as,
n n
Minimize z =   cij xij , where, i = 1, 2, ... n, and j = 1, 2, ... n
i 1 j 1
Subject to the restrictions,
1, if the ith person is assigned jth job
xij = 
0, if not
n
 xij = 1 (one job is done by the ith person)
j 1
n
and  xij = 1 (only one person should be assigned the jth job)
j 1
where, xij denotes that the jth job is to be assigned to the ith person.
Difference between Transportation and Assignment Problems

Transportation problem Assignment problem
1. Number of sources and destinations Since assignment is done on a one to one

need not be equal. Hence, the cost matrix basis, the number of sources and
is not necessarily a square matrix. destinations are equal. Hence, the cost
matrix must be a square matrix.
2. xij, the quantity to be transported from xij, the jth job is to be assigned to the
ith origin to jth destination can take any ith person and can take either the
possible positive value, and it satisfies the value 1 or zero.
rim requirements.
3. The capacity and the requirement value is equal The capacity and the requirement value
to ai and bj for the ith source and jth destination is exactly one, i.e., for each source of
(i = 1, 2, ...m; j = 1, 2, ... n). each destination, the capacity and the
requirement value is exactly one.
4. The problem is unbalanced if the total supply The problem is unbalanced if
and total demand are not equal. the cost matrix is not a square matrix.
Self-Instructional
Material 127
Assignment Problem 7.2.2 Hungarian Method Algorithm
Solution of an assignment problem can be arrived at, by using the Hungarian
method. The steps involved in this method are as follows.
NOTES Step 1 Prepare a cost matrix. If the cost matrix is not a square matrix then
add a dummy row (column) with zero cost element.
Step 2 Subtract the minimum element in each row from all the elements of the
respective rows.
Step 3 Further, modify the resulting matrix by subtracting the minimum element of
each column from all the elements of the respective columns. Thus, obtain
the modified matrix.
Step 4 Then, draw the minimum number of horizontal and vertical lines to cover
all zeros in the resulting matrix. Let the minimum number of lines be N.
Now there are two possible cases.
Case I If N = n, where n is the order of matrix, then an optimal assignment
can be made. So make the assignment to get the required solution.
Case IIIf N < n, then proceed to step 5.
Step 5 Determine the smallest uncovered element in the matrix (element not
covered by N lines). Subtract this minimum element from all uncovered
elements and add the same element at the intersection of horizontal and
vertical lines. Thus, the second modified matrix is obtained.
Step 6 Repeat steps 3 and 4 until we get the case (i) of Step 4.
Step 7 (To make zero assignment) Examine the rows successively until a row-
wise exactly single zero is found. Circle (O) this zero to make the
assignment. Then mark a cross (×) over all zeros if lying in the column
of the circled zero, showing that they cannot be considered for future
assignment. Continue in this manner until all the zeros have been examined.
Repeat the same procedure for columns also.
Step 8 Repeat step 6 successively until one of the following situation arises—
(i) If no unmarked zero is left, then the process ends or
(ii) If there lie more than one unmarked zero in any column or row, circle
one of the unmarked zeros arbitrarily and mark a cross in the cells of
remaining zeros in its row or column. Repeat the process until no
unmarked zero is left in the matrix.
Step 9 Thus, exactly one marked circled zero in each row and each column of
the matrix is obtained. The assignment corresponding to these marked
circled zeros will give the optimal assignment.
Self-Instructional
128 Material
Example 7.1: Using the following cost matrix, determine (a) optimal job assignment Assignment Problem
(b) the cost of assignments.

Job
1 2 3 4 5 NOTES
A 10 3 3 2 8
B 9 7 8 2 7
C 7 5 2 4
Mechanic
6
 
D3 5 8 2 4
E  9 10 9 6 10 
Solution: Select the smallest element in each row and subtract this smallest
element from all the elements in its row.
1 2 3 4 5
A 8 1 1 0 6
B 7 5 6 0 5
 
C 5 3 4 0 2
D 1 3 6 0 2
E 3 4 3 0 4 
Select the minimum element from each column and subtract from all other elements
in its column. With this we get the first modified matrix.
1 2 3 4 5
A 7 0 0 0 4
B 6 4 5 0 3
 
C 4 2 3 0 0
D 0 2 5 0 0
E  2 3 2 0 2 
In this modified matrix we draw the minimum number of lines to cover all zeros
(horizontal or vertical).
1 2 3 4 5
A 7 0 0 0 4
B 6 4 5 0 3
 
C 4 2 3 0 0
D 0 2 5 0 0
E  2 3 2 0 2 
Number of lines drawn to cover all zeros is 4 = N.
The order of matrix is n = 5
Hence, N < n.
Self-Instructional
Material 129
Assignment Problem Now we get the second modified matrix by subtracting the smallest
uncovered element from the remaining uncovered elements and adding it to the
element at the point of intersection of lines.
1 2 3 4 5
NOTES

A 9 0 0 2 6
 3
B 6 2 3 0

C 4 0 1 0 0
D 0 0 3 0 0


E 2 1 0 0 2 

Number of lines drawn to cover all zeros = N = 5
The order of matrix is n = 5.
Hence N = n. Now we determine the optimum assignment.
Assignment
1 2 3 4 5
A 9 0 0 2 6 
 
B 6 2 3 0 3

C 4 0 1 0 0 
D0 0 3 0 0

 
E 2 1 0 0 2 

First row contains more than one zero. So proceed to the 2nd row. It has
exactly one zero. The corresponding cell is (B, 4). Circle this zero thus, making
an assignment. Mark (×) for all other zeros in its column. Showing that they cannot
be used for making other assignments. Now row 5 has a single zero in the cell
(E, 3). Make an assignment in this cell and cross the 2nd zero in the 3rd column.
Now row 1 has a single zero in the column 2, i.e., in the cell (A, 2). Make
an assignment in this cell and cross the other zeros in the 2nd column. This leads
to a single zero in column 1 of the cell (D, 1), make an assignment in this cell and
cross the other zeros in the 4th row. Finally, we have a single zero left in the 3rd
row, making an assignment in the cell (C, 5). Thus, we have the following
assignment.
Optimal assignment and optimum cost of assignment.
Job Mechanic Cost
1 D 3
2 A 3
3 E 9
4 B 2
5 C 4
21
Self-Instructional
130 Material
Therefore, 1  D, 2 A, 3  E, 4  B, 5  C, with minimum cost Assignment Problem
equal to 21.
Example 7.2: A company has 5 jobs to be done on five machines. Any job can
be done on any machine. The cost of doing the jobs on different machines are
NOTES
given below. Assign the jobs for different machines so as to minimize the total
cost.
Machines
Jobs A B C D E
1 13 8 16 18 19
2 9 15 24 9 12
3 12 9 4 4 4
4 6 12 10 8 13
5 15 17 18 12 20
Solution: We form the first modified matrix by subtracting the minimum element
from all the elements in the respective row, and the same with respective columns.
Step 1 A B C D E
1 5 0 8 10 11 
20 6 15 0 3 
 
3 8 5 0 0 0
40 6 4 2 7
5  3 5 6 0 8 
Since each column has the minimum element 0, we have the first modified matrix.
Now we draw the minimum number of lines to cover all zeros.
Step 2 A B C D E
1 5 0 8 10 11
2 0 6 15 0 3 
 
3 8 5 0 0 0
4 0 6 4 2 7
5  3 5 6 0 8 
Step 3 5 0 5 10 8 
0 6 12 0 0 
 
11 8 0 3 0
0 6 1 2 4
3 5 3 0 5 

Here, the smallest uncovered element is ‘3’. So ‘3’ is added to all the
junction points i.e., the prints of intersection of lines.
Number of lines drawn to cover zero is N = 4 < the order of matrix
n = 5.
We find the second modified matrix by subtracting the smallest uncovered
element (3) from all the uncovered elements and adding to the element that is
the point of intersection of lines.
Self-Instructional
Material 131
Assignment Problem A B C D E
Step 4
1 5 0 5 10 8
2 0 6 12 0 0
 
3 11 8 0 3 0
NOTES 4 0 6 1 2 4
5  3 5 3 0 5 
Number of lines drawn to cover all zeros = 5,
which is the order of matrix. Hence, we can form an assignment.
Assignment
A B C D E
15 0 5 10 8 
2  0 6 12 0 0 
3  11 8 0 3 0
 
4 0 6 1 2 4
5  3 5 3 0 5 
All the five jobs have been assigned to 5 different machines.

Here the optimal assignment is,
Job Machine
1 B
2 E
3 C
4 A
5 D
Minimum (Total cost) = 8 + 12 + 4 + 6 + 12 = 42.
Example 7.3: Four different jobs can be done on four different machines and the
take-down time costs are prohibitively high for change overs. The matrix below
gives the cost in rupees for producing job i on machine j.
Machines
Jobs M1 M2 M3 M4
J1 5 7 11 6
J2 8 5 9 6
J3 4 7 10 7
J4 10 4 8 3
How should the jobs be assigned to the various machines so that the total
cost is minimized.
Self-Instructional
132 Material
Solution: We form a first modified matrix by subtracting the least element in Assignment Problem
the respective rows and respective columns.

M1 M2 M3 M4
J1  0 2 6 1 
NOTES
J 2  3 0 4 1 
J3  0 3 6 3
 
J4  7 1 5 0
Since the third column has no zero element, we subtract the smallest element
4 from all the elements.
M1 M2 M3 M4
J1  0 2 2 1 
J 2  3 0 0 1 
J3  0 3 2 3
 
J4  7 1 1 0
Now we draw minimum number of lines to cover all zeros.

M1 M2 M3 M4
J1  0 2 2 1 
J 2  3 0 0 1 
J3  0 3 2 3
 
J4  7 1 1 0
Number of lines drawn to cover all zeros = 3, which is less than the order
of matrix = 4.
Hence, we form the 2nd modified matrix, by subtracting the smallest
uncovered element from all the uncovered elements and adding to the element
that is at the point of intersection of lines.
M1 M2 M3 M4 M1 M2 M3 M4
J1  0 1 1 1  J1  0 0 0 0
J 2  4 0 0 2  J 2  5 0 0 2 
J3  0 2 1 3 J3  0 1 0 2
   
J4  7 0 0 0 J4  8 0 0 0
N=3<n=4 N=4=n=4
Hence, we can make an assignment.
M1 M2 M3 M4
J1 0 0 0 0 
 
J2 5 0 0 2 
J3 0 1 0 2 
 
J4  8 0 0 0 
Self-Instructional
Material 133
Assignment Problem Since no rows and no columns have single zero, we have a different
assignment (Multiple solution).
Optimal assignment
Job Machine
NOTES J1 M4
J2 M2
J3 M1
J4 M3
Minimum (Total cost)

6+5+4+8= 23
Alternate Solution
J1  M1; J2  M2; J3  M3; J4  M4.
Minimum (Total cost)
5 + 5 + 10 + 3 = 23.
Example 7.4: Solve the following assignment problem in order to minimize the
total cost. The cost matrix given below gives the assignment cost when different
operators are assigned to various machines.
Operators
I II IV VIII
A  30 25 35 36 
33
B  23 29 23 26 
38
Machines 
C  30 27 22 22 
22
 
D  25 31 29 27 32 
E  27 29 30 24 32 
Solution: We form the first modified matrix by subtracting the least element from
all the elements in the respective rows and then in the respective columns.
I IIIII IV V
A 5 0 8 10 11 
B  0 6 15 0 3 
C 8 5 0 0 0
 
D0 6 4 2 7 
E  3 5 6 0 8 
Since each column has the minimum element 0, the first modified matrix is
obtained. We draw the minimum number of lines to cover all zeros.
The number of lines drawn to cover all zeros = 4 < the order of matrix
= 5. Hence, we form the second modified matrix by subtracting the smallest
Self-Instructional
134 Material
uncovered element from the remaining uncovered elements and adding to the Assignment Problem
element that is at the point of intersection of lines.

I II III IV V
A 5 0 5 10 8 
B 0 6 12 0 0  NOTES

C 11 8 0 3 0
 
D 0 6 1 2 4
E  3 5 3 0 5 
N = 5, i.e., the number of lines drawn to cover all zeros = order of matrix. Hence,
we can make an assignment.
I II III IV V
A 5 0 5 10 8 
B  0 6 12 0 0 
C 11 8 0 3 0
 
D 0 6 1 2 4
E 3 5 3 0 5 
The optimum assignment is

Operators Machines
I D
II A
III C
IV E
V B
The optimum cost is given by

25 + 25 + 22 + 24 + 26 = 122.
Unbalanced Assignment problem
Any assignment problem is said to be unbalanced if the cost matrix is not a square
matrix, i.e., the number of rows and columns are not equal. To make it balanced,
we add a dummy row or dummy column with all the entries as zero.
Example 7.5: There are four jobs to be assigned to five machines. Only one
job can be assigned to one machine. The amount of time in hours required for the
jobs per machine are given in the following matrix.
Machines
Jobs A B C D E
1 4 3 6 2 7
2 10 12 11 14 16
3 4 3 2 1 5
4 8 7 6 9 6
Self-Instructional
Material 135
Assignment Problem Find an optimum assignment of jobs to the machines to minimize the total
processing time and also find out for which machine no job is assigned. What is
the total processing time to complete all the jobs?
Solution Since the cost matrix is not a square matrix, the problem is unbalanced.
NOTES
We add a dummy job 5 with corresponding entries zero.
Modified Matrix
A B C D E
1 4 3 6 2 7
2 10 12 11 14 16 
34 3 2 1 5
 
48 7 6 9 6
5  0 0 0 0 0 
We subtract the smallest element from all the elements in the respective rows.
A B C D E
1 2 1 4 0 5
2  0 2 1 4 6 
3 3 2 1 0 4
 
4 2 1 0 3 0
5  0 0 0 0 0 
Since each column has minimum element as zero, we draw minimum number of
lines to cover all zeros.
A B C D E
1 2 1 4 0 5
2  0 2 1 4 6 
3 3 2 1 0 4
 
4 2 1 0 3 0
5  0 0 0 0 0 
The number of lines to cover all zeros = 4 < the order of matrix. We form the
2nd modified matrix by subtracting the smallest uncovered element from the
remaining uncovered elements and adding to the element at the point of
intersection of lines.
A B C D E
1 2 0 3 0 4 
 
20 1 0 4 5 
3 3 1 0 0 3 
 
43 1 0 4 0 
5  1 0 0 1 0 
Self-Instructional
136 Material
Here the number of lines drawn to cover all zeros = 5 = Order of matrix. Assignment Problem
Therefore, we can make the assignment

A B C D E
1 2 0 3 0 4
2 0 1 0 4 5  NOTES
3 3 1 0 0 3
 
4 3 1 0 4 0
5 1 0 0 1 0 
Optimum assignment
1 B 3
2 A 10
3 C 2
4 D 6
For machine D, no job is assigned.
Optimum (minimum) cost = 3 + 10 + 1 + 6 = 20.
Example 7.6: A company has 4 machines to do 3 jobs. Each job can be assigned
to only one machine. The cost of each job on each machine is given below.
Determine the job assignments that will minimize the total cost.
Machine
W X Y Z
A  18 24 28 32 
B  8 13 17 18 
Job
C  10 15 19 22 
Solution: Since the cost matrix is not a square matrix, we add a dummy row D
with all the elements 0.
W X Y Z
A  18 24 28 32 
B  8 13 17 18 
C  10 15 19 22 
 
D 0 0 0 0
Subtract the minimum element in each row from all the elements in its row.
W X Y Z
A 0 6 10 14 
B 0 5 9 10 
C 0 5 9 12 
 
D 0 0 0 0
Self-Instructional
Material 137
Assignment Problem Since each column has a minimum element 0, we draw minimum number of lines
to cover all zeros.
W X Y Z
A 0 6 10 14 
NOTES
B  0 5 9 10 
C 0 5 9 12 
 
D0 0 0 0
 The number of lines drawn to cover all zeros = 2 < the order of matrix, we
form a second modified matrix.
W X Y Z
A 0 1 5 9
B  0 0 4 5 
C 0 0 4 7
 
D 5 0 0 0
Here, N = 3 < n = 4.
Again we subtract the smallest uncovered element from all the uncovered
elements and add to the element at the point of intersection
W X Y Z
A 0 1 1 4
B  0 0 0 1 
C 0 0 0 3
 
D 9 4 0 0
Here, N = 4 = n. Hence, we make an assignment.

Assignment
W X Y Z W X Y Z
A 0 1 1 4 A W A 0 1 1 4 AW
B  0 0 0 1  DZ or B  0 0 0 1  DZ
C 0 0 0 3 BX C  0 0 0 3 B Y
   
D 9 4 0 0 C Y D 9 4 0 0 CX
Since D is a dummy job, machine Z is assigned no job.

Therefore, optimum cost = 18 + 13 + 19 = 50.
Maximization in Assignment problem
In this, the objective is to maximize the profit. To solve this, we first convert the
given profit matrix into the loss matrix by subtracting all the elements from the
highest element. For this converted loss matrix we apply the steps in Hungarian
method to get the optimum assignment.
Self-Instructional
138 Material
Example 7.7: The owner of a small machine shop has four mechanics available Assignment Problem
to assign jobs for the day. Five jobs are offered with expected profit for each
mechanic on each jobs, which are as follows:
Job
NOTES
A B CD E
1  62 78 111 82 
50
 84 61 73 59 
Mechanic 2  71
3  87 92 111 71 81 
 
4  48 64 87 77 80 
By using the assignment method, find the assignment of mechanics to the

job that will result in maximum profit. Which job should be declined ?
Solution: The given profit matrix is not a square matrix as the number of jobs is not
equal to the number of mechanics. Hence, we introduce a dummy mechanic 5
with all the elements 0.
Job
A B C D E
1  62 78 50 111 82 
2  71 84 61 73 59 
Mechanic 
3 87 92 111 71 81
 
4  48 64 87 77 80 
5  0 0 0 0 0 
Now we convert this profit matrix into loss matrix by subtracting all the
elements from the highest element 111.
Loss Matrix
A B C D E
1  49 33 61 0 29 
2  40 27 50 38 52 

3  24 19 0 40 30 
 
4  63 47 24 34 31 
5 111 111 111 111 111
We subtract the smallest element from all the elements in the respective rows.
A B C D E
1  49 33 61 0 29 
2  13 0 23 11 25 
3  24 19 0 40 30 
 
4  39 23 0 10 7 
5  0 0 0 0 0 
Self-Instructional
Material 139
Assignment Problem Since each column has minimum element as zero, we draw minimum
number of lines to cover all zeros.
A B C D E
NOTES 1  49 33 61 0 29 
2 13 0 23 11 25
3  24 19 0 40 30 
 
4  39 23 0 10 7
5  0 0 0 0 0 
Here the number of lines drawn to cover all zeros = N = 4, is less than
the order of matrix.
We form the 2nd modified matrix by subtracting the smallest uncovered element
from the remaining uncovered elements and adding to the element that is at the
point of intersection of lines.
A B C D E
1  49 40 68 0 29
2  6 0 23 5 18 
3 17 19 0 33 23
 
4 32 23 0 3 0
5  0 7 7 0 0 
Here, N = 5 = n (the order of matrix).

We make the assignment.
A B C D E
1  49 40 68 0 29 
2  6 0 23 5 18 
3 17 19 0 33 23 
 
4 32 23 0 3 0
5  0 7 7 0 0 
The optimum assignment is

Job Mechanic
A 5
B 2
C 3
D 1
E 4
Since the 5th mechanic is a dummy, job A is assigned to the 5th mechanic,
this job is declined.
The maximum profit is given by, 84 + 111 + 111 + 80 = 386.
Self-Instructional
140 Material
Example 7.8: A marketing manager has 5 salesmen and there are 5 sales districts. Assignment Problem
Considering the capabilities of the salesmen and the nature of districts, the
estimates made by the marketing manager for the sales per month (in 1,000
rupees) for each salesman in each district would be as follows.
A B C D E
NOTES
1 32 38 40 28 40 
2  40 24 28 21 36 
3  41 27 33 30 37 
 
4  22 38 41 36 36 
5  29 33 40 35 39 
Find the assignment of salesmen to the districts that will result in the
maximum sales.
Solution: We are given the profit matrix. To maximize the profit, first we convert
it into a loss matrix, which can be minimized. To convert it into loss matrix, we
subtract all the elements from the highest element 41. Subtract the smallest element
from all the elements in the respective rows and columns, to get the first modified
matrix.
Loss Matrix
A B C D E
1 9 3 1 13 1
2  1 17 13 20 5 
3  0 14 8 11 4
 
4 19 3 0 5 5
5 12 8 1 6 2 
A B C D E
1 8 2 0 12 0
2  0 16 12 19 4 
3  0 14 8 11 4
 
4 19 3 0 5 5
5 11 7 0 5 1 
A B C D E
1  8 0 0 7 0
2  0 14 12 14 4 
3  0 12 8 6 4 
 
4  19 1 0 0 5 
5  11 5 0 0 1 
Self-Instructional
Material 141
Assignment Problem We now draw minimum number of lines to cover all zeros.
A B C D E
18 0 0 7 0
2  0 14 12 4 4 
NOTES
3  0 12 8 6 4
 
4 9 1 0 0 5
5  11 5 0 0 1 
N=4<n=5
We subtract the smallest uncovered element from the remaining uncovered
elements and add to the elements at the point of intersection of lines, to get the
second modified matrix.
A B C D E
1 9 0 1 8 0
2  0 13 12 4 3 
3  0 11 8 6 3
 
4 9 0 0 0 4
5  11 4 0 0 0 
Again, N = 4 n = 5. Repeat the above step.

A B C D
E
1 12 0 1 0
8
2  0 10 9 0 
1
30 8 5 0
3
 
4 12 0 0 0 4
5 14 4 0 0 0 
N = 5 = n = 5. Hence we make the assignment.

Assignment
A B C D E
1 12 0 1 8 0
 
20 10 9 1 0 
30 8 5 3 0
 
4 12 0 0 0 4
5 14 4 0 0 0 
Since no row or column has single zero, we get a multiple solution.

(i) The optimum assignment is:
1  B, 2  A, 3  E, 4  C, 5  D.
With maximum profit (38 + 40 + 37 + 41 + 35) = 191
Self-Instructional
142 Material
(ii) The optimum assignment is: Assignment Problem
1  B, 2  A, 3  E, 4  D, 5  C.
Maximum profit (38 + 40 + 37 + 36 + 40) = 191
7.2.3 Routing Problem: The Travelling Salesman Problem NOTES
A network routing problem consists of finding an optimum route between two or

more nodes in relation to total time, cost, or distance. Various constraints may
exist, such as a prohibition on returning to a node already visited or a stipulation of
passing through every node only once. Network routing problems commonly arise
in communication and transportation systems. Delays that occur at the nodes
(e.g., railroad classification yards or telephone switchboards) may be a function
of the loads placed on them and their capacities. Breakdowns may occur in either
links or nodes. Much studied is the “traveling salesman problem,” which consists
of starting a route from a designated node that goes through each node (e.g., city)
only once and returns to the origin in the least time, cost, or distance. This problem
arises in selecting an order for processing a set of production jobs when the cost
of setting up each job depends on which job has preceded it.
Let us discuss the ‘Traveling Salesman Problem’.
Assuming a salesman has to visit n cities. He wishes to start from a particular
city, visit each city once and then return to his starting point. His objective is to
select the sequence in which the cities are visited in such a way that his total
travelling time is minimized.
To visit 2 cities (A and B, there is no choice. To visit 3 cities we have 2 possible
routes. For 4 cities we have 3 possible routes. In general, to visit n cities there
are (n – 1) possible routes.
Mathematical Formulation
Let Cij be the distance or time or cost of going from city i to city j. Let the decision
variable Xij be 1, if the salesman travels from city i to city j, otherwise let it be 0.
The objective is to minimize the travelling time.
n n
Z=   cij xij
i 1 j 1

n
Z=  X ij = 1, i = 2... n.
j 1
n
Z=  X ij = 1, j = 2 ... n.
i 1
and subject to the additional constraint that Xij is so chosen that, no city is visited
twice before all the cities are visited.
Self-Instructional
Material 143
Assignment Problem In particular, going from i directly to i is not permitted. This means
Cij =, when i = j.
In the travelling salesman problem we cannot choose the element along the
diagonal and this can be avoided by filling the diagonal with infinitely large
NOTES
elements.
The travelling salesman problem is very similar to the assignment problem
except that in the former case, there is an additional restriction, that Xij is so chosen
that no city is visited twice before the tour of all the cities is completed.
Example 7.9: A travelling salesman has to visit 5 cities. He wishes to start from
a particular city, visit each city once and then return to his starting point. Cost of
going from one city to another is shown below. You are required to find the least
cost route.
To City
A B C D E
A   4 10 14 2 
B  12  6 10 4 
From City  
C  16 14  8 14 
 
D  24 8 12  10 
E  2 6 4 16  
Solution: Treat the problem as an assignment problem and solve it using the same
procedures. If the optimal solution of the assignment problem satisfies the additional
constraint, then it is also an optimal solution of the given travelling salesman problem.
If the solution to the assignment problem does not satisfy the additional restriction,
then after solving the problem by assignment technique, we use the method of
enumeration.
First we solve this problem as an assignment problem.
Subtract the minimum element in each row from all the elements in its row.
A B C D E
A  2 8 12 0 
B 8  2 6 0 

C 8 6  0 6
 
D 16 0 4  2
E  0 4 2 14  
Subtract the minimum element in each column from all the elements in its
column.
A B C D E
A  2 8 12 0 
B 8  2 6 0 

C 8 6  0 6
 
D 16 0 4  2
E  0 4 2 14  
Self-Instructional
144 Material
We have the first modified matrix. Draw minimum number of lines to cover Assignment Problem
all zeros.
A B C D E
A  2 6 12 0 
B  8  2 6 0  NOTES
C 8 8  0 6
 
D 16 0 2  2
E  0 4 0 14  
N = 4 < n = 5. Subtract the smallest uncovered element from all the uncovered
elements and add to the element that is at the point of intersection of lines. Hence,
we get the 2nd modified matrix.
A B C D E
A  0 6 12 0 
B  8  0 6 0 
C 8 4  0 6
 
D 18 0 4  4
E  0 4 0 14  
N = 5 = n = 5. We make an assignment.
Assignment
A B C D E
A  0 6 0
12 AE
  BC
B 8  0 6 0
C 8 4  0 6 CD
 
D 18 0 4  4 DB
E  0 4 0 14   EA
A E  A
The salesman should go from A to E and then come back to A without

covering B, C, D. But this is contradicting the constraint that no city is visited twice
before all the cities are visited.
Hence, we obtain the next best solution by bringing the next minimum non-
zero element namely 4.
A B C D E
A  0 6 12 0 
B 8  0 6 0 

C 8 4  0 6
 
D 18 0 4  4
E  0 4 0 14  
A  B, B  C, C  D, D  E, E  A
Self-Instructional
Material 145
Assignment Problem Since all the cities have been visited and no city is visited twice before
completing the tour of all the cities, we have an optimal solution for the travelling
salesman.
The least cost route is A  B  C  D  E  A.
NOTES
Total cost = 4 + 6 + 8 + 10 + 2 = 30.
Example 7.10: A machine operator processes five types of items on his machine
each week and must choose a sequence for them. The set-up cost per change
depends on the items presently on the machine and the set-up to be made,
according to the following table.
To Item
A B C D E
A  4 7 3 4 
B 4  6 3 4
From Item  
C 7 6  7 5
 
D3 3 7  7
E  4 4 5 7  
If he processes each type of item only once in each week, how should he
sequence the items on his machine in order to minimize the total set-up cost?
Solution: Reduce the cost matrix and make assignments in rows and columns
having single row.
Modify the matrix by subtracting the least element from all the elements in its
row and also in its column.
A B C D E
A  1 4 0 1 
B  1  3 0 1 
C 2 1  2 0
 
D 0 0 4  4
E  0 0 1 3  
A B C D E
A  1 3 0 1 
B  1  2 0 1 
C 2 1  2 0
 
D 0 0 3  4
E  0 0 0 3  
Here, N = 4 n = 5, i.e., N < n.
Self-Instructional
146 Material
Subtract the smallest uncovered element from all the uncovered elements Assignment Problem
and add to the element that is at the point of intersection of lines and get the
reduced 2nd modified matrix.
A B C D E
NOTES
A  0 2 0 1 
B  0  1 0 1 
C 1 0  1 0
 
D 0 0 3  5
E  0 0 0 4  
N = 5 = n = 5. We make the assignment.

Assignment
A B C D E
A 0 2 0 1  AB
 
B0  1 0 1 BD
C1 0  1 0 CE
 
D 0 0 3  5 DA
E  0 0 0 4   E  C
We get the solution A  B  D  A.

This schedule does not provide the required solution as each item is not
processed only once in a week.
Hence, we make a better solution by considering the next smallest non-zero
element by considering 1.
A B C D E
A 0 2 0 1 
 
B0  1 0 1
C1 0  1 0
 
D 0 0 3  5
E  0 0 0 4  
A  E, E  C, C  B, B  D, D  A,
i.e., A  E  C  B  D  A.
The total set-up cost comes to 21.
Check Your Progress

1. State any one difference between transportation and assignment problems.
2. What is an unbalanced assignment problem?
3. What do you mean by routing problem?
Self-Instructional
Material 147
Assignment Problem
QUESTIONS
NOTES 1. The one main difference between transportation problem and assignment
problem is that in transportation problem the number of sources and
destinations need not be equal. Hence, the cost matrix is not necessarily a
square matrix. While in assignment problem the assignment is done on a
one to one basis, the number of sources and destinations are equal. Hence,
the cost matrix must be a square matrix.
2. Any assignment problem is said to be unbalanced if the cost matrix is not a
square matrix, i.e., the number of rows and columns are not equal. To
make it balanced, we add a dummy row or dummy column with all the
entities as zero.
3. A network routing problem consists of finding an optimum route between
two or more nodes in relation to total time, cost, or distance. Various
constraints may exist, such as a prohibition on returning to a node already
visited or a stipulation of passing through every node only once.
7.4 SUMMARY
 An assignment problem is one of the fundamental combinatorial optimization

problems and helps to find a maximum weight identical in nature in a weighted
bipartite graph.
 The solution of an assignment problem can be arrived at using the Hungarian
method.
 An assignment problem is balanced if the cost matrix is a square matrix;
otherwise, it is termed as unbalanced.
 To convert an unbalanced assignment problem into a balanced problem,
dummy rows or columns are added with all entries as 0s.
7.5 KEY WORDS
 Assignment problem: It helps to find a maximum weight identical in nature

in a weighted bipartite graph.
 Unbalanced assignment problem: Any assignment problem is said to be
unbalanced if the cost matrix is not a square matrix.
 Square matrix: In mathematics, a square matrix is a matrix with the same
number of rows and columns.
Self-Instructional
148 Material
Assignment Problem
EXERCISES
1. State the differences between a transportation problem and an assignment

problem.
2. Give a mathematical formulation of the assignment problem.
3. How can you maximize an objective function in the assignment problem?
1. Discuss the assignment problem with a suitable example.
2. Describe the algorithm for the solution of the assignment problem.
3. Explain the nature of i in travelling salesman problem and give its mathematical
formulation.
4. Solve the following assignment problem.
Men
A B C D A B C D
I 1 4 6 3 I  10 25 15 20 
(a) II  9 7 10 9  (b) Jo b II  15 30 5 15 
   
III  4 5 11 7  III  35 20 12 24 
IV  8 7 8 5  IV  17 25 24 20 
5. Men
A B C D E
I 1 3 2 8 8
II  2 4 3 1 5 
Tasks III  5 6 3 4 6
 
IV  3 1 4 2 2
V  1 5 6 5 4 
6. There are five jobs to be assigned, one each to 5 machines and the associated
cost matrix is as follows.
Machine
1 2 3 4 5
A  11 17 8 16 20 
B  9 7 12 6 15 
Job C  13 16 15 12 16 
 
D  21 24 17 28 26 
E  14 10 12 11 15 
Self-Instructional
Material 149
Assignment Problem 7. A salesman has to visit five cities A, B, C, D and E. The distance (in hundred
miles) between the five cities is as follows.
To
A B C D E
A- 7 6 8 4
NOTES
B  7 - 8 5 6 
From C  6 8 - 9 7 0
 
D 8 5 9 - 8
E  4 6 7 8 - 
If the salesman starts from city A and has to come back to his starting point,
which route should he select so that the total distance travelled is minimum?
8. Determine the optimum assignment schedule for the following assignment
problem. The cost matrix is given below.
Machine
 1 2 3 4 5 6 
 17 8 16 20 15 
A  11
B 9 7 12 6 15 13 
 
Job C  13 16 15 12 16 8 
 24 17 28 2 15 
D  21 
E  14 10 12 11 15 6 
If the job C cannot be assigned to machine 6, will the optimum solution

change?

House.
Self-Instructional
150 Material
Integer Programming
Problem
BLOCK - II
INTEGER PROGRAMMING AND
QUEUING CONCEPTS NOTES
UNIT 8 INTEGER PROGRAMMING

PROBLEM
Structure
8.0 Introduction
8.1 Objectives
8.2 Importance of Integer Programming Problems
8.2.1 Types of Integer Programming Problem
8.2.3 Gomory’s Fractional Cut Algorithm or Cutting Plane Method for Pure
(All) IPP
8.4 Summary
8.5 Key Words
8.0 INTRODUCTION
A linear programming problem in which all or some of the decision variables are
constrained to assume non-negative integer values is called an Integer
Programming Problem (IPP).
In a linear programming problem, if all variables are required to take integral
values then it is called the Pure (all) Integer Programming Problem (Pure IPP).
If only some of the variables in the optimal solution of a LPP are restricted to
assume non-negative integer values, while the remaining variables are free to take
any non-negative values, then it is called a Mixed Integer Programming Problem
(Mixed IPP).
Further, if all the variables in the optimal solution are allowed to take values 0
or 1, then the problem is called the 0–1 Programming Problem or Standard
Discrete Programming Problem.
The general integer programming problem is given by, Max Z = CX
Axb
x  0 and some or all variables are integers. Self-Instructional
Material 151
Integer Programming
Problem 8.1 OBJECTIVES

NOTES  Understand the importance of integer programming problems
 Discuss the applications and types of integer programming problem
 Explain the concept of mixed integer programming problem
 Describe the Branch and Bound Method
8.2 IMPORTANCE OF INTEGER PROGRAMMING

PROBLEMS
In LPP, all the decision variables were allowed to take any non-negative real
values as it is quite possible and appropriate to have fractional values in many
situations. There are several frequently occurring circumstances in business and
industry that lead to planning models involving integer-valued variables. For
example, in production, manufacturing is frequently scheduled in terms of batches,
lots or runs. In allocation of goods, a shipment must involve a discrete number
of trucks or aircrafts. In such cases the fractional values of variables like 13/3
may be meaningless in the context of the actual decision problem.
Applications of Integer programming

Integer programming is applied in business and industry. All assignment and
transportation problems are integer programming problems, as in the assignment
and travelling salesmen problem, all the decision variables are either zero or one.
i.e., xij = 0 or 1
Other examples are capital budgeting and production scheduling problems.
In fact, any situation involving decisions of the type ‘either to do a job or not’
can be viewed as an IPP In all such situations,
xij = 1, if the jth activity is performed,
0 if the jth activity is not performed.
In addition, allocation problems involving the allocation of men or machines
give rise to IPP, since such commodities can be assigned only in integers and
not in fractions.
Note: If the non-integer variable is rounded off, it violates the feasibility and
there is no guarantee that the rounded off solution will be optimal. Due to these
difficulties, there is a need for developing a systematic and efficient procedure
for obtaining the exact optimal integer solution to such problems.
Self-Instructional
152 Material
8.2.1 Types of Integer Programming Problem Integer Programming
Problem
There are two methods used to solve IPP, namely,
(i) Gomory’s Cutting Plane Method
(ii) Branch and Bound Method (Search Method). NOTES
A systematic procedure for solving pure IPP was first developed by R.E. Gomory,
in 1956, which he later used to deal with the more complicated case of mixed
integer programming problem. This method consists of first solving the IPP as an
ordinary LPP by ignoring the restriction of integer values and then introducing a
new constraint to the problem such that the new set of feasible solution includes
all the original feasible integer solutions, but does not include the optimum non-
integer solution initially found. This new constraint is called ‘Fractional cut’ or
‘Gomorian constraint’. Then the revised problem is solved using the simplex
method, till an optimum integer solution is obtained.
Search Method
This is an enumeration method in which all feasible integer points are enumerated.
The widely used search method is the Branch and Bound method. It was
developed in 1960, by A.H. Land and A.G. Doig. This method is applicable to
both pure and mixed IPP. It first divides the feasible region into smaller subsets
that eliminate parts containing no feasible integer solution.
8.2.3 Gomory’s Fractional Cut Algorithm or Cutting Plane Method

for Pure (All) IPP
Step 1 Convert the minimization IPP into an equivalent maximization IPP. Ignore
the integrality condition.
Step 2 Introduce slack and/or surplus variables if necessary, to convert the given
LPP in its standard form and obtain the optimum solution of the given
LPP by using simplex method.
Step 3 Test the integrality of the optimum solution.
(i) If all xBi  0 and are integers, an optimum integer solution is
obtained.
(ii) If all xBi  0 and at least one xBi is not an integer, then go to the
next step.
Step 4 Rewrite each xBi as xBi = [xBi] + fi where xBi is the integral part of xBi
and fi is the positive fractional part of xBi 0  fi < 1.
Choose the largest fraction of xBi’s, i.e., Choose max ( fi ), if there is a
tie, select arbitrarily. Let max ( fi ) = fK, corresponding to xBK (the Kth
row is called the ‘source row’).
Step 5 Express each negative fraction, if any, in the source row of the optimum
simplex table as the sum of a negative integer and a non-negative fraction. Self-Instructional
Material 153
Integer Programming Step 6 Find the fractional cut constraint (Gomorian Constraint)
Problem
n
From the source row  ak j
x j = xBi
j 1
NOTES n
i.e.,  ([akj ]  fkj) x j = [xBK] + fK
j 1
n n
in the form  fkj x j  fK   f kj x j  – fK
j 1 j 1
n
or,   fkj x j  G1 = –fK
j 1
where, G1 is the Gomorian slack.

Step 7 Add the fractional cut constraint obtained in step (6) at the bottom of
the simplex table obtained in step (2). Find the new feasible optimum
solution using dual simplex method.
Step 8 Go to step (3) and repeat the procedure until an optimum integer solution
is obtained.
Example 8.1: Find the optimum integer solution to the following LPP.
Max Z = x1 + x2
Subject to constraints, 3x1 + 2x2  5
x2  2
x1, x2  0 and are integers.
Solution: Introducing the non-negative slack variable S1, S2  0, the standard
form of the LPP becomes,
Max Z = x1 + x2 + 0S1 + 0S2
Subject to, 3x1 + 2x2 + S1 = 5
0x1 + x2 + S2  2
x1, x2, S1, S2  0
Ignoring the integrality condition, solve the problem by simplex method. The
initial basic feasible solution is given by,
S1 = 5 and S2 = 2.
Since all Zj – Cj  0 an optimum solution is obtained, given by
Max Z = 7/3, x1 = 1/3, x2 = 2.
To obtain an optimum integer solution, we have to add a fractional cut constraint
in the optimum simplex table.
Since xB = 1/3, the source row is the first row.
Self-Instructional
154 Material
Expressing the negative fraction –2/3 as a sum of negative integer and positive Integer Programming
Problem
fraction, we get
–2/3 = –1 + 1/3
NOTES
Since x1 is the source row, we have,

1/3 = x1 + 1/3 S1 – 2/3 S2
i.e., 1/3 = x1 + 1/3S1 + ( –1 + 1/3) S2
The fractional cut (Gomorian) constraint is given by
1/3S1 + 1/3S2  1/3
 – 1/3S1 – 1/3S2  –1/3
 – 1/3S1 – 1/3 S2 + G1 = –1/3
where, G1 is the Gomorian slack. Add this fractional cut constraint at the bottom
of the above optimal simplex table.
We apply dual simplex method. Since G1 = –1/3, G1 leaves the basis. To

find the entering variable we find,
 Z j  C j   1/ 3 1/ 3 
Max  , aik  0  = Max  , 
 aik   1/ 3 1/ 3 
Max {–1, –1} = –1
Self-Instructional
Material 155
Integer Programming We choose S1 as the entering variable arbitrarily.
Problem
NOTES
Since all Zj – Cj  0 and all xBi  0, we obtain an optimal feasible integer

solution.
 The optimum integer solution is,
Max Z = 2, x1 = 0, x2 = 2.
Example 8.2: Find an optimum integer solution to the following LPP.
Max Z = x1 + 2x2
Subject to the contraints, 2x2  7
x1 + x2  7
2x1  11
x1, x2  0 and x1, x2 are integers.
Solution: Introducing slack variables S1, S2, S3  0, we get,
Max Z = x1 + 2x2 + 0S1 + 0S2 + 0S3
Subject to, 2x2 + S1 = 7
x1 + x2 + S2 = 7
2x1 + S3 = 11
Ignoring the integer condition, we get the optimum solution of the given LPP,
with initial basic feasible solution as, S1 = 7, S2 = 7, S3 = 11.
Self-Instructional
156 Material
Since all Zj – Cj  0, an optimum solution is obtained which is given by, Integer Programming
Problem
7 7
Max Z = 21, x1 = , x2 =
2 2 2
Since the optimum solution obtained above is not an integer, we now select a NOTES
constraint corresponding to
Max {fi} = Max {f1, f2, f3}
x 1 = 7/2 = 3 + 1/2
x 2 = 7/2 = 3 + 1/2
S3 = 4 = 4 + 0
 Max {fi} = Max  1 , 1 , 0  = 1/2

 
2 2 
Since the max fraction is same for both x1 and x2 rows, we choose x1 row
as the source row arbitrarily. From this row we have,
7/2 = x1 + 0x2 – 1/2 S1 + 1S2 + 0S3.
On expressing the negative fraction as a sum of negative integer and a positive
fraction, we have,
3 + 1/2 = x1 + 0x2 + (–1 + 1/2) S1 + 1S2 + 0S3
 The Gomorian constraint is given by,
1/2 S1  1/2
i.e., –1/2 S1  –1/2  –1/2 S1 + G1 = –1/2
where, G1 is the Gomorian slack. Adding this new constraint at the bottom of
the above optimal simplex table, we get a new table.
We apply dual simplex method. Since G1 = –1/2, G1 leaves the basis. Entering
variable is given by,
 Z j  C j   1/ 2 
Max  , ak j  0  = Max  
 ak j    1/ 2 
gives the non-basic variable S1 to enter into the basis. Drop G1 and introduce S1. Self-Instructional
Material 157
Integer Programming
Problem
NOTES
Since all Zj – Cj  0, an optimum solution has been obtained in integers. Hence,

the integer optimum solution is given by,
Max Z = 10, x1 = 4, x2 = 3.
Example 8.3: Solve the following integer programming problem.
Max Z = 2x1 + 20x2 – 10x3
Subject to, 2x1 + 20x2 + 4x3  15
6x1 + 20x2 + 4x3 = 20
x1, x2, x3  0 and are integers.
Solution: Introducing slack variable S1  0 and an artificial variable A1  0,
the initial basic feasible solution is S1 = 15, A1 = 20. Ignoring the integer condition,
solve the problem by simplex method.
Max Z = 2x1 + 20x2 – 10x3 + 0S1 – MA1
Subject to, 2x1 + 20x2 + 4x3 + S1 = 15
6x1 + 20x2 + 4x3 + A1 = 20
x1, x2, x3, S1, A1  0
Self-Instructional
158 Material
Integer Programming
Problem
NOTES
Since all Zj – Cj  0, the solution is optimum but the variables are non-integer.
 The non-integer optimum solution is given by,
x1 = 5/4, x2 = 5/8, x3 = 0, Max Z = 15
To obtain an integer optimum solution, we proceed as follows.
Max {f1 f2} = Max {5/8, 1/4} = 5/8
 The source row is the first row, namely, x2 row. From this source row
we have,
5/8 = 0x1 + 1x2 + (1/5)x3 + (3/40)S1.
The fractional cut constraint is given by,
(1/5)x3 + (3/40)S1  5/8
(–1/5)x3 – (3/40)S1  –5/8  (–1/5)x3 – (3/40)S1 + G1 = 5/8
We apply dual simplex method. Since G1 = –5/8, G1 leaves the basis.

 Z j  C j   14 1  40
Also, Max  , ai k  0  = Max   = Max 
 ai k   1/5 3/40  3
gives the non-basic variable S1, this enters the basis.
Self-Instructional
Material 159
Integer Programming
Problem
NOTES
Again since the solution is non-integer, we add one more fractional cut
constraint.
Max{fi} = Max { 0, 1/3, 1/3}
Since the max fraction is same for both the rows x1 and S1, we choose S1
arbitrarily.
 From the source row we have,
25/3 = 0x1 + 0x2 + (8/3)x3 + 1S1 – (40/3)G1
Expressing the negative fraction as the sum of negative integer and positive
fraction we have,
(8 + 1/3) = 0x1 + 0x2 + (2 + 2/3)x3 + 1S1 + (–14 + 2/3)G1
The corresponding fractional cut is given by,
–2/3x3 – 2/3 G1 + G2 = –1/3.
Add this second Gomorian constraint at the bottom of the above simplex table
and apply dual simplex method.
Since G2 = –1/3, G2 leaves the basis. Also,

 Z j  Cj   34 / 3 40 / 3 
Max  , ai k  0  = Max     = – 17
 ai k   2 / 3 2 / 3 
gives the non-basic variable x3 which enters the basis. Using dual simplex method,
introduce x3 and drop G2.
Self-Instructional
160 Material
Since the solution is still a non-integer, a third fractional cut is required. It is Integer Programming
Problem
given from the source row (x3 row) as,
– 1/2 = –1/2 G2 + G3
Insert this additional constraint at the bottom of the table, the modified simplex NOTES
table is shown below.
Using dual simplex method, we drop G3 and introduce G2.
Since all Zj – Cj  0 and also the variables are integers, the optimum integer
solution is obtained and given by, x1 = 2, x2 = 0, x3 = 2 and Max Z = 16.
Example 8.4: Solve the integer programming problem.
Max Z = 7x1 + 9x2
Subject to,– x1 + 3x2 6
7x1 + x2  35
x1, x2  0 are integers.
Solution: Introducing slack variables S1, S2  0, we get the standard form of
LPP as,
Max Z = 7x1 + 9x2 + 0S1 + 0S2
Subject to, – x1 + 3x2 + S1 = 6
7x1 + x2 + S2 = 35
Self-Instructional
Material 161
Integer Programming Now ignoring the integer conditions, solve the given LPP by simplex method.
Problem
NOTES
9
Since all Zj – Cj  0, optimum solution is obtained as x1 =
2
x2 = 7 and Max Z = 63.

2
Since the optimum solution obtained above is not an integer solution, we select
a constraint corresponding to,
Max {fi} = Max {f1 f2}
= Max  ,   xB1   [3]  , xB 2   [4]  

1 1 7 1 9 1
2 2  2 2 2 2
Since both the equations have the same value of fi, either one of the two
equations can be used. Let us consider the x2 row as source row.
From x2 row we have,
7 7 1
 0 x1  x2  S1  S2
2 22 22
There is no negative fraction.
The Gomorian constraint is given by,
7 1 1
S1  S2 
22 22 2
7 1 1
i.e., S1  S2  
22 22 2
7 1 1
  S1  S 2  G1 = 
22 22 2
Self-Instructional
162 Material
where, G1 is the Gomorian slack. Adding this new constraint at the bottom of Integer Programming
Problem
the above optimal simplex table, we have the new table.
NOTES
We apply dual simplex method, since G1 = – 1/2, G1 leaves the basis. Also,
 28 15 
 Z j  C j   
Max  , aik  0  = Max  11 11

 aik   7 1 

 22 22 
= MAX (–8, – 30) = – 8
gives the non-basic variable S1 to enter into the basis.
Applying dual simplex method, drop G1 and introduce S1.
The optimal solution obtained by dual simplex method as above is still a

non-integer. Thus a new Gomory’s constraint is to be reconsidered.
4
Max {fi} = Max , ,  = 7
4 4
 7 7
Choose the x1 row as source row arbitrarily as both the fraction values are
the same. From the source row we have,
4 1 6
= 1x1  0 x2  0S1  S2  G1
7 7 7
There is no negative fraction in the source row.
Self-Instructional
Material 163
Integer Programming The Gomory’s constraint is given by,
Problem
1 6 4 1 6 4
S 2  G1  i.e., S 2  G1  G2 = 
7 7 7 7 7 7
NOTES where, G2 is the Gomorian slack. Adding this constraint in the above simplex
table we get a modified table.
We again apply the dual simplex method.

4
Since G2 =  , G2 leaves the basis. Also,
7
 
 Z j  C j   1 8 
Max  , aik  0  = Max  , 
 aik   1  6 
 7 7
= Max (–7, –9) = –7
gives the non-basic variable S2 to enter into the basis.
Since all Zj – Cj  0 and also the solution is an integer, we obtain an optimum

integer solution given by, x1 = 4, x2 = 3 and Max Z = 55.

In mixed IPP only some of the variables are restricted to integer values, while
the other variables may take integer or other real values.
Mixed integer cutting plane procedure The iterative procedure for the solution
of mixed integer programming problem is as follows.
Step 1 Reformulate the given LPP into a standard maximization form and then
determine an optimum solution using simplex method.
Self-Instructional
164 Material
Step 2 Test the integrality of the optimum solution. Integer Programming
Problem
(i) If all xBi  0 (i = 1, 2,... m) and are integers, then the current solution
is an optimum one.
(ii) If all xBi  0 (i = 1, 2,... m) but the integer restricted variables are
NOTES
not integers, then go to the next step.
Step 3 Choose the largest fraction among those xBi, which are restricted to
integers. Let it be xBk = fk (assume)
Step 4 Find the fractional cut constraints from the source row, namely Kth row.
From the source row,
n
 akj k j = xBk
j 1
n
i.e,   akj   fki  rj = [xBK] + fk
j 1
n
in the form  fki x j  fk

j 1
 f 
i.e.,  f kj x j   f k k1   f kj x j  + fk
j  j   j  j
 f 
  f kj x j   f k k1   f kj x j  – fk
j j   j  j
 f 
  f kj x j   f k k1   f kj x j  Gk = – fk
j  j   j j
where, Gk is Gomorian slack
j    j / f kj  0 
 
j 1   j / f kj  0 
Step 5 Add this cutting plane generated in step K at the bottom of the optimum
simplex table obtained in step 1. Find the new optimum solution using
dual simplex method.
Step 6 Go to step 2 and repeat the procedure until all xBi  0 (i = 1, 2,... m)
and all restricted variables are integers.
Example 8.5:
Max Z = x1 + x2
Subject to,3x1 + 2x2 5
x2  2
x1 + x2  0 and x1 is an integer.
Self-Instructional
Material 165
Integer Programming Solution: Introducing slack variables S1, S2  0 the standard form of LPP is,
Problem
Max Z = x1 + x2 + 0S1 + 0S2
Subject to,3x1 + 2x2 + S1 = 5
NOTES x2 + S2 = 2
x1, x2, S1, S2  0
Initial basic feasible solution,
S1 = 5, S2 = 2
Ignore the integer condition and solve the problem using simplex method, to
obtain optimum solution.
Since all Zj – Cj  0, the current basic feasible solution is optimum. But x1 is

non-integer. From the source row (first row) we have,
1/3 = x1 + 0 x2 + 1/3 S1 – 2/3 S2
 1 
1    2  1
S1   3    S2 
3  1  1   3  3
3 
1 1 1 1 1 1
S1  S 2    S1  S2  
3 3 3 3 3 3
1 1 1
S1  S 2  G1 = 
3 3 3
Self-Instructional
166 Material
Adding this Gomorian constraint at the bottom of the above simplex table, Integer Programming
Problem
we have,
NOTES
Using the dual simplex method, since G1 = –1/3 < 0, G1 leaves the basis.
Also,
 Z j  C j 
Max  , aik  0 
 aik 
 1 1 
 3  = Max {–1, –1} = –1
Max  3 , 
 1 1 
 3 3 
As this corresponds to both S1 and S2, we choose S1 arbitrarily as the entering
variable.
Drop G1 and introduce S1.
Since all Zj – Cj  0 and all xBi  0, the current solution is feasible and optimal.
The required optimal integer solution is given by,
x1 = 0, x2 = 2 and Max Z = 2.
Example 8.6:
Max Z = 4x1 + 6x2 = 2x3
Subject to, 4x1– 4x2  5,
–x + 6x2  
x1 + x2 + x3  
x1, x2, x3  and x1, x3 are integers.
Solution Introducing slack variables S1, S2, S3  0 the standard form of LPP
is,
Max Z = 4x1 + 6x2 + 2x3 + 0S1 + 0S2 + 0S3
Self-Instructional
Material 167
Integer Programming Subject to, 4x1 – 4x2 + S1 = 5
Problem
– x1 + 6x2 + S2 = 5
– x1 + x2 + x3 + S3 = 5
The initial basic feasible solution is given by S1 = 5, S2 = 5 and S3 = 5. Ignoring
NOTES
the integer condition, the optimum solution of given LPP is obtained by the simplex
method.
Since all Zj – Cj  the solution is optimum. But the integer constrained
variables x1 and x3 are non-integer.
x 1 = 5/2 = 2 + 1/2
x 2 = 25/4 = 6 + 1/4
Max (f1 f3) = Max (1/2, 1/4) = 1/2
From the first row we have,
(2 + 1/2 ) = x1 + 0x2 + 0x3 + (3/10) S1 + (1/5) S2

3/10 S1 + 1/5 S2  1/2
–3/10S1 – 1/5 S2  –1/2
i.e., –3/10 S1 – 1/5 S2 + G1 = – 1/2, where G1 is the Gomorian slack. Introduce
this new constraint at the bottom of the above simplex table.
Self-Instructional
168 Material
Using dual simplex method, since G1 = – 1/2 < 0, Integer Programming
Problem
G1 leaves the basis. Also,
 
 Z j  C j   2 2   20   20
Max  , aik  0  = Max  ,  = Max  ,  10 = NOTES
 aik    3  1   3  3
 10 5 
corresponding to S1.Therefore, the non-basic variable S1 enters the basics. Drop

G1 and introduce S1.
Since all Zj – Cj  0, the solution is optimum and also the integer restricted
variable x3 = 35/6 is not an integer, therefore, we add another Gomorian
constraint
x3 = 35/6 = 5 + 5/6
The source row is the third row.
From this row we have,
5 1 5
5 = 0 x1  0 x2  x3  0S1  S2  S3  G1
6 6 6
 5 
 6   1  5 5
    S2  G1 
 5  1   6  6 6
6 
Self-Instructional
Material 169
Integer Programming 5 5 
Problem  S 2  G1 
6 6 6
5 5 5
 S 2  G1  G2 
6 6 6
NOTES where, G2 is the Gomorian slack.
Add this second cutting plane constraint at the bottom of the above optimum
simplex table.
Use dual simplex method. ( G2 = – 5/6 < 0)

G2 leaves the basics.
 2 20 
 Z j  C j   3  = Max   4 ,  8 =  4 ,
Also, Max  aik  0  = Max  3 ,   
 aik   5 5   5  5
 6 6 
which corresponds to S2.
Drop G2 and introduce S2.
Since all Zj – Cj  and also all the restricted variables x1 and x3 are integers,
an optimum integer solution is obtained.
The optimum integer solution is,
x1 = 2, x2 = 1, x3 = 6 and Max Z = 26

This method is applicable to both, pure as well as mixed IPP. Sometimes a few
or all the variables of an IPP are constrained by their upper or lower bounds.
The most general method for the solution of such constrained optimization
problems is called ‘Branch and Bound method’.
Self-Instructional
170 Material
This method first divides the feasible region into smaller subsets and then Integer Programming
Problem
examines each of them successively, until a feasible solution that gives an optimal
value of objective function is obtained.
Let the given IPP be,
NOTES
Max Z = CX
Subject to,Ax  b
X  0 are integers.
In this method, we first solve the problem by ignoring the integrality condition.
(i) If the solution is in integers, the current solution is optimum for the given
IPP.
(ii) If the solution is not in integers, say one of the variable Xr is not an integer,
then xr* < xr < x*r +1 where x*r , x*r +1 are consecutive non-negative
integers.
Hence, any feasible integer value of xr must satisfy one of the two conditions.
xr  xr* or xr  x*r + 1.
These two conditions are mutually exclusive (both cannot be true
simultaneously). By adding these two conditions separately to the given LPP, we
form different sub-problems.
Sub-problem 1 Sub-problem 2
Max Z = Cx Max Z = Cx
Subject to, Ax  b Subject to, Ax  b
x r  x*r x r  x*r + 1
x  0. x  9.
Thus, we have branched or partitioned the original problem into two sub-
problems. Each of these sub-problems is then solved separately as LPP.
If any sub-problem yields an optimum integer solution, it is not further branched.
But if any sub-problem yields a non-integer solution, it is further branched into
two sub-problems. This branching process is continued until each problem
terminates with either an integer optimal solution or there is an evidence that it
cannot yield a better solution. The integer-valued solution among all the sub-
problems, which gives the most optimal value of the objective function is then
selected as the optimum solution.
Note: For minimization problem, the procedure is the same except that upper
bounds are used. The sub-problem is said to be fathomed and is dropped from
further consideration if it yields a value of the objective function lower than that
of the best available integer solution and it is useless to explore the problem any
further.
Self-Instructional
Material 171
Integer Programming Example 8.7: Use branch and bound technique to solve the following:
Problem
Max Z = x1 + 4x2
Subject to, 2x1 + 4x2  7
NOTES 5x1 + 3x2  15
x1, x2  0 and are integers.
Solution: Ignoring the integrality condition we solve the LPP,
Max Z = x1 + 4x2
5x1 + 3x2  15
x1, x2  0
Introducing slack variables S1, S2  0, the standard form of LPP becomes,
Max Z = x1 + 4x2 + 0S1 + 0S2
Subject to, 2x1 + 4x2 + S1 = 7
5x1 + 3x2 + S2 = 15
Since all Zj – Cj  0, an optimum solution is obtained.

x1 = 0, x2 = 7/4 and Max Z = 7
7
Sincex 2 = 4
, this problem should be branched into two sub-problems.
7
For x 2 = 4
, 1 < x2 < 2 = x2  1, x2  2
Applying these two conditions separately in the given LPP we get two sub-
problems.
Sub-problem (1) Sub-problem (2)
Max Z = x1 + 4x2 Max Z = x1 + x2
Self-Instructional
172 Material
Subject to,2x + 4x2 7 Subject to, 2x1 + 4x2 7 Integer Programming
Problem
5x1 + 3x2  15 5x1 + 3x2  15
x2  1 x 2 2
x1, x2  0 x1, x2  0 NOTES
Sub-Problem (1)
Since all Zj – Cj  the solution is optimum, given by x1 = 3/2

x2 = 1, and Max Z = 11/2
Since x1 = 3/2 is not an integer, this sub-problem is branched again.
Sub-Problem (2)
Max Z = x1 + 4x2
5x1 + 3x2  15
x2  2
x1, x2  0.
Self-Instructional
Material 173
Integer Programming
Problem
NOTES
Since all Zj – Cj  0, but an artificial variable A1 is in the basis at positive

level, there exists no feasible solution. Hence, this sub-problem is dropped.
In sub-problem (1) Since, x1 = 3/2
we have, 1  x1  2
= x1  , x1  2
Applying these two conditions separately in the sub-problem (1), we get two
sub-problems.
Max Z = x1 + 4x2 Max Z = x1 + x2
Subject to,2x1 + 4x2  7 Subject to, 2x1 + 4x2 7
5x1 + 3x2  15 5x1 + 3x2  15
x2  1 x 2 1
x1  1 x 1 
x1, x2  0. x1, x2 
Sub-Problem (3)
Since all Zj – Cj  , an optimum solution is obtained. It is given by, x1 = 1, x2
= 1 and Max Z = 5. Since this solution is integer-valued this sub-problem cannot
be branched further. The lower bound of the objective function is 5.
Self-Instructional
174 Material
Integer Programming
Problem
NOTES
Sub-Problem (4)
Max Z = x1 + 4x2
5x1 + 3x2  15
x2  1
x1  2
x1, x2  0
Since all Zj – Cj  0,the solution is optimum and is given by,
x1 = 2, x2 = 3/4
Self-Instructional
Material 175
Integer Programming
Problem
NOTES
Since x2 = 3/4 is not an integer, this sub-problem is branched further.

In sub-problem (4) since x2 = 3/4 , 0  x2  1
= x2  0, or x2  1
Applying these two conditions in the sub-problem (4)
We get two sub-problems.
Max Z = x1 + 4x2 Max Z = x1 + x2
Subject to,2x + 4x2 7 Subject to, 2x1 + 4x2 7
5x1 + 3x2  15 5x1 + 3x2  15
x2  1 x 2 1
x1  2 x1  2
x2  0 x2  1
x1, x2  0 x1, x2  0.
Self-Instructional
176 Material
Sub-Problem (5) Integer Programming
Problem
NOTES
Since all Zj – Cj  the solution is optimum and is given by x1 = 3, x2 = 0

and Max Z = 3. This sub-problem yields an optimum integer solution. Hence,
this sub-problem is dropped.
Sub-Problem (6)
Max Z = x1 + 4x2
5x1 + 3x2  15
x2  1
x1  2
x2  1
x1, x2  0
Self-Instructional
Material 177
Integer Programming This sub-problem has no feasible solution. Hence, this sub-problem is also
Problem
fathomed.
Original Problem
NOTES
Among the available integer-valued solutions, the best integer solution is given
by sub-problem (3).
The optimum integer solution is,
Max Z = 5, x1 = 1 and x2 = 1.
The best available integer optimal solution is,
Max Z = 5, x1 = 1 and x2 = 1.
Check Your Progress

1. What do you mean by integer programming problem?
2. Define a pure integer programming problem.
3. Define a mixed integer programming problem.
4. Differentiate between pure and mixed IPP.
5. Mention a few applications of IPP.
6. Name the two methods used in solving IPP.
Self-Instructional
178 Material
Integer Programming
8.3 ANSWERS TO CHECK YOUR PROGRESS Problem
QUESTIONS
1. An LPP in which some or all the variables in the optimal solution are restricted NOTES
to assume non-negative integer values is called an integer programming
problem.
2. In a LPP, if all the variables in the optimal solution are restricted to assume
non-negative integer value, then it is called a pure IPP.
3. In an LPP, if only some of the variables in the optimal solution are restricted
to assume non-negative integer values, while the remaining variables are
free to take any non-negative values, then it is called a mixed integer
programming problem.
4. In a pure IPP all the variables in the optimal solution are restricted to assume
non-negative integer values. Whereas in mixed IPP, only some of the variables
in the optional solution are restricted to assume non-negative integer values.
5. Some applications of IPP are as follows:
(i) In product mix problem
(ii) Sequencing and routing decisions
(iii) All allocation problems involving the allocation of goods, men and
machine.
6. The two methods used in solving IPP are as follows:
(i) Cutting methods (Gomany’s cutting plane algorithm)
(ii) Search method (Branch and Bound Technique)
8.4 SUMMARY
 A linear programming problem in which all or some of the decision variables

are constrained to assume non-negative integer values is called an Integer
Programming Problem (IPP).
 In a linear programming problem, if all variables are required to take integral
values then it is called the Pure (all) Integer Programming Problem (Pure
IPP).
 If only some of the variables in the optimal solution of a LPP are restricted
to assume non-negative integer values, while the remaining variables are
free to take any non-negative values, then it is called a Mixed Integer
Programming Problem (Mixed IPP).
 If all the variables in the optimal solution are allowed to take values 0 or 1,
then the problem is called the 0–1 Programming Problem or Standard
Discrete Programming Problem.
Self-Instructional
Material 179
Integer Programming  In LPP, all the decision variables were allowed to take any non-negative
Problem
real values as it is quite possible and appropriate to have fractional values in
many situations. There are several frequently occurring circumstances in
business and industry that lead to planning models involving integer-valued
NOTES variables.
 Integer programming is applied in business and industry. All assignment and
transportation problems are integer programming problems, as in the
assignment and travelling salesmen problem, all the decision variables are
either zero or one.
 There are two methods used to solve IPP, namely Gomory’s Cutting Plane
Method and Branch and Bound Method (Search Method).
 Cutting method is a systematic procedure of solving pure IPP was first
developed by R.E. Gomory, in 1956, which he later used to deal with the
more complicated case of mixed integer programming problem.
 Cutting Method consists of first solving the IPP as an ordinary LPP by
ignoring the restriction of integer values and then introducing a new constraint
to the problem such that the new set of feasible solution includes all the
original feasible integer solutions, but does not include the optimum non-
integer solution initially found. This new constraint is called ‘Fractional cut’
or ‘Gomorian constraint’.
 Search method is an enumeration method in which all feasible integer points
are enumerated. The widely used search method is the Branch and Bound
method. It was developed in 1960, by A.H. Land and A.G. Doig. This
method is applicable to both pure and mixed IPP. It first divides the feasible
region into smaller subsets that eliminate parts containing no feasible integer
solution.
 In mixed IPP only some of the variables are restricted to integer values,
while the other variables may take integer or other real values.
 Branch and Bound method is applicable to both, pure as well as mixed IPP.
Sometimes a few or all the variables of an IPP are constrained by their
upper or lower bounds. The most general method for the solution of such
constrained optimization problems is called ‘Branch and Bound method’.
 If any sub-problem yields an optimum integer solution, it is not further
branched. But if any sub-problem yields a non-integer solution, it is further
branched into two sub-problems.
 The integer-valued solution among all the sub-problems, which gives the
most optimal value of the objective function is then selected as the optimum
solution.
Self-Instructional
180 Material
Integer Programming
8.5 KEY WORDS Problem
 Integer Programming Problem (IPP): It refers to a mathematical

optimization or feasibility program in which some or all of the variables are NOTES
restricted to be integers.
 Cutting-plane method: It refers to any of a variety of optimization methods
that iteratively refine a feasible set or objective function by means of linear
inequalities, termed cuts.
 Branch and Bound Method: It refers a systematic method for solving
optimization problems.

EXERCISES

1. What are the applications of integer programming?
2. What is Standard Discrete Programming Problem?
3. Name the widely used search method and when was it developed.
4. Write an informative note on mixed integer programming problem.
5. What is branch and bound technique?
1. Explain Gomarian constraint and also explain its geometrical interpretation.
2. Why not round off the optimum values instead of resorting to integer
programming? Explain.
3. Discuss the importance of IPP.

Self-Instructional
Material 181
Infinite Queuing Models
UNIT 9 INFINITE QUEUING

MODELS
NOTES
Structure
9.0 Introduction
9.1 Objectives
9.2 Queuing Theory and Models
9.3 Operating Characteristics and Constituents of a Queueing system
9.3.4 Customer’s Behaviour
9.5 Summary
9.6 Key Words
9.0 INTRODUCTION
A flow of customers from finite/infinite population towards the service facility forms
a queue (waiting line), on account of a lack of capability to serve them all at a time.
In the absence of a perfect balance between the service facilities and the customers,
waiting time is required either for the service facilities or for the customer’s arrival.
The arriving unit that requires some service to be performed is called
customer. The customer may be persons, machines, vehicles, etc. Queue (waiting
line) stands for the number of customers waiting to be serviced. This does not
include the customer being serviced. The process or system that provides services
to the customer is termed as service channel or service facility.
In this unit, you will learn about the fundamentals of queuing theory and
models.
9.1 OBJECTIVES

 Understand the basic concepts of queuing theory and models
 Classify queuing models into finite and infinite models
 Explain the operating characteristics and constituents of a queuing system
Self-Instructional
182 Material
9.2 QUEUING THEORY AND MODELS
Queuing theory is the mathematical study of the congestion and delays of waiting
in line. Queuing theory (or “queueing theory”) examines every component of waiting NOTES
in line to be served, including the arrival process, service process, number of
servers, number of system places and the number of “customers” (which might be
people, data packets, cars, etc.). As a branch of operations research, queuing
theory can help users make informed business decisions on how to build efficient
and cost-effective workflow systems. Real-life applications of queuing theory cover
a wide range of applications, such as how to provide faster customer service,
improve traffic flow, efficiently ship orders from a warehouse and the design of
telecommunications systems, from data networks to call centres.
Queues happen when resources are limited. In fact, queues make economic
sense; no queues would equate to costly overcapacity. Queuing theory helps in
the design of balanced systems that serve customers quickly and efficiently but do
not cost too much to be sustainable. All queuing systems are broken down into the
entities queuing for an activity.
At its most elementary level, queuing theory involves the analysis of arrivals
at a facility, such as a bank or fast food restaurant, then the service requirements
of that facility, e.g., tellers or attendants. By applying queuing theory, a business
can develop more efficient queuing systems, processes, pricing mechanisms, staffing
solutions and arrival management strategies to reduce customer wait times and
increase the number of customers that can be served.
Kendall’s notation for representing Queueing Models
Generally, queueing model may be completely specified in the following symbol
form (a/b/c): (d/e) where,
a = probability law for the arrival (inter-arrival) time
b = probability law according to which the customers are being served
c = number of channels (or service stations)
d = capacity of the system, i.e., the maximum number allowed in the system
(in service and waiting)
e = queue discipline
The queueing models are classified as follows:
Model I: (M/M/1): ( /FCFS): This denotes Poisson arrival (exponential inter-
arrival), Poisson departure (exponential service time), Single server, Infinite capacity
and First come first served service discipline. The letter M is used due to Markovian
property of exponential process.
Self-Instructional
Material 183
Infinite Queuing Models Model II: (M/M/1): (N/FCFS): In this model, the capacity of the system is
limited (finite), say N. Obviously, the number of arrivals will not exceed the number
N in any case.
NOTES Model III: Multiservice Model (M/M/S):(/FCFS): This model takes the
number of service channels as S.
Model IV: (M/M/S): (N/FCFS): This model is essentially the same as model
II, except the maximum number of customers in the system is limited to N, where,
(N > S).
Of these four models, Model II and Model IV are finite models, whereas
Models I and III are infinite models. we will discuss these models in subsequent
units.
Check Your Progress

1. What is the queuing theory?
2. State some real-life applications of queuing theory.
3. How is a business benefitted by queuing theory?
9.3 OPERATING CHARACTERISTICS AND

CONSTITUENTS OF A QUEUEING SYSTEM
In daily life, customers wait in lines for different purposes such as for service at a
bank ATM counter, for purchasing rail / air tickets, submitting telephone / electricity
bills, etc. In factories, jobs wait in lines to be worked in different machines and
machines wait their turn to be overhauled, trucks wait to unload / load their cargo,
ships wait at docks for unloading / loading, etc. Planners analyse service requirements
and establish service facilities keeping in view the customers’ requirements and
satisfaction level. In other words, the ultimate goal is to achieve an economic
balance between the cost of service and the cost associated with the waiting for
that service.
The central problem in waiting line situation is a trade-off decision. The
manager weighs the added cost of providing more rapid service against the cost
of waiting. Queue discipline is a priority rule for determining the order of the service
to customers in a waiting line to have better overall performance.
A queuing system depends upon the source population and customer arrival
pattern. Arrivals in a service system may be drawn from finite population when
limited number is involved or from infinite population when a large number is
involved. Any increase or reduction does not have any effect on population size. A
constant arrival distribution is periodic with exactly the same time between arrivals.
However, random or variable distributions are common.
Self-Instructional
184 Material
Operating characteristics of a queuing system are as follows: Infinite Queuing Models
 Queue length
 Number of customers in the system
 Waiting time in a queue NOTES
 Waiting time in system
 Service facility utilization
A queueing system can be completely described by,
(i) Input (arrival pattern)
(ii) Service mechanism (service pattern)
(iii) Queue discipline
(iv) Customer’s behaviour
Input describes the way in which the customers arrive and join the system. Generally,
customers arrive in a more or less random fashion, which is not worth predicting.
Thus, the arrival pattern can be described in terms of probabilities, and consequently,
the probability distribution for inter-arrival times (the time between two successive
arrivals) must be defined. We deal with those queueing systems in which the customers
arrive in Poisson fashion. The mean arrival rate is denoted by .

This means, the arrangement of service facility to serve customers. If there is an
infinite number of servers, then all the customers are served instantaneously on
arrival, and there will be no queue.
If the number of servers is finite then the customers are served according to a
specific order, with service time as a constant or random variable. Distribution of
service time that is important in practice is the negative exponential distribution.
The mean service rate is denoted by m.
It is a rule according to which the customers are selected for service when a queue
has been formed. The most common disciplines are:
— First come first served (FCFS)
— First in first out (FIFO)
— Last in first out (LIFO)
— Selection for service in random order (SIRO).
There are various other disciplines according to which a customer is served in
preference over the other. Under priority discipline, the service is of two types,
namely pre-emptive and non-pre-emptive. In pre-emptive system, the high priority
Self-Instructional
Material 185
Infinite Queuing Models customers are given service over the low priority customers; in non-pre-emptive
system, a customer of low priority is serviced before a customer of high priority. In
the case of parallel channels ‘fastest server rule’ is adopted.
NOTES 9.3.4 Customer’s Behaviour

The customers generally behave in the following four ways:
(i) Balking A customer who leaves the queue because the queue is too
long and he has no time to wait or does not have sufficient waiting space.
(ii) Reneging This occurs when a waiting customer leaves the queue due to
impatience.
(iii) Priorities In certain applications some customers are served before
others, regardless of their arrival. These customers have priority over
others.
(iv) Jockeying Customers may jockey from one waiting line to another. This
is most common in a supermarket.
Transient and steady states A system is said to be in a transient state when its
operating characteristics are dependent on time.
A steady state system is the one in which the behaviour of the system is
independent of time. Let Pn(t) denote the probability that there are n customers in
the system, at time t. Then in steady state,
lim pn (t ) = pn (independent of t)
t
dpn (t ) dpn
 =
dt dt
 lim pn (t ) =0
t
Traffic intensity (or utilization factor) An important measure of a simple queue

is its traffic intensity given by,
Mean arrival rate
Traffic intensity  = =
Mean service rate
The unit of traffic intensity is Erlang.
Some useful suggestions for managing queues are as follows:
 Determine an acceptable waiting time of customers
 Divert customers’ attention while they are waiting
 Keep customers informed about difficulty when waiting time is longer than
normal
 Segment customers
 Train servers to be polite and friendly
 Encourage customers to come during slack periods
 Take a long-term perspective towards reduction in queues.
Self-Instructional
186 Material
Check Your Progress

4. State the factors governing a queuing system.
5. How are customers served if the number of servers is finite? NOTES
6. When is a system said to be in a transient state?

QUESTIONS
1. Queuing theory is the mathematical study of the congestion and delays of

waiting in line.
2. Real-life applications of queuing theory cover a wide range of applications,
such as how to provide faster customer service, improve traffic flow,
efficiently ship orders from a warehouse and the design of telecommunications
systems, from data networks to call centres.
3. By applying queuing theory, a business can develop more efficient queuing
systems, processes, pricing mechanisms, staffing solutions and arrival
management strategies to reduce customer wait times and increase the
number of customers that can be served.
4. A queuing system depends upon the source population and customer arrival
pattern.
5. If the number of servers is finite then the customers are served according to
a specific order, with service time as a constant or random variable.
6. A system is said to be in a transient state when its operating characteristics
are dependent on time.
9.5 SUMMARY
 As a branch of operations research, queuing theory can help users make

informed business decisions on how to build efficient and cost-effective
workflow systems.
 Queuing theory helps in the design of balanced systems that serve customers
quickly and efficiently but do not cost too much to be sustainable.
 Queuing theory involves the analysis of arrivals at a facility, such as a bank
or fast food restaurant, then the service requirements of that facility, e.g.,
tellers or attendants.
 The central problem in waiting line situation is a trade-off decision. The
manager weighs the added cost of providing more rapid service against the
cost of waiting. Self-Instructional
Material 187
Infinite Queuing Models  Arrivals in a service system may be drawn from finite population when
limited number is involved or from infinite population when a large number
is involved.
 Operating characteristics of a queuing system are as follows:
NOTES
o Queue length
o Number of customers in the system
o Waiting time in a queue
o Waiting time in system
o Service facility utilization
 Distribution of service time that is important in practice is the negative
exponential distribution. The mean service rate is denoted by m.
9.6 KEY WORDS
 Queuing theory: It examines every component of waiting in line to be

served, including the arrival process, service process, number of servers,
number of system places and the number of ‘customers’ (which might be
people, data packets, cars, etc.).
 Queue discipline: It is a priority rule for determining the order of the
service to customers in a waiting line to have better overall performance.
 Steady state system: It refers to the condition in which the behaviour of
the system is independent of time.

EXERCISES

1. State the Kendall’s notation for representing queuing models.
2. What are the operating characteristics of a queuing system?
3. Provide some useful suggestions for managing queues.
1. Classify the various queuing models and state their important characteristics.
2. Discuss the various components of a queuing system.
3. Write short notes on: (i) Customer’s behaviour, (ii) transient and steady
states and (iii) Traffic intensity.
Self-Instructional
188 Material

Singh, J. K. 2009. Business Mathematics. New Delhi: Himalaya Publishing NOTES
House.
Self-Instructional
Material 189
Mathematical Analysis
of Queuing Theory
UNIT 10 MATHEMATICAL
ANALYSIS OF
NOTES
QUEUING THEORY
Structure
10.0 Introduction
10.1 Objectives
10.2 Model I: (M/M/I) (/FCFS) (Birth and Death Model)
10.3 Model III: (Multiservice Model) (M/M/S): (/FCFS)
10.4 Properties of Queuing System
10.5 Single and Multiple Channel Queuing Model
10.6 Erlang Family of Distribution of Service Times
10.8 Summary
10.9 Key Words
10.0 INTRODUCTION
As you studied in the last unit, queuing theory is the study of queues or waiting
lines. One can derive a few analysis using queuing theory like expected waiting
time in the queue, the average time in the system, the assumed que length, the
number of customers served at one time, the probability of balking customers. In
this unit, you will be studying about the mathematical analysis of queuing theory,
properties of queuing theory, single channel queuing model and multiple channel
queuing model. In addition to this, you will also learn about the Erlang Family of
Distribution of Service Times and will examine the applications and limitations of
the queuing model.
10.1 OBJECTIVES

 Understand mathematical analysis of queuing theory
 Discuss the properties of queuing system
 Analyse the single channel queuing model and multiple channel queuing model
 Examine the Erlang Family of Distribution of Service Times
 Discuss the applications and limitations of queuing model
Self-Instructional
190 Material
Mathematical Analysis
10.2 MODEL I: (M/M/I) (/FCFS) (BIRTH AND of Queuing Theory
DEATH MODEL)
To obtain the steady state equations: The probability that there will be n units NOTES
(n > 0) in the system at time (t + t), may be expressed as the sum of three
independent compound probabilities by using the fundamental properties of
probability, Poisson arrivals and exponential service times.
The following are the three cases:
Time (t) Arrival Service Time (t + t)
No. of units No. of units
n–1 0 0 n
n–1 1 0 n
n+1 0 1 n
Now, by adding the above three independent compound probabilities, we obtain

the probability of n units in the system at time (t + t).
Pn(t + t) = Pn(t) (1 – ( + ) t) + Pn–1(t) t + Pn+1(t)t + 0(t)
Pn (t  t )  Pn (t ) 0( t )
= – ( + ) Pn(t) + Pn–1(t) + Pn+1(t) +
t t
Pn (t  t )  Pn (t )  0(t )
Lt = Lt  (  ) Pn (t )  Pn 1(t )  Pn 1(t ) 
t  0 t t  0  t
dPn (t )
= – ( + )Pn(t) + Pn–1(t) + Pn+1(t)
dt
 0 (t ) 
where, n > 0  Lt  0
 t  0 t 
In the steady state,
Pn(t)  0, Pn(t) = Pn
0= –( + )Pn + Pn–1 + Pn+1

(1)
In a similar fashion, the probability that there will be n units (i.e., n = 0) in the
system at time (t +t), will be the sum of the following two independent
probabilities.
(i) Probability (that there is no unit in the system at time t and no arrival in
time t)
= P0(t) (1 – t)
(ii) Probability (that there is one unit in the system at time t, one unit serviced in
t and no arrival in t)
= Pi(t)t (1 – t)
= Pi(t)t – 0(t)
Self-Instructional
Material 191
Mathematical Analysis Adding these two probabilities,
of Queuing Theory
P0(t – t) = P0(t)(1 – t) + P1(t)t + 0 (t)
P0 (t t ) P0 (t ) 0( t )
= –P0(t) + P1(t) +
NOTES t t
P0 (t t ) P0 (t )
Lt = –P0(t) + P1(t) for n = 0
t 0 t
dP0 (t )
= –P0(t) + P1(t)
d (t )
Under steady state, we have,
0 = –P0 + P1 (2)
Equations (1) and (2) are called steady state difference equations for this
model.
From (2), P1 = P0
2

From (1), P2 = P1 =   P0


Generally, Pn =    P
  0

Since, Pn = 
n 0
2
 
 P0 + P0    P0 + ... = 1
 
2
P0 1 ... = 1
1
i.e., P0 =1
1
Since, < 1, sum of infinite G.P. is valid.
P0 = 1 – =1–


Also, Pn =   P0 = 1

Pn = n(1 – )
Self-Instructional
192 Material
Measures of Model I Mathematical Analysis
of Queuing Theory
1. Expected (average) number of units in the system LS.
LS = nPn
n 1
NOTES
n
= n 1
n 1
n 1
= 1 n
n 1
2
= 1 1 2 3 ...
2
= 1 1
= = ,= < 1.
1
1
LS =
1
2. Expected (average) queue length Lq .

2
Lq = Ls –
1
Lq =
1
3. Expected waiting line in the queue,
Wq =
( ) 1
4. Expected waiting line in the system,
1
WS = Wq +
1 1
=
( )
1
WS =
5. Expected waiting time of a customer who has to wait (w/w > 0)
1 1
(W/W > 0) =
(1 )
Self-Instructional
Material 193
Mathematical Analysis 6. Expected length of non-empty queue,
of Queuing Theory
1
(L/L > 0) =
1
N
7. Probability of queue size  N = 
NOTES
( )
= ( )e d
t
8. Probability of waiting time in the queue  t
9. Probability (waiting time in the system  t)
( )
= ( )e d
t
10. Traffic intensity,  =
Inter-relationship between LS, Lq, WS, Wq (Littles formula)

We know,
LS =
1
1
WS =
 L s = Ws
Similarly, Lq = Wq hold in general,

Wq =
( )
1
WS =
1 1 1 ( )
 WS – = – = =
( ) ( )
1
 Wq = WS –
Multiplying both sides by , we have,
 1
Wq =   WS  
 
Lq =  WS – = LS –
Lq = LS –
Self-Instructional
194 Material
Example 10.1: A T.V. mechanic finds that the time spent on his jobs has an Mathematical Analysis
of Queuing Theory
exponential distribution with mean 30 minutes, if he repairs sets in the order in
which they come in. If the arrival of sets is approximately Poisson with an average
rate of 10 per eight-hour day, what is the mechanic’s expected idle time each day?
How many jobs are ahead of the average set just brought in? NOTES
Solution:
10 1
Here,  = 1/30,  =
8 60 48
Expected number of jobs are,
LS =
1
1
= 48 = 1 2/3 jobs.
1/ 30 1/ 48

Since the fraction of the time the machanic is busy equals to  ’ the number of
hours for which the repairman remains busy in an eight-hour day,

30
=8 =8× = 5 hours
48
Therefore, the time for which the machanic remains idle in an eight-hour day =
(8 – 5) hours = 3 hours.
Example 10.2: At what average rate must a clerk at a supermarket work, in

order to insure a probability of 0.90 that the customers will not have to wait longer
than 12 minutes? It is assumed that there is only one counter, to which customers
arrive in a Poisson fashion at an average rate of 15 per hour. The length of service
by the clerk has an exponential distribution.
Solution:
15 1
Here, = = customer/minute  = ?
60 4
Prob. (waiting time  12) = 1 – 0.9 = 0.10

 
   1    e
(  ) 
d  = 0.1
12

  e ( )  
 1     = 0.1
    (  ) 12
 12(
(e ) = 0.10

e(3–12) = 0.4
1
= 2.48 minutes per service. Self-Instructional
Material 195
Mathematical Analysis Example 10.3: Arrivals at a telephone booth are considered to be Poisson, with
of Queuing Theory
an average time of 10 minutes between one arrival and the next. The length of a
phone call is assumed to be distributed exponentially with mean three minutes.
NOTES (i) What is the probability that a person arriving at the booth will have to
wait?
(ii) What is the average length of the queue that forms from time to time?
(iii) The telephone department will install a second booth when convinced that an
arrival would have to wait at least three minutes for the phone. By how much
must the flow of arrivals be increased in order to justify a second booth?
Solution:
Given,  = 1/10,  = 1/3
(i) Probability (w > 0)= 1 – P0 = = 1 3

= 3/10 = 0.3
10 1
 1/3
(ii) (L/L > 0)= = = 1.43 persons
 1/3  1/10
(iii) Wq =
( )
1
Since, Wq = 3,  = ,  =  for second booth,
3
3=  = 0.16
1 1
3 3
Hence, increase in the arrival rate = 0.16 – 0.10 = 0.06 arrival per minute.
Example 10.4: As in example 10.3, in a telephone booth with Poisson arrivals

spaced 10 minutes apart on the average and exponential call length averaging
three minutes.
(i) What is the probability that an arrival will have to wait for more than 10
minutes before the phone becomes free?
(ii) What is the probability that it will take him more than 10 minutes in total to wait
for the phone and complete his call?
(iii) Estimate the fraction of a day that the phone will be in use.
(iv) Find the average number of units in the system.
Solution: Given, n = 0.1 arrival/minute
 = 0.33 service/minute
(i) Probability (waiting time  10) = 1 e ( )

d
10

 ( )
10

=– e

= 0.3 e–2.3 = 0.03
Self-Instructional
196 Material
(ii) Probability (waiting time in the system  10) Mathematical Analysis
of Queuing Theory
= ( )e ( d
10
= e–10( – ) = e–2.3 = 10 NOTES
(iii) The fraction of a day that the phone will be busy = traffic intensity
= = 0.3.
(iv) Average number of units in the system,
1
10
LS = = 1 1 = 3/7 = 0.43 customer..
3 10
Example 10.5: Customers arrive at a one-window drive-in bank according to

Poisson distribution with mean 10 per hour. Service time per customer is exponential
with mean five minutes. The space in front of the window including that for the
serviced car can accommodate a maximum of three cars. Others can wait outside
this space.
(i) What is the probability that an arriving customer can drive directly to the
space in front of the window?
(ii) What is the probability that an arriving customer will have to wait outside
the indicated space?
(iii) How long is an arriving customer expected to wait before starting service?
Solution: Given,  = 10 per hour
1
= × 60 = 12 per hour
5
10
=
12
(i) The probability that an arriving customer can drive directly to the space in
front of the window,
2
 
P0 + P1 + P2 = P0 + P0    P0 or 1 – Px
 
2
= P0 1
2
= 1 1  P0 = 1 –
10 10 100
= 1 1 = 0.42
12 12 144
Self-Instructional
Material 197
Mathematical Analysis (ii) Probability that an arriving customer will have to wait outside the indicated
of Queuing Theory
space,
S = 1 – 0.42 = 0.58
NOTES (iii) Average waiting time of a customer in a queue,
1 10 1 5
= = =
12 12 10 12
= 0.417 hours.
Example 10.6: In a supermarket, the average arrival rate of customers is 10 every

30 minutes, following Poisson process. The average time taken by a cashier to list
and calculate the customer’s purchase is two and a half minutes following exponential
distribution. What is the probability that the queue length exceeds six?
What is the expected time spent by a customer in the system?
Solution:
10
= per minute
30
1
= per minute
2.5
1
3
= = 0.8333
1/ 2.5
(i) The probability of queue size > 6 = 6
1
WS = = (0.8333)6 = 0.3348.
–
LS 1 0.833
(ii) WS = = = 3
 1 0.8333
= 14.96 minutes.
Example 10.7: On an average, 96 patients per 24-hour day require the service
of an emergency clinic. Also, on an average a patient requires 10 minutes of active
attention. Assume that the facility can handle only one emergency at a time. Suppose
that it costs the clinic 100 per patient treated, to obtain an average servicing time
of 10 minutes and thus, each minute of decrease in this average time would cost
10 per patient treated. How much would have to be budgeted by the clinic to
decrease the average size of the queue from 1 13 patients to 12 patients?
Solution:
96 1
Given, = = patient/minute
24 60 15
1
= patient/minute
10
Self-Instructional
198 Material
Average number of patients in the queue, Mathematical Analysis
of Queuing Theory
Lq =
2
1 NOTES
15 1
= 1 1 1 = 1 patients
3
10 15 10
1
But, Lq = 1 is reduced to Lq = 1/2
3
1
 Substituting Lq = 1/2,  =  = in the formula
15
12
Lq =
( )
2
1
1 15
=  = 2/15 patients/minute
2 ( 1/15)
1
Hence, the average rate of treatment required is,  7.5 minutes. Decrease in

time required by each patient
15 5
= 10 – minutes
2 2
 The budget required for each patient
5
= 100 + × 10 = 125
2
So, in order to get the required size of the queue, the budget should be increased
from 100 to 125 per patient.
Example 10.8: In a public telephone booth, the arrivals on an average are 15 per
hour. A call on an average takes three minutes. If there is just one phone, find
(i) the expected number of callers in the booth at any time (ii) the proportion of the
time, the booth is expected to be idle?
Solution:
Given,  = 15 per hour
1
= × 60 per hour
3
(i) Expected length of the non-empty queue
20
= 4
20 15
15 3
(ii) The service is busy =
20 4
 the booth is expected to be idle for 1 – 3 1 hours = 15 minutes.

4 4 Self-Instructional
Material 199
Mathematical Analysis Example 10.9: In a railway marshalling yard, goods trains arrive at a rate of 30 trains
of Queuing Theory
per day. Assuming that inter-arrival time and service time distribution follows an
exponential distribution with an average of 30 minutes, calculate the following.
NOTES (i) The mean queue size.

(ii) The probability that queue size exceeds 10.
(iii) If the input of the train increases to an average of 33 per day, what will be the
changes in (i) and (ii)?
Solution:
30 1
Given, = trains/minute
60 24 48
1
= trains/minute
30
 30 5
=  
 48 8
 5/8 5/8 5
(i) LS =    = 1.66 trains = 2 trains (app.)
1   1  5/8 5/8 3
(ii) P ( 10) = (0.75)10 = 0.056

(iii) When the input increases to 33 trains per day,
33 1 1
we have, = ,= trains/min
60 24 480 30
11
 LS = where, = =
1 16
11/16 11
LS =  = 2.1 trains
5/16 5
10
( 10) = 10 =   = 0.1460.
11
also,
 16 
10.3 MODEL III: (MULTISERVICE MODEL)

(M/M/S): (/FCFS)
When there are n units in the system, difference equations may be obtained in the
following two situations.
(i) if n  s, all the customers may be served simultaneously. There will be no
queue. (s – n) number of servers may remain idle and then,
n = n, n = 0, 1, 2, ... S;
(ii) if n  s, all the servers are busy, and the maximum number of customers
waiting in queue will be (n – s), then, n = s
Self-Instructional
200 Material
Also, n = [n = 0, 1, 2 ... ] Mathematical Analysis
of Queuing Theory
The steady state difference equations are,
P0(t + t) = P0(t)(1 – t) + P1(t)t + 0(t)
for n = 0 NOTES
Pn(t + t) = Pn(t)[1 – ( + n)t] + Pn–1(t)t
+ Pn+1(t)(n + 1)t + 0(t) for n = 1, 2 ... S – 1
Pn(t + t) = Pn(t)[1 – ( + S)t] + Pn–1(t)t
+ Pn+1(t)St + 0(t) for n = S, S + 1, S + 2 ...
Now, dividing these equations by t and taking limit as t0, the difference
equations are,
P01(t) = –P0(t) + P1(t) for n = 0
Pn1(t) = –( + n)Pn(t) + Pn–1(t) + (n + 1)Pn+1(t)
for, n = 1, 2 ... S – 1.
Pn1(t) = –( + S)Pn(t) + Pn–1(t) + SPn+1(t)
for, n = S, S + 1, S + 2 ...
Considering the case of steady state, i.e., when t  , Pn(t)  Pn and hence,
1
Pn (t)for all n, above equations become,
O = –0 + P1 for n = 0 (1)
O = –( + n)Pn + Pn–1 + (n + 1)Pn+1 (2)
for, 1  n  S – 1
O = –( + S)Pn + Pn–1 + SPn+1 for n  S (3)
Here, P0 = P0 initially,
P1 = P0 from (1)
2
P2 = P1 = P0 (Put n = 1 in (2))
2 2! 2
3
P3 = P2 = 3
P0 (Put n = 2 in (2))
3 3!
n
In general, Pn = Pn–1 = 1 P0
n n!
1nS
S
1
PS = PS–1 = P0
S S!
Self-Instructional
Material 201
Mathematical Analysis S 1
of Queuing Theory 1 1 
PS+1 = PS =    P0
S S S !  
S 2
1 1 
PS+2 = PS+1 =   . P0
NOTES S S 2 S !  
n
1 1
In general, Pn = PS+(n–S) = P0 n S
Sn S S!
Now, find Pn using the fact,

 Pn =1
n 0
S 1 
i.e.,  Pn   Pn = 1
n 0 nS
This gives the steady state distribution of arrivals (n) as,

n
1
P0 , n 0, 1, 2 ... S 1
n!
Pn = n
1 1
P0 , if n S , S 1...
S! Sn S
Measures of Model III

1. Length of the queue,
P
Lq = Ps
(1 P)2
S
P0
where, PS =
S!
2. Length of the system,
LS = + Lq
3. Waiting time in the queue,

Lq
Wq =
4. Waiting time in the system,
Ls
WS =
5. The mean number of individuals who actually wait is given by,

1
L(L > 0) =
1
Self-Instructional
202 Material
6. The mean waiting time in the queue for those who actually wait is given by Mathematical Analysis
of Queuing Theory
W(W > 0),
1
=
S
NOTES
Ps
7. Probability (W > 0) =
1
Ps 
8. Probability that there will be someone waiting =
1 
9. Average number of idle servers = S – (average number of customers served)
average number of customers served
10. Efficiency of M/M/S model =
total number of customers served
Example 10.10: A supermarket has two girls ringing up sales at the counters. If
the service time for each customer is exponential with mean four minutes and if
people arrive in a Poisson fashion at the counter, at the rate of 10 per hour, then
calculate,
(i) the probability of having to wait for service.
(ii) the expected percentage of idle time for each girl.
(iii) if a customer has to wait, find the expected length of his waiting time.
Solution:
(i) Probability of having to wait for service is,
S
P(W > 0) = P0
S !(1 )
1
= ,  = 1/4, S = 2
6
1
= =
s 3
1
 S 1 ( S )n ( S ) S 
0 =    
 n0 n ! S !(1  ) 
1
  1 n  1 
2
 1 2   2   
=    3    3  
n! 2!(1  1/ 3)
 n0 
 
4
9 1
= 1 2/3
2 2/3 2
2 1
(4 / 6) 
2  1/ 6
Thus, probability (W > 0) =
2!(1  1/ 3) Self-Instructional
Material 203
Mathematical Analysis (ii) The fraction of the time the service remains busy,
of Queuing Theory
1
(i.e., traffic intensity) is given by,  =
S 3
NOTES  The fraction of the time the service remains idle is,
1 2
1 = 67% (approximately)
3 3
1 1 1 1
(iii) (W/W > 0) =    = 3 minutes.
1   S 1  1 2  1
3 4
Example 10.11: A petrol station has two pumps. The service time follows the
exponential distribution with mean four minutes and cars arrive for service in a
Poisson process at the rate of 10 cars per hour. Find the probability that a customer
has to wait for service. What proportion of time do the pumps remain idle?
Solution:
Given, S = 2,  = 10 per hour
1 60
= per minute = = 15 per hour
4 4
10 1
=
S 2 15 3
S
PS
(i) Probability (W > 0) = , PS P0
1 S!
S 1 1
( S )n ( S ) S
where, P0 =
n 0
n ! S !(1 )
1 1
( S ) n ( S )2
=
n 0
n ! 2!(1 )
1
( S )1 ( S )2
= 1
1! 2!(1 )
1
1
2
1 3 1
= 1 2
3 1 2
2! 1
3
Self-Instructional
204 Material
S Mathematical Analysis
of Queuing Theory
PS 1
P(W > 0) = P0
1 1 S!
NOTES
  2 2 1 
 
1   3  2  1 3
=    0.167 (approximately) .
1 
1   2!  9 2
3
 
1
(ii) The duration of time for which the pumps are busy =
S 3
 The duration of time for which the pumps remain idle
1 2
=1– = = 67% (approximately)
3 3
Example 10.12: A telephone exchange has two long distance operators. The
telephone company finds that during the peak load, long-distance calls arrive in a
Poisson fashion at an average rate of 15 per hour. The length of service on these
calls is approximately distributed with mean length five minutes.
(i) What is the probability that a subscriber will have to wait for his long-
istance call during the peak hours of the day?
(ii) If the subscribers will wait and be serviced in turn, what is the expected
waiting time?
Establish the formula used.
Solution:
15 1 1
Here, = , ,S 2
60 4 5
5
 =
S 8
S 1 1
( S )n (S )S
First, calculate P0 =
n 0
n! S !(1 )
1 1
= (5 / 4) n (5 / 4)2
n 0
n! 2!(1 5 / 8)
1
= 1
3 /13
1 5 / 4 (5 / 4) 2 8/3
2
Self-Instructional
Material 205
Mathematical Analysis S
of Queuing Theory
(i) P(W > 0) = P0

S !(1 )
2
NOTES 5
   3 /13
4
=    25/52  0.48
2!(1  5 / 8)
(ii) Wq = Lq/
1 (S )S
= P0
S !(1 )2
5 / 8 (5 / 4)2 3 125
=4 3.2 minutes.
2!(1 5 / 8)2 13 39
Example 10.13: Four counters are being run on the frontier of a country to check
the passports and necessary papers of the tourists. The tourists choose any counter
at random. If the arrival at the frontier is Poisson at the rate  and the service time
is exponential with parameter /2, what is the steady state average queue at each
counter?
Solution:
Here, S = 4,  = ,  = /2
1
 =
S 2
4 1
1
3
2n 4 4
2
P0 =
n 0 n! 4! 1 1/ 2
1
20 21 22 23 256 1
=
0! 1! 2! 3! 24 8
= [1 + 2 + 2 + 8/6 + 4/3]–1 = [5 + 8/3]–1 = 3/23
But, the expected queue length q is,
1
( /  ) S  24 2 3
= P0    4 / 23
S ! 1  2 4!  1  2 23
 
2
10.4 PROPERTIES OF QUEUING SYSTEM
The essential characteristics or properties of a queuing system are as follows:

 Arrival Pattern: It expresses the way of arrival of customers at the service
facility. These arrivals may occur in batches as opposed to one at a time.
Self-Instructional
206 Material
 Service Pattern: This mechanism describes the way in which service is Mathematical Analysis
of Queuing Theory
being provided. The units are served either singly or in batches. Service
time is a term used for the time required for servicing a unit or a batch.
 Queue Discipline: It refers to the mode which is used to select the
NOTES
customers for service after the formation of queue. First in first out (FIFO)
and last in first out (LIFO) are the common discipline.
 Service Channels: There may be several service channels in a queuing
system to provide service. The arrangement of these service channels may
be parallel or in series or both. The arrangement is basically depend on the
design of the system’s service mechanism.
10.5 SINGLE AND MULTIPLE CHANNEL QUEUING

MODEL
Single Channel Queuing Model

It is the one of the naivest queuing model and also the common one. Following
assumptions are made in the model:
 In this model, the arriving customers are served as First in First Out (FIFO)
basis.
 The rate of arrival is constant and does not change with time.
 Arrivals follow Poisson’s distribution and are not of infinite population.
 All the customers are served as per their different service time requirements
and are independent of each other.
 Negative exponential probability distribution is used to describe service
time.
Multiple Channel Queuing Model
The multiple channel queuing model exhibits a variety of queue disciplines. In this
model, the possibilities of unusual customer behaviour is realized frequently. Classical
multiple-channel models primarily undertake the operation of a FIFO rule, as
compared to LIFO and SIRO rules. They follow the principle that all arriving
customers would join a single queue and this queue would serve all service-channels
with customers being drawn from it according to some clearly defined rule of the
FIFO, LIFO or SIRO type.
10.6 ERLANG FAMILY OF DISTRIBUTION OF

SERVICE TIMES
There are number of phases comprised in many queuing models. For instance, a
number of sequential steps are required to repair or revamp a certain machine.
Self-Instructional
Material 207
Mathematical Analysis The time taken at each phase of repair is a random variable following an exponential
of Queuing Theory
distribution with parameter . Let there be k phases, the service time of the ith
phase being Si ( i = 1,2, . . ., k ). The total service time S can be expressed as
S = S1 + S2 + … + Sk.
NOTES
It is also assumed that Si (i = 1,2, . . ., k ) are independently distributed.
Hence, the probability distribution of S is a Gamma distribution with parameter
k and p. The Gamma distribution is also a member of Erlang family of distribution
and is denoted by Ek when k exponential variables are involved.

 The queuing theory techniques can help in determining suitable number and
type of service facilities to be provided to different types of customers.
 Queuing theory is helpful in obtaining complete details concerning setting
up workstations, requirement of manpower and the people who would visit
that domain. Though, all these details are based on the concept of probability.
 The theory helps in developing a scientific acknowledgement of the problem
which further helps in making optimal usage of available facilities which also
reduces the total waiting time.
Although, the queuing theory offers a scientific method of understanding the queues
and solving the queues’ problems, still the theory has certain limitations which are
stated as follows:
 Mathematical distributions are only a close approximation of the behaviour
of customers, the time between their arrival and service time is required by
each customer.
 Many real life queuing problems are complex situation are very difficult to
use the queuing theory technique.
Self-Instructional
208 Material
 In industry and service, many situations are multi-channel queuing problems. Mathematical Analysis
of Queuing Theory
When a customer has been attended to and the service provided, it may still
have to get some other service from another service and may have to fall in
queue once again. Here, the departure of one channel queue becomes the
arrival of the channel queue. In such situations, the problem becomes still NOTES
more difficult to analyse.
Check Your Progress

1. What is arrival pattern?
2. How the units are served in the service pattern?
3. What do you understand by service channels?

QUESTIONS
1. It expresses the way of arrival of customers at the service facility. These

arrivals may occur in batches as opposed to one at a time.
2. In service pattern, the units are served either singly or in batches. Service
time is a term used for the time required for servicing a unit or a batch.
3. There may be several service channels in a queuing system to provide service.
The arrangement of these service channels may be parallel or in series or
both. The arrangement is basically depend on the design of the system’s
service mechanism.
10.8 SUMMARY
 Arrival pattern, service pattern, queue discipline, service channels etc. are
some of the properties of queuing system.
 Single channel queuing model is the one of the naivest queuing model and
also the common one.
 The multiple channel queuing model exhibits a variety of queue disciplines.
In this model, the possibilities of unusual customer behaviour is realized
frequently. Classical multiple-channel models primarily undertake the
operation of a FIFO rule, as compared to LIFO and SIRO rules.
 Multiple channel queuing model follow the principle that all arriving customers
would join a single queue and this queue would serve all service-channels
with customers being drawn from it according to some clearly defined rule
of the FIFO, LIFO or SIRO type.
Self-Instructional
Material 209
Mathematical Analysis  The queuing theory techniques can help in determining suitable number and
of Queuing Theory
type of service facilities to be provided to different types of customers.
 Queuing theory helps in developing a scientific acknowledgement of the
problem which further helps in making optimal usage of available facilities
NOTES
which also reduces the total waiting time.
10.9 KEY WORDS
 Queuing Theory: It refers to the mathematical study of waiting lines or

queues.
 Queue Discipline: It refers to the rules a company or other organization
follows to process orders, inventory, or other things that it receives.
 Erlang Distribution: It refers to a two-parameter family of the continuous
probability distributions with support.

EXERCISES

1. Write a brief note on single channel queuing model?
2. What is queue discipline?
3. Give an overview of multiple channel queuing model.
4. What is Model I and Model III of the queuing model?
1. Discuss the properties of queuing model.
2. What assumptions are made in single channel queuing model?
3. Analyse the Erlang Family of Distribution of Service Times.
4. What are the applications and limitations of the queuing theory? Discuss.

House.
Self-Instructional Vikas Publishing House Pvt. Ltd.
210 Material
Finite Queuing Models
UNIT 11 FINITE QUEUING MODELS

Structure NOTES
11.0 Introduction
11.1 Objectives
11.2 Model II (M/M/1): (N/FCFS)
11.3 Model IV (M/M/S): (M/FCFS)
11.4 Methodology to Simulation
11.6 Summary
11.7 Key Words
11.0 INTRODUCTION
Finite queuing system is the combination of the queue and the service channels.
Under this model, it is assumed that the total number of customers that can be
present in the system is limited to some maximum number. That is to say that the
size of the queue is finite and it is limited by the maximum number in the system less
than the number of servers. In this unit, you will study about finite queuing models
and the concept of simulation.
11.1 OBJECTIVES
 Understand the measures of finite queuing models
 State the concept of simulation
 Discuss the types and examples of simulation
 Summarize the concept of random variable
11.2 MODEL II (M/M/1): (N/FCFS)

This model differs from model I in the sense that, the maximum number of customers
in the system is limited to N. Therefore, the difference equations of model I are
valid for this model as long as n < N. Arrivals will not exceed N in any case. The
various measures of this model are,
1
1. P0 = N 1
where,  = 1 is allowed
1
1 n
2. PN = N 1
, for n = 0, 1, 2, ... N
1 Self-Instructional
Material 211
Finite Queuing Models 3. Lq = /
N
n
4. LS = 0 n
n 0
NOTES 
5. Lq = Ls 

6. WS = LS /
 Wq = Lq/
Example 11.1: In a railway marshalling yard, goods trains arrive at the rate of 30
trains per day. Assume that the inter-arrival time follows an exponential distribution
and the service time is also to be assumed as exponential with mean of 36
minutes.Calculate,
(i) the probability that the yard is empty.
(ii) the average queue length, assuming that the line capacity of the yard is
nine trains.
Solution: for model II
30 1 1
Given =   trains per minute
60  24 48 36
36
= 0.75
48
(i) The probability that the queue is empty is given by,
1 
P0 = where N = 9
1  N 1
1 0.75 0.25
= 9 1
0.28.
1 (0.75) 0.90
(ii) Average queue length is given by,

N
1 n
Ls = N 1
n
1 n 0
9
1 0.75
= n(0.75)n
1 (0.75)10 n 0
= 0.28 × 9.58 = 3 trains.
11.3 MODEL IV (M/M/S): (M/FCFS)
This model is essentially the same as model III, except that the maximum number
of customers in the system is limited to N, where N > S(S = Number of channels)
, 0 n N
 n =
0, n N
Self-Instructional
212 Material
n ,0 n S
n =
S ,S n N
 1   n 
   P0 , 0  n  S 
n     NOTES
Pn =  n 
 1  
 n  S    P0 , S  n  N 
S S !  
n n 1
S 1 N
1 1
P0 =
n 0 n! n C Sn S
S!
( S )C N S 1 N S
Lq = 2
[1 (1 )( N S 1) ]P0
S !(1 )
S 1
( S n)( S )n
LS = Lq + S – P0
n 0 n!
LS
WS = , where  = (1 – PN)

1
Wq = WS –
Example 11.2: A barber shop has two barbers and three chairs for customers.
Assume that the customers arrive in a Poisson fashion at a rate of five per hour and
each barber services customers according to an exponential distribution with mean
15 minutes. Further, if a customer arrives and there are no empty chairs in the
shop, he will leave. What is the expected number of customers in the shop?
Solution:
5 1
Here, S = 2, N = 3,  = customer/min
60 12
1
= / min
15
n n 1
2 1 3
1 1 1
P0 = n 2
n 0 !
n n 22 2!
1
 5 1 5
2
1 5 
3
= 1  1.       
 4 2!  4  2.2!  4  
1
5 1 1
= 1 (5 / 4) 2 (5 / 4)3
4 2 4
256
= 0.28
901
Self-Instructional
Material 213
1 n
 n ! (5 / 4)  0.28 0n2
Pn =  1
 (5 / 4) n  0.28 2  n  3
 2n  2 2!
NOTES 1
= (1.25) n 0.28 0n<2
n!
1
= n 2
(1.25)n 0.28 2n3
2 2!
n
S 1 ( S n)
LS = Lq + S – P0
n 0 n!
3 2 1
(2 n)(1.25)n
= (n 2) Pn 2 P0
n 2 n 0 n!
= P3 + 2 – 3.2 P0
1
= (1.25)3 0.28 2 3.2 0.28 1.226 customers.
2.2!
11.4 METHODOLOGY TO SIMULATION
Simulation is a representation of reality through the use of a model or other device,

which will react in the same manner as reality under a given set of conditions.
Simulation is also defined as the use of a system model that has the designed
characteristics of reality, in order to produce the essence of an actual operation.
Simulation is mainly of two types.
(i) Analog (environmental) simulation
(ii) Computer (system) simulation
Some examples of simulation models are given below:
(i) Testing an aircraft model in a wind tunnel.
(ii) Children cycling in a park with various signals and crossings—to model a
traffic system.
(iii) Planetorium.
To determine the behaviour of a real system in actual environment, a number
of experiments are performed on simulated models either in the laboratories or on
the computer itself.
Self-Instructional
214 Material
Random Variable Finite Queuing Models
The random variable is a real-valued function, defined over a sample space associated
with the outcome of a conceptual chance experiment. Random variables are classified
according to their probability density function.
NOTES
Random number: It refers to a uniform random variable or a numerical value
assigned to a random variable, following uniform probability density function. In
other words, it is a number in a sequence of numbers, whose probability of
occurrence is the same as that of any other number in that sequence.
Pseudo-random numbers: Random numbers are called pseudo-random
numbers when they are generated by some deterministic process, but have already
qualified the pre-determined statistical test for randomness.
The methodology, procedure and application of simulation have been
discussed in the next unit.
Check Your Progress

1. What are the two types of simulation?
2. What do you mean by random variable?
3. Define random number.
4. Which numbers are called pseudo-random numbers?

QUESTIONS
1. Simulation is mainly of two types: analog (environmental) simulation and

computer (system) simulation.
2. The random variable is a real-valued function, defined over a sample space
associated with the outcome of a conceptual chance experiment. Random
variables are classified according to their probability density function.
3. Random number is a number whose probability of occurrence is the same
as that of any other number in the collection.
4. Random numbers are called pseudo-random numbers when they are
generated by some deterministic process, but have already qualified the
pre-determined statistical test for randomness.
11.6 SUMMARY
 Model II (M/M/1) differs from model I in the sense that, the maximum
number of customers in the system is limited to N. Therefore, the difference
Self-Instructional
Material 215
Finite Queuing Models equations of model I are valid for this model as long as n < N. Arrivals will
not exceed N in any case.
 The various measures of model II are as follows:
1    
NOTES 1. P0 = , where  =  1 is allowed 
1  N 1    
1 
2. PN = N 1
 n for n = 0 1, 2, ..., N.
1 

3. Lq =

N
4. LS = 0  n n
n 0

5. Lq = LS –

6. WS = LS /
7. Wq = Lq /
 Model IV is essentially the same as model II, except that the maximum
number of customers in the system is limited to N, where N>S (S=Number
of channels).
 The various measures of model IV are as follows:
 , 0  n  N
1. n = 
0, n  N
 n , 0  n  S
2. n = 
S , S  n  N
 1   n 1 
n
3. Pn =    P0 , 0  n  S n  S S !   P0 , S  n  N
 n    S 
1
 s 1 1   n N
1  
n
4. P0 =       n  s   
 n  0 n !    nC S S !    
( S  )c  
5. Lq = 2 
1   N  S  1  (1   )( N  S  1)  N  S  P0
S !(1   ) 
S 1
( S  n) ( SP) n
6. LS = Lq + S – P0  n!
n 0
LS
7. wS = where   =  (1– PN)

1
8. wq = wS –

 Simulation is a representation of reality through the use of a model or other
device, which will react in the same manner as reality under a given set of
conditions.
Self-Instructional
216 Material
 Simulation is also defined as the use of a system model that has the designed Finite Queuing Models
characteristics of reality, in order to produce the essence of an actual

operation.
 Simulation is mainly of two types: analog (environmental) simulation and
NOTES
computer (system) simulation.
 To determine the behaviour of a real system in actual environment, a number
of experiments are performed on simulated models either in the laboratories
or on the computer itself.
 The random variable is a real-valued function, defined over a sample space
associated with the outcome of a conceptual chance experiment. Random
variables are classified according to their probability density function.
 Random number refers to a uniform random variable or a numerical value
assigned to a random variable, following uniform probability density function.
In other words, it is a number in a sequence of numbers, whose probability
of occurrence is the same as that of any other number in that sequence.
 Random numbers are called pseudo-random numbers when they are
generated by some deterministic process, but have already qualified the
pre-determined statistical test for randomness.
11.7 KEY WORDS
 Simulation: It refers to an approximate imitation of the operation pf a

process or system.
 Pseudo-random numbers: These numbers are generated by computers.
These are not truly random, they just follow certain deterministic formula
and thence it looks random.
 Analog Simulation: It refers to the representation of physical systems and
phenomena by variables such as translation, rotation, resistance, and voltage.
 Computer Simulation: It refers to a computer program that attempts
to simulate an abstract model of a particular system.

EXERCISES

1. Define simulation.
2. Give some examples of simulation models.
3. List the various measures used in Model II.
Self-Instructional
Material 217
Finite Queuing Models Long Answer Questions
1. Let there be an automobile inspection site with three inspection stalls. Assume
that cars wait in such a way that when a stall becomes vacant, the car at the
NOTES head of the pulls up to it. The station can accommodate four cars at the
most, to be waiting (seven in station) at one time. The arrival pattern is
Poisson with a mean of one car every minute during peak hours. The service
time is exponential with mean six minutes. Find the average number of cars
in the site during peak hours, the average waiting time and the average
number of cars per hour that cannot enter the station because of full capacity.
2. Customers arrive at a one-window drive-in bank according to Poisson
distribution with mean 10 per hour. Service tome per customer is exponential
with mean 5 minutes. The car space in front of the window, including that
for the service, can accommodate a maximum of 3 cars. Other cars can
wait outside this space.
(a) What is the probability that an arriving customer can drive directly to
the space in front of the window?
(b) What is the probability that an arriving customer can drive directly to
the space in front of the window?
(c) How long is an arriving customer expected to wait before starting the
service?

House.
Self-Instructional
218 Material
Simulation: Monte Carlo
Method
BLOCK - III
SIMULATION GAME THEORY CONCEPTS
NOTES
UNIT 12 SIMULATION: MONTE
CARLO METHOD
Structure
12.0 Introduction
12.1 Objectives
12.2 Importance and Uses of Simulation Technique
12.3 Monte Carlo Simulation
12.4 Application of Simulation
12.6 Summary
12.7 Key Words
12.0 INTRODUCTION
As already discussed in the previous unit, simulation is a technique which is used in

an attempt to duplicate the features, appearance and characteristics of a real system.
In this unit, you will learn more about simulation technique and Monte Carlo method.
The unit covers the importance and uses of simulation technique and areas of
application of Monte Carlo simulation. In addition, the unit also discusses about
the applications of simulation, advantages and disadvantages of simulation.
12.1 OBJECTIVES

 Assess the importance and uses of simulation technique
 Discuss the Monte Carlo Method
 Describe the application of simulation
 Discuss the advantages and disadvantages of simulation
Self-Instructional
Material 219
Method 12.2 IMPORTANCE AND USES OF SIMULATION
TECHNIQUE
NOTES Monte Carlo simulation is often used by modern management when it cannot use
other techniques. There are many industrial problems which defy mathematical
solutions. The reason is that either they are too complicated or that the data cannot
be expressed in mathematical terms. In such cases, it is still possible to reach valid
conclusions by using the Monte Carlo technique. A considerable help is thus
obtained at practically no cost in taking decisions concerning the functioning of a
business system. The data and conclusions can be obtained through simulation of
an actual operation on the basis of its own past working. It paves the way for
predicting the changes in its behaviour and the result is evaluated from innovations
that we want to introduce.
By using a fresh series of random numbers at the appropriate junctures, we
can also examine the reactions of the simulated model just as if the same alterations
had actually been made in the system itself. Monte Carlo simulation, therefore,
provides a tool of knowing in advance as to whether or not the expense to be
incurred or the investment to be made in making the changes is envisaged. Through
this technique you can introduce the innovations on a piece of paper, examine their
effects and then may decide to adopt or not to adopt such innovations in the
functioning of real system. The usefulness of simulation lies in the fact that it allows
us to experiment with a model of the system rather than the actual system; in case
we are convinced about the results of our experiments, we can put the same into
practice. Thus the effect of the actual decisions are tested in advance through the
technique of simulation by resorting to the study of the model representing the real
life situation or the system.
The main purpose of simulation in management is to provide feedback,
which is vital for the learning process. It creates an atmosphere in which managers
play a dynamic role by enriching their experience through involvement in reckoning
with actual conditions through experimentation on paper. The technique permits
trying out several alternatives as the entire production for service process can be
worked out on paper, without dislocating the system in any way. Thus, Monte
Carlo technique transforms the manager from a blind folded driver of an automobile,
reacting to instructions of a fellow passenger to one who can see fairly, clearly,
where he is going.
Monte Carlo simulation has been applied to a wide diversity of problems ranging
from queuing process, inventory problem, risk analysis concerning a major capital
investment such as the introduction of a new product, expansion of the capacity
and many other problems. Budgeting is another area where simulation can be
very useful. In fact, the system of flexible budgeting is an exercise in simulation.
Self-Instructional
220 Material
Simulation can as well be used for preparing the master budget through functional Simulation: Monte Carlo
Method
budgets.
Over and above, the greatest contribution of simulation is in the analysis of
complex systems. Many real-world problems involve systems made up of many
NOTES
component parts that are interrelated. The system may be dynamic and changing
over time and may involve probabilistic or uncertain events. Simulation is the only
technique for quantitative analysis of such problems.
12.3 MONTE CARLO SIMULATION
Monte Carlo methods are basically the algorithms used in the computation of
result to be calculated from repeated random sampling. These methods help in
computerized calculations because these can perform repeated computation using
random or pseudo-random numbers. It is also used when it is not feasible to
compute correct result with a deterministic algorithm. Monte Carlo simulation
methods are used to study systems having degrees of freedom and in the situations
when there is significant ambiguity in inputs, for example, calculating risk factor in
a business.
Various simulation models, based on the principle of similitude (such as
model of aeroplanes initiating flight conditions in a wind tunnel) have been in use
for a long time. But Monte Carlo simulation is a recent operations research
innovation. The novelty lies in making use of pure chance to contact a simulated
version of the process under analysis, in exactly the same way as pure chance
operates the original system under working conditions. Only models under
uncertainty can be evaluated using Monte Carlo technique.
‘Monte Carlo’ is the code name given by John von Neumann and S.M.
Ulam to the technique of solving problems though it is too expensive for experimental
solutions and too complicated for analytical treatment.
Monte Carlo method is not one single method. It involves various widely-
used classes of approaches to follow a specific model. Using it we can,
 Define a domain with feasible inputs
 Randomly generate inputs from the domain
 Perform deterministic computation with the inputs
 Combine the results of the personal computations into the final result
Monte Carlo methods are used to solve various mathematical problems
based on sampling experiments in statistics using the sequences of random numbers
for simulation and are termed as statistical simulation methods. Thus, Monte Carlo
method is not one single method but it is a collection of various methods and is
basically used to perform similar procedure. Some of these methods are discussed
here with the help of solved examples.
Self-Instructional
Material 221
Simulation: Monte Carlo The Monte Carlo simulation technique can as well be used to solve
Method
probabilistic problems. Suppose, we are to evaluate the probability P that a tank
will be knocked out by either a first or second shot from an antitank gun assumed
to possess a constant kill probability of 1/2. The probability analysis will say that
NOTES the chance of tank being knocked out by either a first or second shot from an
antitank gun is 1/2+1/2 (1–1/2) =3/4. But we can also work out this probability
by simulating each round of the antitank gun by the flip of a coin through Monte
Carlo simulation technique.
Since the probability of a ‘head’ is the same as that of a kill, we may call it a hit
when the coin turns up a head and otherwise a miss. If we flip the coin a large
number of times, the value of P may be calculated by merely counting the number of
times a head turns up at least in two successive throws and then dividing this number
by the total pairs of throws of the coin. Monte Carlo method in this simple case, is
indeed a poor substitute for the theoretical probability analysis. But many real-life
systems are so complicated that even the well-defined probability analysis very often
fails but such situations can be handled by Monte Carlo simulation, particularly the
Monte Carlo technique that provides the simplest possible solutions for queuing
problems. Problems of corporate planning, inventory control, capital investment,
consumer behaviour and quality control can also be handled through simulation.
Monte Carlo simulation uses random number tables to reproduce on paper
the operation of any given system under its own working conditions. This technique
is used to solve problems that depend upon probability where formulation of
mathematical model is not possible. It involves first, the determination of the
probability distribution of the concerned variables and then sampling from this
distribution by means of random numbers to obtain data. It may, however, be
emphasized here that the probability distributions to be used should closely resemble
the real world situation.
One should always remember that simulation is not a perfect substitute but
rather an alternative procedure for evaluating a model. Analytical solution produces
the optimal answer to a given problem, while Monte Carlo simulation yields a
solution which should be very close to the optimal but not necessarily the exact
correct solution. Monte Carlo Simulation solution converge with to the optimal
solution as the number of simulated trials goes to infinity.
In context of Queuing Theory: Monte Carlo simulation which uses random number
tables can better be illustrated by considering any concrete operation subject to
chance. Let us take the arrival of scooters at a service station. First of all, observe
the actual arrival of scooters on number of days, say for five days, then put this
information in the following two ways:
 Group the number of scooters arriving every hour, say between 7–8, 8–9
a.m. and so on till 4–5 p.m. (assuming the working hours of the service
station is from 7 a.m. to 5 p.m. and also assuming that no scooter arrives
Self-Instructional
222 Material
before 7 a.m. nor any scooter after 5 p.m.). Work out the mean number of Simulation: Monte Carlo
Method
scooters arriving during 7–8 a.m., 8–9 a.m., 9–10 a.m. and so on. Let us
suppose, we get the following information:
Table 12.1 Mean Number of Scooters Arriving per Hour
NOTES
7-8 a.m. 5.6
8-9 a.m. 5.4
9-10 a.m. 3.4
10-11 a.m. 3.6
11-12 noon 2.0
12-1 p.m. 3.0
1-2 p.m. 4.0
2-3 p.m. 6.0
3-4 p.m. 3.0
4-5 p.m. 4.0
 Obtain the deviation of actual arrivals during a particular hour from the
corresponding mean and do it for all the hours from 7 a.m. to 5 p.m. There
will thus be 10 sets of 5 deviations (because of the observation on 5 days)
from each of the 10 mean hourly arrivals and then prepare a frequency
distribution of such deviations is to be prepared. Suppose we get the
frequency distribution of such deviations as under:
Table 12.2 Random Number Allotment
Deviation Frequency Percentage Probability Random
Nos. from Mean Frequency Allotted
+4.0 2 4 0.04 00–03
+3.5 3 6 0.06 04–09
+3.0 5 10 0.10 10–19
+2.5 4 8 0.08 20–27
+2.0 6 12 0.12 28–39
+1.5 2 4 0.04 40–43
+1.0 3 6 0.06 44–49
+0.5 4 8 0.08 50–57
–0.5 3 6 0.06 58–63
–1.0 2 4 0.04 64–67
–1.5 4 8 0.08 68–75
–2.0 2 4 0.04 76–79
–2.5 3 6 0.06 80–85
–3.0 1 2 0.02 86–87
–3.5 2 4 0.04 88–91
–4.0 4 8 0.08 92–99
Total 50 100 1.00
Self-Instructional
Material 223
Simulation: Monte Carlo In the last column of the above table, we have allotted 100 random numbers
Method
from 00 to 99, both inclusive according to the percentage frequency of the deviation
or the probability distribution of deviations. Thus deviation to the extent of +4.0
from the mean, having a frequency of 2 out of 50 (and as such 4 out of 100) or a
NOTES probability of 0.04, has been allotted 4 random numbers, 00 to 03. The next
deviation of +3.5 with a probability of 0.06 has the next 6 random numbers, 04 to
09, allotted to it. The same treatment has been done to all the remaining deviations.
The last deviation – 4.0 with a probability of 0.08 has been allotted to the last 8
random numbers, from 92 to 99.
The object of doing all this is to derive, by simulation, the actual number of
scooters that may be expected to arrive during any given hour. Suppose we want
to know the expected number of scooters arriving on a particular day during the
hour 8 to 9 a.m. The table giving the mean number of scooters arrival shows that
the mean arrival during this hour is 5.4. If we can ascertain the deviation of the
actual arrivals from the mean, we can easily work out the actual number of scooters
arrived. To do so, we look at the table of random numbers and select any two-
digit number at random. Suppose, the random number so selected is 84
corresponding to which the value of the deviation of actual arrivals from the mean
as per the above table is –2.5. In other words the actual arrivals for the hour 8-9
a.m. will be (5.4) – (2.5) = 2.9 (or approximately 3), so that we may say that 3
scooters will arrive between 8-9 a.m. on the day in question. The underlying rationale
of this simulation procedure is that every deviation from its corresponding mean
has the same chance of occurring by random number selection as in the actual
case because each deviation has as many random numbers allotted to it as its
frequency percentage in the general pool of all deviations as stated above.
We can go a step further and using the random number technique can even
simulate and tell the actual arrival time of the scooter coming to the service station
during any particular hour. If we are satisfied to note the arrival time correct to
within, say, 5 minutes the required numbers of minutes past the hour can take a
value only in one of the 12 intervals (0-5, 6-10, 11-15, ..., 56 to 60 minutes) into
which any hour can be divided. Keeping this in view and the actual observations
for all 5 days under consideration we can prepare a frequency table showing how
many scooters arrive within 5 minutes, how many within 6–10 minutes past the
hour and so on for the remaining intervals into which we choose to split the hour.
Let the observed information for 5 days period on this basis be put as shown in
Table 12.3.
Random numbers in the last column of the Table 12.3 have been allotted in
a similar manner as we did in an earlier table. We have already seen that 3 scooters
arrive during the hour 8–9 a.m. and now we want to know the actual time of their
coming to service station.
Self-Instructional
224 Material
Table 12.3 Number of Scooters Arriving Within the Number of Minutes Past the Hour Simulation: Monte Carlo
Method
Deviation Frequency Percentage Probability Random
Nos. from Mean Frequency Allotted
0-5 mts. 20 10 0.10 00-09
NOTES
6-l0 mts. 30 15 0.15 10-24
11-15 mts. 10 5 0.05 25-29
16-20 mts. 40 20 0.20 30-49
21-25 mts. 16 8 0.08 50-57
26-30 mts. 14 7 0.07 58-64
31-35 mts. 18 9 0.09 65-73
36-40 mts. 12 6 0.06 74-79
41-45 mts. 16 8 0.08 80-87
46-50 mts. 14 7 0.07 88-94
51-55 mts. 6 3 0.03 95-97
56-60 mts. 4 2 0.02 98-99
Total 200 100 1.00
For this purpose we pick a two-digit random number from the table of
random numbers and let us say it is 25. A reference to the above table shows that
this number occurs in the range of 25–29 which belongs to the interval 11–15
minutes and this means that the first of the 3 scooters arriving between 8–9 a.m.
arrives at 15 minutes past 8 a.m. Similarly picking two more random numbers viz.,
36 and 96 we find from the above table that they are related to intervals 16–20
minutes and 51–55 minutes respectively. Thus the second scooter arrives at 20
minutes past 8 a.m. and the third scooter arrives at 55 minutes past 8 a.m.
Proceeding in a similar manner we can also make a frequency table showing
the number of scooters serviced within intervals of varying magnitudes. Each of
these intervals can then be allotted a set of 100 random numbers (i.e., 00 to 99)
according to the probability distribution to provide a basis for simulating the pattern
of available service.
The above is an example of how arrival as well as service pattern in a
queuing process may be derived by Monte Carlo simulation. Remember that there
is no regularity either in the arrival of the scooters or in rendering service and
because of this there may be times when scooters have to wait for service while at
other times the service attendant may remain idle. If in such a case we want to add
one more service point to the service station then certainly we would first like to
assess whether the same would be economical or not. Simulation technique can
assist us in the matter. How does simulation help can better be made clear by the
following example:
Self-Instructional
Material 225
226
Method
Material
NOTES
Self-Instructional
Table 12.4 Simulation Worksheet for Arrival Time, Service Time and Waiting Time
Random Time Till Next Arrival Service Random Service Waiting Time Length of
Number Arrival Time Begins Number Time Ends Attendant Customers the Line
(Minutes) (a.m.) (a.m.) (Minutes) (a.m.) (Minutes) (Minutes)
44 5 10.05 10.05 50 5 10.10 5 – –

84 6 10.11 10.11 95 7 10.18 1 – 1
82 6 10.17 10.18 58 5 10.23 – 1 1
50 5 10.22 10.23 44 5 10.28 – 1 –
83 6 10.28 10.28 77 6 10.34 – – 1
40 5 10.33 10.34 11 4 10.38 – 1 –
96 8 10.41 10.41 08 3 10.44 3 – –
88 7 10.48 10.48 38 5 10.53 4 – 1
16 4 10.52 10.53 87 6 10.59 – 1 1
16 4 10.56 10.59 45 5 11.04 – 3 –
97 8 11.04 11.04 09 3 11.07 – – –
92 7 11.11 11.11 99 7 11.18 4 – 1
39 5 11.16 11.18 81 6 11.24 – 2 1
33 5 11.21 11.24 97 7 11.31 – 3 1
83 6 11.27 11.31 30 4 11.35 – 4 1
42 5 11.32 11.35 36 5 11.40 – 3 1
16 4 11.36 11.40 75 6 11.46 – 4 1
07 4 11.40 11.46 72 6 11.52 – 6 1
77 6 11.46 11.52 79 6 11.58 – 6 1
66 6 11.52 11.58 83 6 12.04 – 6 –
20 112 107 17 41 13
Example 12.1: A firm has a single channel service station with following empirical Simulation: Monte Carlo
Method
data available to its management:
(i) The mean arrival rate is 6.2 minutes.
(ii) The mean service time is 5.5 minutes. NOTES
(iii) The arrival and service time probability distributions are as follows:
Arrivals Probability Service Time Probability

(Minutes) (Minutes)
3-4 0.05 3-4 0.10
4-5 0.20 4-5 0.20
5-6 0.35 5-6 0.40
6-7 0.25 6-7 0.20
7-8 0.10 7-8 0.10
8-9 0.05 8-9 0.00
1.00 1.00
The queuing process begins at 10.00 a.m. and proceeds for nearly 2 hours.
An arrival goes to the service facility immediately if it is empty otherwise it will wait
in a queue. The queue discipline is, First come first served.
If the attendant’s wage is 8 per hour and the customer’s waiting time cost
9 per hour, would it be an economical proposition to engage second attendant?
Answer on the basis of Monte Carlo simulation technique. You may use the figures
based upon the simulated period for 2 hours.
Solution: From the given probability distributions of arrivals and service times,
first of all we allot the random numbers to the various intervals. This has been done
as follows:
Arrivals Probability Random Service Probability Random
(Minutes) Nos. Allotted Time Allotted Nos.
(Minutes)
3-4 0.05 00-04 3-4 0.10 00-09

4-5 0.20 05-24 4.5 0.20 10-29
5-6 0.35 25-59 5-6 0.40 30-69
6-7 0.25 60-84 6-7 0.20 70-89
7-8 0.10 85-94 7-8 0.10 90-99
8-9 0.05 95-99 8-9 0.00
These allotted random numbers now become the basis for generating arrival
and service times in conjunction with a table of random numbers. A simulation
worksheet as shown on the next page has been developed in the following manner:
As the given mean arrival rate is 6.2 minutes which means that in an hour
approximately 10 units arrive and as such in a simulation period for 2 hours about
Self-Instructional
Material 227
Simulation: Monte Carlo 20 units are expected to arrive. Hence, the number of arrivals for our simulation
Method
exercise is 20.
A table of random numbers (given in table 12.4) is used for developing the
simulation worksheet. The first random number for arrival time is 44. This number
NOTES
lies between 25–59 and this indicates a simulated arrival time of 5 minutes. All
simulated arrival and service times have been worked out in a similar fashion.
The next step is to list the arrival time in the appropriate column. The first
arrival comes in 5 minutes after the starting time. It means the attendant waited for
5 minutes. This has been shown under the column ‘Waiting Time: Attendant’. The
simulated service time for the first arrival is 5 minutes which results in the service
being completed by 10.10 a.m. The next arrival comes at 10.11 a.m. which indicates
that no one has waited in queue but the attendant has waited for 1 minute from
10.10 a.m. to 10.11 a.m. The service time ends at 10.18 a.m. But the third arrival
comes at 10.17 a.m. and the service of the second continues upto 10.18 a.m.,
hence the third arrival has to wait in the queue. This is shown in the column
‘Waiting Time: Customer’ of the simulation worksheet. One customer waiting in
queue is shown in the column—‘Length of the Line.’ The same procedure has
been followed throughout the preparation of the simulation worksheet.
The following information can be derived from the above stated simulation
worksheet.
1. Average length of queue:
No. of customers in line 13
=   0.65
No. of arrivals 20
2. Average waiting time of customer before service:
Customer waiting time 41
=   2.05 minutes
No. of arrivals 20
3. Average service time:
Total service time 107
=   5.35 minutes
No. of arrivals 20
4. Time a customer spends in the system:
= Average service time + Average waiting time before service
= 5.35 + 2.05 = 7.40 minutes.
Simulation worksheet developed above also states that if one more attendant
is added then there is no need for a customer to ‘Wait in Queue’. But the cost of
having one more attendant in addition to the existing one is to be compared with
the cost of one attendant and the customer waiting time. This can be worked out
as under:
Self-Instructional
228 Material
Two Hour Period Cost with Method
One Attendant Two Attendants
Customer waiting time
(41 mts. × 9 per hour) 6.15 nil NOTES
Attendant’s cost
(2 hours × 8 per hour) 16 32
Total 22.15 32
If the above analysis is based on simulation for a period of 2 hours and is
representative of the actual situation, then it can be concluded that the cost with
one attendant is lower than with two attendants. Hence, it would not be an
economical proposition to engage as additional attendant.
12.4 APPLICATION OF SIMULATION
Simulation is used in almost all fields. Some of the areas where simulation is used
are as follows:
 Manufacturing Applications
 Semiconductor Manufacturing
 Construction Engineering and project management
 Business Process Simulation
 Healthcare
 Logistics, Supply Chain and Distribution Application
 Risk Analysis
 Computer Simulation
 Transportation modes and Traffic
Main advantages of simulation are as follows:
 Simulation helps to study the behaviour of a system without building it.
 The results of simulation are accurate in general, in contrast to analytical
model.
 It also helps to find an unexpected phenomenon, behaviour of the system.
 The process of simulation also eases the procedure of ‘What-If’ analysis.
Main disadvantages of simulation are as follows:
 To build a simulation model is a bit expensive.
 In addition to this, it is also expensive to conduct simulation.
 At times, it is difficult to interpret the results of the simulation.
Self-Instructional
Material 229
Method
Check Your Progress
1. What is the main purpose of simulation in management?
NOTES 2. List the problems where Monte Carlo simulation is applied.
3. What is the greatest contribution of simulation?
4. Who introduced the name ‘Monte Carlo’?
5. State one of the uses of the Monte Carlo methods?

QUESTIONS
1. The main purpose of simulation in management is to provide feedback which

is vital for the learning process. It creates an atmosphere in which managers
play a dynamic role by enriching their experience through involvement in
reckoning with actual conditions through experimentation on paper.
2. Monte Carlo simulation has been applied to a wide diversity of problems
ranging from queuing process, inventory problem, risk analysis concerning
a major capital investment such as the introduction of a new product,
expansion of the capacity and many other problems. Budgeting is another
area where simulation can be very useful. In fact, the system of flexible
budgeting is an exercise in simulation. Simulation can as well be used for
preparing the master budget through functional budgets.
3. The greatest contribution of simulation is in the analysis of complex systems.
4. ‘Monte Carlo’ is the code name given by John von Neumann and S.M.Ulam.
5. Monte Carlo methods are used to solve various mathematical problems
based on sampling experiments in statistics using the sequences of random
numbers for simulation.
12.6 SUMMARY
 Monte Carlo simulation is often used by modern management when it cannot

use other techniques.
 The main purpose of simulation in management is to provide feedback,
which is vital for the learning process. It creates an atmosphere in which
managers play a dynamic role by enriching their experience through
involvement in reckoning with actual conditions through experimentation on
paper.
Self-Instructional
230 Material
 The technique permits trying out several alternatives as the entire production Simulation: Monte Carlo
Method
for service process can be worked out on paper, without dislocating the
system in any way.
 Monte Carlo technique transforms the manager from a blind folded driver
NOTES
of an automobile, reacting to instructions of a fellow passenger to one who
can see fairly, clearly, where he is going.
 Monte Carlo simulation has been applied to a wide diversity of problems
ranging from queuing process, inventory problem, risk analysis concerning
a major capital investment such as the introduction of a new product,
expansion of the capacity and many other problems. Budgeting is another
area where simulation can be very useful. In fact, the system of flexible
budgeting is an exercise in simulation. Simulation can as well be used for
preparing the master budget through functional budgets.
 The greatest contribution of simulation is in the analysis of complex systems.
Many real-world problems involve systems made up of many component
parts that are interrelated. The system may be dynamic and changing over
time and may involve probabilistic or uncertain events. Simulation is the
only technique for quantitative analysis of such problems.
 Monte Carlo methods are basically the algorithms used in the computation
of result to be calculated from repeated random sampling. These methods
help in computerized calculations because these can perform repeated
computation using random or pseudo-random numbers. It is also used when
it is not feasible to compute correct result with a deterministic algorithm.
 ‘Monte Carlo’ is the code name given by John von Neumann and S.M.
Ulam to the technique of solving problems though it is too expensive for
experimental solutions and too complicated for analytical treatment.
 Monte Carlo methods are used to solve various mathematical problems
based on sampling experiments in statistics using the sequences of random
numbers for simulation and are termed as statistical simulation methods.
 Monte Carlo simulation uses random number tables to reproduce on paper
the operation of any given system under its own working conditions. This
technique is used to solve problems that depend upon probability where
formulation of mathematical model is not possible.
 Monte Carlo simulation which uses random number tables can better be
illustrated by considering any concrete operation subject to chance.
 Simulation is used in almost all fields. Some of the areas where simulation is
used are manufacturing applications, semiconductor manufacturing, business
process simulation healthcare, computer simulation etc.
 Simulation helps to study the behaviour of a system without building it and
the results of simulation are accurate in general, in contrast to analytical
model.
Self-Instructional
Material 231
Method 12.7 KEY WORDS
 Monte Carlo Simulation: It refers to a method that is used to model the

NOTES probability of different outcomes in a process that cannot easily be predicted
due to the intervention of random variables.
 Sampling: It refers to the process of selecting a representative group from
the population under study. The target population is the total group of
individuals from which the sample might be drawn.

EXERCISES

1. State the areas of application of Monte Carlo simulation.
2. List the areas where simulation technique is used.
3. What are the advantages and disadvantages of the simulation?
1. Discuss Monte Carlo simulation with reference to queuing theory.
2. The following data is observed in a tea serving counter. The arrival is for
one minute interval.
No. of Persons Arriving 0 1 2 3 4 5
Probability 0.05 0.15 0.40 0.20 0.15 0.05
The service is taken as 2 person for one-minute interval. Using the following
random numbers simulate for 15-minutes period.
09, 54, 94, 01, 80, 73, 20, 26, 90, 79, 25, 48, 99, 25, 89
Calculate also the average number of persons waiting in queue per minute.

House.
Self-Instructional
232 Material
Project Scheduling and
UNIT 13 PROJECT SCHEDULING PERT-CPM
AND PERT-CPM
NOTES
Structure
13.0 Introduction
13.1 Objectives
13.2 PERT/CPM Network Components and Precedence Relationship
13.2.1 Basic Difference between PERT and CPM
13.2.2 Basic Terms
13.2.7 Time Analysis
13.3 Critical Path Method (CPM)
13.4 Project Management-PERT
13.6 Summary
13.7 Key Words
13.0 INTRODUCTION
Network scheduling is a technique used for planning and scheduling large projects,
in the fields of construction, maintenance, fabrication and purchasing of computer
systems, etc. It is a method of minimizing the trouble spots such as production,
delays and interruptions, by determining critical factors and co-ordinating various
parts of the overall job.
There are two basic planning and control techniques that utilize a network
to complete a predetermined project or schedule. These are Programme Evaluation
Review Technique (PERT) and Critical Path Method (CPM). A project is defined
as a combination of interrelated activities, all of which must be executed in a certain
order for its completion.
The work involved in a project can be divided into three phases,
corresponding to the management functions of planning, scheduling and controlling.
Planning: This phase involves setting the objectives of the project as well as the
assumptions to be made. It also involves the listing of tasks or jobs that must be
performed in order to complete a project under consideration. In this phase, in
addition to the estimates of costs and duration of the various activities, the manpower,
Self-Instructional
machines and materials required for the project are also determined. Material 233
Project Scheduling and Scheduling: This consists of laying the activities according to their order of
PERT-CPM
precedence and determining the following:
(i) The start and finish times for each activity
NOTES (ii) The critical path on which the activities require special attention.
(iii) The slack and float for the non-critical paths.
Controlling: This phase is exercised after the planning and scheduling. It involves
the following:
(i) Making periodical progress reports
(ii) Reviewing the progress
(iii) Analyzing the status of the project
(iv) Making management decisions regarding updating, crashing and resource
allocation, etc.
13.1 OBJECTIVES

 Differentiate between PERT and CPM
 Identify the components of PERT/CPM network
 Explain the precedence relationship in PERT/CPM
 Apply PERT for project management
13.2 PERT/CPM NETWORK COMPONENTS AND

PRECEDENCE RELATIONSHIP
Program Evaluation Review Technique (PERT) and Critical Path Method (CPM)
are two planning and control techniques for keeping a project schedule on track
to complete within the scheduled time. Network scheduling is a technique used
for planning and scheduling large projects in the field of construction, maintenance,
fabrication, purchasing computer system, etc. PERT is a probabilistic method where
the activity times are represented by a probability distribution.
Critical path method (CPM) is a graphical technique for planning and
scheduling of projects. This technique involves the preparation of the network in
the form of arrow diagram and its analysis to indicate the critical path. It has the
potential for scheduling of a task in minimum time and/or cost in accordance with
specified constraints. After preparing the network diagram and indicating the times
for each activity, you can mention the various possible paths for determining the
critical path. The critical path being the longest path can easily be found out from
the possible paths as the one taking the maximum time in the completion of the
project.
Self-Instructional
234 Material
13.2.1 Basic Difference between PERT and CPM Project Scheduling and
PERT-CPM
Although PERT and CPM were developed separately and originally there were
many differences between the two, over time these project planning techniques
have become almost identical. CPM also is a display of a complex project as a NOTES
network, with one “time estimate” used for each step in the project, as compared
to PERT where three “time estimates” are used. In this technique the times and
cost of activities are known. It accommodates situations in which sets of
standardized activities are a part of the project for which the times for completion
of these activities can be more accurately calculated. However, since there is only
one time estimate, which may or may not be accurate, the probability for completing
the project on time is difficult to compute. In all other aspects PERT and CPM
one similar.
13.2.2 Basic Terms
To understand the network techniques, one should be familiar with a few basic
terms of which both CPM and PERT are special applications.
Network It is the graphic representation of logically and sequentially connected
arrows and nodes, representing activities and events in a project. Networks are
also called arrow diagrams.
Activity An activity represents some action and is a time consuming effort
necessary to complete a particular part of the overall project. Thus, each and
every activity has a point of time where it begins and a point where it ends.
It is represented in the network by an arrow,
Here A is called the activity.

Event The beginning and end points of an activity are called events or nodes.
Event is a point in time and does not consume any resources. It is represented by
a numbered circle. The head event called the jth event always has a number higher
than the tail event, which is also called the ith event.
Merge and burst events It is not necessary for an event to be the ending event
of only one activity as it can be the ending event of two or more activities. Such an
event is defined as a merge event.
Self-Instructional
Material 235
Project Scheduling and If the event happens to be the beginning event of two or more activities, it is
PERT-CPM
defined as a burst event.
NOTES
Preceding, succeeding and concurrent activities Activities that must be

accomplished before a given event can occur, are termed as preceding activities.
Activities that cannot be accomplished until an event has occurred, are termed
as succeeding activities.
Activities that can be accomplished concurrently, are known as concurrent
activities.
This classification is relative, which means that one activity can be preceding to a
certain event, and the same activity can be succeeding to some other event or it may be
a concurrent activity with one or more activities.
Dummy activity Certain activities, which neither consume time nor resources
but are used simply to represent a connection or a link between the events are
known as dummies. It is shown in the network by a dotted line. The purpose of
introducing dummy activity is:
(i) To maintain uniqueness in the numbering system, as every activity may have a
distinct set of events by which the activity can be identified.
(ii) To maintain a proper logic in the network.

Following are the three common errors in a network construction:
Looping (cycling) In a network diagram, a looping error is also known as cycling
error. Drawing an endless loop in a network is known as error of looping. A loop
can be formed if an activity is represented as going back in time.
Self-Instructional
236 Material
Dangling To disconnect an activity before the completion of all the activities in Project Scheduling and
PERT-CPM
a network diagram, is known as dangling.
NOTES
Redundancy If a dummy activity is the only activity emanating from an event and can
be eliminated, it is known as redundancy.

There are a number of rules in connection with the handling of events and activities
of a project network that should be followed.
(i) Try to avoid arrows that cross each other.
(ii) Use straight arrows.
(iii) No event can occur until every activity preceding it has been completed.
(iv) An event cannot occur twice, i.e., there must be no loops.
(v) An activity succeeding an event cannot be started until that event has
occurred.
(vi) Use arrows from left to right. Avoid mixing two directions, vertical and
standing arrows may be used if necessary.
(vii) Dummies should be introduced only if it is extremely necessary.
(viii) The network has only one entry point called the start event and one point
of emergence called the end or terminal event.
After the network is drawn in a logical sequence, every event is assigned a number.
The number sequence must be such so as to reflect the flow of the network. In
numbering the events, the following rules should be observed.
(i) Event numbers should be unique.
(ii) Event numbering should be carried out on a sequential basis, from left to
right.
(iii) The initial event, which has all outgoing arrows with no incoming arrow is
numbered as 1. Self-Instructional
Material 237
Project Scheduling and (iv) Delete all arrows emerging from all the numbered events. This will create
PERT-CPM
at least one new start event, out of the preceding events.
(v) Number all new start events 2, 3 and so on. Repeat this process until the
terminal event without any successor activity is reached. Number the
NOTES terminal node suitably.
Note: The head of an arrow should always bear a number higher than the one
assigned to the tail of the arrow.
Example 13.1 Construct a network for the project whose activities and
precedence relationships are as given below:
Activities A B C D E F G H I
Immediate – A A – D B, C, E F D G, H
Predecessor
Solution From the given constraints, it is clear that A and D are the starting
activities and I the terminal activity. B and C are starting with the same event and
are both the predecessors of the activity F. Also, E has to be the predecessor of
both F and H. Hence, we have to introduce a dummy activity.
Self-Instructional D1 is the dummy activity.

238 Material
Finally, we have the following network. Project Scheduling and
PERT-CPM
NOTES
Example 13.2 Construct a network for each of the projects whose activities
and their precedence relationships are given below.
Activity A B C D E F G H I J K
Predecessor – – – A B B C D E H, I F, G
Solution A, B and C are the concurrent activities as they start simultaneously. B

becomes the predecessor of activities E and F. Since the activities J and K have
two preceding activities, a dummy may be introduced (if possible).
Self-Instructional
Material 239
PERT-CPM
NOTES
Finally we have,
Example 13.3 A < C, D, I; B < G, F; D < G, F; F < H, K; G, H < J; I, J,

K<E
Solution Given A < C, which means that C cannot be started until A is completed.
That is, A is the preceding activity to C. The above constraints can be given in the
following table.
Predecessor – – A A I, J, K B, D B, D F A G, H F
A and B are the starting activities, and E is the terminal activity.
Self-Instructional
240 Material
PERT-CPM
NOTES
Finally, we have,
Self-Instructional
Material 241
Project Scheduling and 13.2.7 Time Analysis
PERT-CPM
Once the network of a project is constructed, the time analysis of the network
becomes essential for planning various activities of the project. Activity time is a
forecast of the time an activity is expected to take from its starting point to its
NOTES
completion (under normal conditions).
We shall use the following notation for basic scheduling computations.
(i, j) = Activity (i, j) with tail event i and head event j
Tij = Estimated completion time of activity (i, j)
ESij = Earliest starting time of activity (i, j)
EFij = Earliest finishing time of activity (i, j)
LSij = Latest starting time of activity (i, j)
LFij = Latest finishing time of activity (i, j).
The basic scheduling computation can be put under the following three groups.
Before starting computations, the occurrence time of the initial network event is
fixed. The forward pass computation yields the earliest start and the earliest finish
time for each activity (i, j) and indirectly the earliest occurrence time for each
event namely Ei. This consists of the following three steps:
Step 1 The computations begin from the start node and move towards the end node.
Let zero be the starting time for the project.
Step 2 Earliest starting time (ES)ij = Ei is the earliest possible time when an
activity can begin, assuming that all of the predecessors are also started at
their earliest starting time. Earliest finish time of activity (i, j) is the earliest
starting time + the activity time.
(EF)ij = (ES)ij + tij
Step 3 Earliest event time for event j is the maximum of the earliest finish time of
all the activities, ending at that event.
Ej = Max (Ei + tij)
i
The computed ‘E’ values are put over the respective rectangles
representing each event.
The latest event time (L) indicates the time by which all activities entering into that
event must be completed without delaying the completion of the project. These
can be calculated by reversing the method of calculations used for the earliest
event time. This is done in the following steps.
Step 1 For ending event assume, E = L.
Self-Instructional
242 Material
Step 2 Latest finish time for activity (i, j) is the target time for completing the Project Scheduling and
PERT-CPM
project,
(LFij) = Lj
Step 3 Latest starting time of the activity (i, j ) = latest completion time of (i, j) NOTES
– the activity time
LSij = LFij – tij
= Lj – tij
Step 4 Latest event time for event i is the minimum of the latest start time of all
activities originating from the event.
Li = Min
j
(Lj – tij)
The computed ‘L’ values are put over the respective triangles representing
each event.
Float is defined as the difference between the latest and the earliest activity time.
Slack is defined as the difference between the latest and the earliest event time.
Hence, the basic difference between the slack and float is that slack is used for
events only; whereas float is used for activities.
There are mainly three kinds of floats as given below.
Total float It refers to the amount of time by which the completion of an activity
could be delayed beyond the earliest expected completion time, without affecting
the overall project duration time.
Mathematically, the total float of an activity (i, j) is the difference between the
latest start time and the earliest start time of that activity.
Hence, the total float for an activity (i, j ) denoted by (TF)ij is calculated by the
formula,
(TF)ij = (Latest start – Earliest start) for activity (i, j)
i.e., (TF)ij = (LS)ij – (ES)ij

or (TF)ij = (Lj – Ei) – tij
where, Ei and Lj are the earliest time and latest time for the tail event i and head
event j and tij is the normal time for the activity (i, j ). This is the most important type
of float as it concerns the overall project duration.
Free float The time by which the completion of an activity can be delayed beyond
the earliest finish time, without affecting the earliest start of a subsequent succeeding
activity.
Self-Instructional
Material 243
Project Scheduling and Mathematically, the free float for activity (i, j) denoted by (FF)ij can be calculated
PERT-CPM
by the formula,
FFij = (Ej – Ei) – tij

NOTES
(FF)ij = Total float – Head event slack
Head event slack = Lj – Ej

This float is concerned with the commencement of the subsequent activity.
The free float can take values from zero up to total float, but it cannot exceed
total float. This float is very useful for rescheduling an activity with minimum
disruption in earlier plans.
Independent float The amount of time by which the start of an activity can be
delayed, without affecting the earliest start time of any immediately following
activities, assuming that the preceding activity has finished at its latest finish time.
Mathematically, independent float of an activity (i, j) denoted by (IF)ij can be
calculated by the formula,
IFij = (Ej – Li) – tij

or
(IF)ij = Free Float – Tail event slack
where tail event slack is given by,
Tail event slack = Li – Ei
The negative independent float is always taken as zero. This float is concerned
with prior and subsequent activities.
IFij  FFij  TFij
Note: (i) If the total float TFij for any activity (i, j) is zero, then such an activity is
called critical activity.
(ii) The float can be used to reduce project duration. While doing this, the
float of not only that activity, but that of other activities will also change.
Critical activity An activity is said to be critical if a delay in its start cause a
further delay in the completion of the entire project.
Critical path The sequence of critical activities in a network is called the critical
path. It is the longest path in the network, from the starting event to the ending event
and defines the minimum time required to complete the project. In the network it is
denoted by a double line and identifies all the critical activities of the project. Hence,
for the activities (i, j) to lie on the critical path, following conditions must be satisfied.
(a) ESi = LFi
(b) ESj = LFj
Self-Instructional
244 Material
(c) ESj – ESi = LFj – LFi = tij Project Scheduling and
PERT-CPM
ESi and ESj are the earliest start and finish time of the events i and j.
LFi and LFj are the latest start and finish time of the events i and j.
NOTES
13.3 CRITICAL PATH METHOD (CPM)
The iterative procedure of determining the critical path (CP) is as follows:
Step 1 List all the jobs and then draw an arrow (network) diagram. Each job is
indicated by an arrow with the direction of the arrow showing the sequence
of jobs. The length of the arrows has no significance. The arrows are
placed based on the predecessor, successor and concurrent relation within
the job.
Step 2 Indicate the normal time (tij ) for each activity (i, j) above the arrow, which
is deterministic.
Step 3 Calculate the earliest start time and the earliest finish time for each event
and write the earliest time Ei for each event i in the . Also calculate
the latest finish and latest start time. From this we calculate the latest time
Lj for each event j and put it in the .
Step 4 Tabulate the various times, namely, normal time, earliest time and latest
time on the arrow diagram.
Step 5 Determine the total float for each activity by taking the difference between
the earliest start and the latest start time.
Step 6 Identify the critical activities and connect them with the beginning and the
ending events in the network diagram by double line arrows. This gives
the critical path.
Step 7 Calculate the total project duration.
Note: The earliest start and finish time of an activity, as well as the latest start and
finish time of an activity are shown in the table. These are calculated by using the
following hints.
To find the earliest time, we consider the tail event of the activity. Let the starting
time of the project namely ESi = 0. Add the normal time with the starting time, to
get the earliest finish time. The earliest starting time for the tail event of the next
activity is given by the maximum of the earliest finish time for the head event of the
previous activity.
Similarly, to get the latest time, we consider the head event of the activity.
The latest finish time of the head event of the final activity is given by the target
time of the project. The latest start time can be obtained by subtracting the normal
time of that activity. The latest finish time for the head event of the next activity is
Self-Instructional
Material 245
Project Scheduling and given by the minimum of the latest start time for the tail event of the previous
PERT-CPM
activity.
Example 13.4 A project schedule has the following characteristics.

NOTES
Activity 1–2 1– 3 2– 4 3–4 3– 5 4– 9 5–6 5–7 6– 8 7– 8 8–10 9–10
Time (days) 4 1 1 1 6 5 4 8 1 2 5 7
From the above information, you are required to:

1. Construct a network diagram.
2. Compute the earliest event time and latest event time.
3. Determine the critical path and total project duration.
4. Compute total and free float for each activity.
Solution First we construct the network with the given constraints (here we get
it by just connecting the event numbers).
The following table gives the critical path as well as total and free floats calculation.
Self-Instructional
246 Material
The earliest and latest calculations are shown below. Project Scheduling and
PERT-CPM
Forward pass calculation In this we estimate the earliest start (Esi) and finish
times (Esj). The earliest time for the event i is given by,
Ei = Max (ESi + tij) NOTES
i
ES1 = 0 = E1 = 0
E2 = ES2 = ES1 + t12 = 0 + 4 = 4
E3 = ES3 = ES1 + t13 = 0 + 1 = 1
E4 = ES4 = Max (ES3 + t34, ES2 + t24)
= Max (1 + 1, 4 + 1) = 5
E5 = (E3 + t35) = 1 + 6 = 7
E6 = E5 + t56 = 7 + 4 = 11
E7 = E5 + t57 = 7 + 8 = 15
E8 = Max (E6 + t68, E7 + t78)
= Max (11 + 1, 15 + 2) = 17
E9 = E4 + t49 = 5 + 5 = 10
E10 = Max (E9 + t9,10, E8 + t8, 10)
= Max (10 + 7, 17 + 5) = 22.
Backward pass calculation In this we calculate the latest finish and the latest
start time. The latest time L for an event i is given by Li = Min
j
(LFj – tij),
where, LFj is the latest finish time for the event j, tij is the normal time of the
activity.
L10 = 22
L9 = L10 – t9, 10 = 22 – 7 = 15
L8 = L10 – t8, 10 = 22 – 5 = 17
L7 = L8 – t7, 8 = 17 – 2 = 15
L6 = L8 – t6, 8 = 17 – 1 = 16
L5 = Min (L6 – t5, 6, L7 – t5, 7)
= Min (16 – 4, 15 – 8) = 7
L4 = L9 – t4, 9 = 15 – 5 = 10
L3 = Min (L4 – t3, 4, L5 – t3, 5 )
= Min (10 – 1, 7 – 6) = 1
L2 = L4 – t2, 4 = 10 – 1 = 9
L1 = Min (L2 – t12, L3 – t13) = Min (9 – 4, 1 – 1) = 0.
These calculations are shown in the above table.
Self-Instructional
Material 247
Project Scheduling and To find the TF (Total Float) Considering the activity 1 – 2, TF of (1 – 2) Latest
PERT-CPM
start Earliest start
5 0 5
Similarly TF (2 – 4) LS ES
NOTES
9 4 5
Free float TF Head event slack.
Consider the activity 1 – 2
FF of 1 – 2 TF of 1 – 2 Slack for the head event 2
5 (9 – 4) (from the figure for event 2)
5 5 0
FF of 2 – 4 TF of 2 – 4 Slack for the head event 4
5 (10 – 5) 5 5 0
Like this we calculate the TF and FF for the remaining activities.
From the above table we observe that the activities 1 – 3, 3 – 5, 5 – 7, 7 – 9, 8 –
10 are the critical activities as their total float is 0.
Hence, we have the following critical path.
1 – 3 – 5 – 7 – 8 – 10, with the total project duration of 22 days.
Example 13.5 A small maintenance project consists of the following jobs, whose
precedence relationships are given below.
Job 1–2 1–3 2–3 2–5 3–4 3–6 4–5 4–6 5–6 6–7
Duration 15 15 3 5 8 12 1 14 3 14
(days)
1. Draw an arrow diagram representing the project.

2. Find the total float for each activity.
3. Find the critical path and the total project duration.
Solution
Self-Instructional
248 Material
Forward pass calculation In this we estimate the earliest start and the earliest Project Scheduling and
PERT-CPM
finish time ESj given by,
ESj = Maxi
(ESi + tij) where, ESi is the earliest start time and tij is the normal
time for the activity (i, j).
NOTES
ES1 = 0
ES2 = ES1 + t15 = 0 + 15 = 15
ES3 = Max (ES2 + t23, ES1 + t13)
= Max (15 + 3, 0 + 15) = 18
ES4 = ES3 + t34 = 18 + 8 = 26
ES5 = Max (ES2 + t25, ES4 + t45)
= Max (15 + 5, 26 + 1) = 27
ES6 = Max (ES3 + t36, ES4 + t46, ES5 + t56)
= Max (18 + 12, 26 + 14, 27 + 3)
= 40
ES7 = ES6 + t67 = 40 + 14 = 54.
Backward pass calculation In this we calculate the latest finish and latest start
time LFi, given by LFi = Min
i
(LFj – tij) where, LFj is the latest finish time for the
event j
LF7 = 54
LF6 = LF7 – t67 = 54 – 14 = 40
LF5 = LS6 – t56 = 40 – 3 = 37
LF4 = Min (LS5 – t45, LS6 – t46)
= Min (37 – 1, 40 – 14) = 26
LF3 = Min (LF4 – t34, LF6 – t36)
= Min (26 – 8, 40 – 12) = 18
LF2 = Min (LF5 – t25, LF3 – t23)
= Min (37 – 5, 18 – 3) = 15
LF1 = Min (LF3 – t13, LF2 – t12)
= Min (18 – 15, 15 – 15) = 0
Self-Instructional
Material 249
Project Scheduling and The following table gives the calculations for critical path and total float.
PERT-CPM
NOTES
From the above table, we observe that the activities 1—2, 2—3, 3—4, 4—6, 6—7
are the critical activities and the critical path is given by, 1—2—3—4—6—7
The total time taken for project completion is 54 days.
Example 13.6 Tasks A, B, ... H, I constitute a project. The notation X < Y

means that the task X must be completed before Y is started. With the notation,
A < D, A < E, B < F, D < F, C < G, C < H, F < I, G < I
Draw a graph to represent the sequence of tasks and find the minimum time of
completion of the project, when the time (in days) of completion of each task is as
follows.
The above constraints can be given as in the following table:
Task A B C D E F G H I
Time (days) 8 10 8 10 16 17 18 14 9
Solution The above constraints are given in the following table:
Activity A B C D E F G H I
Preceding – – – A A B, D C C F, G
Activity
Self-Instructional
250 Material
PERT-CPM
NOTES
Time calculation Using forward and backward pass calculation, we first estimate
the earliest and the latest time for each event.
ES1 = E1 = 0
E2 = E1 + t12 = 0 + 8 = 8
E3 = Max (E1 + E13, E2 + t23)
= Max (0 + 10, 8 + 10) = 18
E4 = E1 + t14 = 0 + 8 = 8
E5 = Max (E3 + t35, E4 + t45)
= Max (18 + 17, 8 + 18) = 35
E6 = Max (E2 + t26, E4 + t46, E5 + t56)
= Max (8 + 16, 8 + 14, 35 + 9) = 44
The value of the latest time can now be obtained.
L6 = E6 = 44 (Target completion time for the project)
L5 = L6 – t56 = 44 – 9 = 35
L4 = Min (L6 – t46, L5 – t45)
= Min (44 – 14, 35 – 18) = 17
L3 = L5 – t35 = 35 – 17 = 18
L2 = Min (L6 – t26, L3 – t23)
= Min (44 – 16, 18 – 10 = 8
L1 = Min (L4 – t14, L3 – t13, L2 – t12)
= Min (17 – 8, 18 – 10, 8 – 8) = 0.
Self-Instructional
Material 251
Project Scheduling and To evaluate the critical events, all these calculations are put in the following
PERT-CPM
table.
NOTES
The above table shows that the critical events are the tasks 1 – 2, 2 – 3, 3 – 5, 5
– 6 as their total float is zero.
The critical path is given by 1–2–3–5–6 or A–D–F–I, with the total project
duration as 44 days.
13.4 PROJECT MANAGEMENT-PERT
The network methods discussed so far may be termed as deterministic, since

estimated activity times are assumed to be known with certainty. However, in the
research project or design of a gear box or a new machine, various activities are
based on judgement. It is difficult to obtain a reliable time estimate due to the changing
technology since time values are subject to chance variations. For such cases, where
the activities are non-deterministic in nature, PERT was developed. Hence, PERT
is a probabilistic method, where the activity times are represented by a probability
distribution. This distribution of activity times is based on three different time
estimates made for each activity, which are as follows:
(i) Optimistic time estimate
(ii) Most likely time estimate
Self-Instructional
252 Material
(iii) Pessimistic time estimate
PERT-CPM
NOTES
Time distribution curve
Optimistic time estimate It is the smallest time taken to complete the activity, if
everything goes well. There is very little chance that an activity can be completed in a
time less than the optimistic time. It is denoted by to or a.
Most likely time estimate It refers to the estimate of the normal time the activity
would take. This assumes normal delays. It is the mode of the probability
distribution. It is denoted by tm or m.
Pessimistic time estimate It is the longest time that an activity would take, if
everything goes wrong. It is denoted by tp or b. These three time values are
shown in the following figure.
From these three time estimates, we have to calculate the expected time of an
activity. It is given by the weighted average of the three time estimates,
to  4tm  t p
te =
6
[ distribution with weights of 1, 4 and 1, for to , tm and tp estimates respectively.]
Variance of the activity is given by,
2
 tp  to 
2=  
 6 
The expected length (duration), denoted by to Tc of the entire project is the
length of the critical path, i.e., the sum of the tc’s of all the activities along the
critical path.
The main objective of the analysis through PERT is to find the completion for a
particular event within the specified date Ts, given by P (Z  D) where,
Due date – Expected date of completion
D=
Project variance
Here, Z stands for standard normal variable.
PERT Procedure
Step 1 Draw the project network.
Step 2 Compute the expected duration of each activity using the formula,
t  4tm  tp
te = o 2
6
2  t p  to 
Also calculate the expected variance  =   of each activity..
 6 
Step 3 Compute the earliest start, earliest finish, latest start, latest finish and total
float of each activity.
Self-Instructional
Material 253
Project Scheduling and Step 4 Find the critical path and identify the critical activities.
PERT-CPM
Step 5 Compute the project length variance 2, which is the sum of the variance
of all the critical activities and hence, find the standard deviation of the
project length .
NOTES T T
Step 6 Calculate the standard normal variable Z = s e , where Ts is the

scheduled time to complete the project.
Te = Normal expected project length duration.
 = Expected standard deviation of the project length.
Using the normal curve, we can estimate the probability of completing the
project within a specified time.
Example 13.7 The following table shows the jobs of a network along with their
time estimates.
Job 1–2 1–6 2–3 2–4 3–5 4–5 6–7 5–8 7–8
a (days) 1 2 2 2 7 5 5 3 8
m (days) 7 5 14 5 10 5 8 3 17
b (days) 13 14 26 8 19 17 29 9 32
Draw the project network and find the probability of the project completing in
40 days.
Solution First we calculate the expected time and standard deviation for each
activity.
2
to + 4t m + t p t p - to
Activity te = 2 =
6 6
2
1  (4  7)  13  13  1 
1–2 =7   =4
6  6 
2
2  (4  5)  14  14  2 
1–6 =6   =4
6  6 
2  (4  14)  26  26  2 
2–3 = 14   = 16
6  6 
2
2  (5  4)  8 8 2
2–4 =5   =1
6  6 
2
7  (4  10)  19  19  7 
3–5 = 11   =4
6  6 
2
5  (5  4)  17  17  5 
4–5 =7   =4
6  6 
2
5  (8  4)  29  29  5 
6–7 = 11   = 16
6  6 
3  (3  4)  9  9  3
5–8 =4   =1
6  6 
2
8  (4  17)  32  32  8 
7–8 = 18   = 16
Self-Instructional 6  6 
254 Material
PERT-CPM
NOTES
Expected project duration = 36 days

Critical path 1–2–3–5–8
Project length variance, 2 = 4 + 16 + 4 + 1 = 25
 =5
The probability that the project will be completed in 40 days is given by,
P(Z  D)
Ts  Te 40  36 4
D= = = = 0.8
 5 5
Area under the normal curve for the region Z  0.8

P(Z  0.8)
= 0.5 + (0.8) [(8) = 0.2881 (from table)]
= 0.5 + 0.2881 = 0.7881
= 78.81%
Conclusion If the project is performed 100 times, under the same conditions,
there will be 78.81 occasions for this job to be completed in 40 days.
Example 13.8 A small project is composed of seven activities, whose time
estimates are listed in the table as follows:
Self-Instructional
Material 255
Estimated duration (weeks)
PERT-CPM Activity
Optimistic (a) Most likely (m) Pessimistic (b)
1–2 1 1 7
1–3 1 4 7
NOTES
2–4 2 2 8
2–5 1 1 1
3–5 2 5 14
4–6 2 5 8
5–6 3 6 15
You are required to:

1. Draw the project network.
2. Find the expected duration and variance of each activity.
3. Calculate the earliest and latest occurrence for each event and the expected
project length.
4. Calculate the variance and standard deviations of project length.
5. What is the probability that the project will be completed,
(i) 4 weeks earlier than expected?
(ii) Not more than 4 weeks later than expected?
(iii) If the project’s due date is 19 weeks, what is the probability of meeting
the due date?
Solution The expected time and variance of each activity is computed as shown
in the table below:
2
Activity a m b a + 4t m + b b - a
te = s2 =  
6  6 
1–2 1 1 7 2 1
1–3 1 4 7 4 1
1–4 2 2 8 3 1
2–5 1 1 1 1 0
3–5 2 5 14 6 4
4–6 2 5 8 5 1
5–6 3 6 15 7 4
The earliest and the latest occurrence time for each is calculated as below:
E1 = 0;
E2 = 0 + 2 + 2
E3 = 0 + 4 = 4
E4 = 0 + 3 = 3
E5 = Max (2 + 1, 4 + 6) = 10
Self-Instructional
E6 = Max (10 + 7, 3 + 5) = 17.
256 Material
To determine the latest expected time, we start with E6 being the last event and Project Scheduling and
PERT-CPM
move backwards subtracting te from each activity. Hence, we have,
L6 = E6 = 17
L5 = L6 – 7 = 17 – 7 = 10 NOTES
L4 = 17 – 5 = 12
L3 = 10 – 6 = 4
L2 = 10 – 1 = 9
L1 = Min (9 – 2, 4 – 4, 12 – 3) = 0
Using the above information we get the following network, where the critical
path is shown by the double-line arrows.
We observe the critical path of the above network as 1     6.

The expected project duration is 17 weeks, i.e., Te = 17 weeks.
The variance of the project length is given by,
2 = 1 + 4 + 4 = 9.
(i) The probability of completing the project within 4 weeks earlier than expected
is given by,
Ts  T
P(Z  D), where D =
s
Due date – Expected date of completion
D=
Project variance
17  4  17 12  17 4
D= = =
3 3 3
= – 1.33
Self-Instructional
Material 257
Project Scheduling and  P(Z  – 1.33) = 0.5 – (1.33)
PERT-CPM
= 0.5 – 0.4082 (from the table)
= .0918 = 9.18%.
NOTES
Conclusion If the project is performed 100 times under the same conditions,
then there will be 9 occasions for this job to be completed in 4 weeks earlier than
expected.
(ii) The probability of completing the project not more than 4 weeks later than
expected is given by,
P(Z  D), where
Ts  Te
D=

Here, Ts = 17 + 4 = 21
21  17 4
0= = = 1.33
3 3
P(Z  1.33)
= 0.5 + (1.33)
= 0.5 + 0.4082 (from the table)
= 0.9082 = 90.82%.
then there will be 90.82 occasions when this job will be completed not more than
4 weeks later than expected.
(iii) The probability of completing the project within 19 weeks, is given by,
19  17 2
P(Z  D), where D = = [ Ts = 19]
3 3
= 0.666.
i.e., P(Z  0.666) = 0.5 + (0.666)
= 0.5 + 0.2514 (from the table)
= 0.7514 = 75.14%.
Conclusion If the project is performed 100 times, under the same conditions,
then there will be 75.14 occasions for this job to be completed in 19 weeks.
Self-Instructional
258 Material
Example 13.9 Consider the following project. Project Scheduling and
PERT-CPM
Time estimate in weeks
Activity Predecessor
to tm tp
A 3 6 9 None NOTES
B 2 5 8 None
C 2 4 6 A
D 2 3 10 B
E 1 3 11 B
F 4 6 8 C, D
G 1 5 15 E
Find the path and standard deviation. Also find the probability of completing
the project by 18 weeks.
Solution First we calculate the expected time and variance of each activity as
in the following table.
2
to tm tp t o + 4t m + t p  t p - to 
Activity te = 2 =  
6  6 
2
3  (4  6)  9 9  3
A 3 6 9 =6   =1
6  6 
2
2  (4  5)  8 8 2
B 2 5 8 =5   =1
6  6 
2
2  (4  4)  6 6  2
C 2 4 6 =4   = 0.444
6  6 
2
2  (4  3)  10  10  3 
D 2 3 10 =4   = 1.777
6  6 
2
1  (4  3)  11  11  1 
E 1 3 11 =4   = 2.777
6  6 
2
4  (4  6)  8 8 4
F 4 6 8 =6   = 0.444
6  6 
2
1  (4  5)  15  15  1 
G 1 5 15 =6   = 5.444
6  6 
We construct the network with the help of the predecessor relation given in the
data.
Self-Instructional
Material 259
Project Scheduling and Critical path is 1–2–4–6 or A–C–F.
PERT-CPM
The project length = 18 weeks.
Project length variance, 2 = 1 + 0.444 + 0.444 = 1.888
NOTES Standard deviation,  = 1.374
The probability of completing the project in 18 weeks is given by,
Ts  Te
P(Z  D), where D =

Ts = 18; Te = 16;  = 1.374

18  16
D= = 1.4556
1.374
P(Z  D) = P(Z  1.4556) = 0.5 + (1.4456)

= 0.5 + 0.4265 = 0.9265 = 92.65%.
then there will be 92.65 occasions when this job will be completed by 18 weeks.
Check Your Progress

1. Define the preceding, succeeding and concurrent activities.
2. What is the purpose of introducing dummy activity?
3. What does the latest event time (L) indicate?
4. How is the critical path denoted in a network?

QUESTIONS
1. Activities that must be accomplished before a given event can occur, are
termed as preceding activities. Activities that cannot be accomplished until
an event has occurred, are termed as succeeding activities. Activities that
can be accomplished concurrently, are known as concurrent activities.
Self-Instructional
260 Material
2. The purpose of introducing dummy activity is: Project Scheduling and
PERT-CPM
(i) To maintain uniqueness in the numbering system, as every activity may
have a distinct set of events by which the activity can be identified.
(ii) To maintain a proper logic in the network. NOTES
3. The latest event time (L) indicates the time by which all activities entering
into that event must be completed without delaying the completion of the
project.
4. In the network critical path is denoted by a double line and identifies all the
critical activities of the project.
13.6 SUMMARY
 Network is the graphic representation of logically and sequentially connected

arrows and nodes, representing activities and events in a project. Networks
are also called arrow diagrams.
 An activity represents some action and is a time consuming effort necessary
to complete a particular part of the overall project. Thus, each and every
activity has a point of time where it begins and a point where it ends.
 The beginning and end points of an activity are called events or nodes.
Event is a point in time and does not consume any resources. It is represented
by a numbered circle. The head event called the jth event always has a
number higher than the tail event, which is also called the ith event.
 It is not necessary for an event to be the ending event of only one activity as
it can be the ending event of two or more activities. Such an event is defined
as a merge event.
 If the event happens to be the beginning event of two or more activities, it is
defined as a burst event.
 A certain event, and the same activity can be succeeding to some other
event or it may be a concurrent activity with one or more activities.
 Certain activities, which neither consume time nor resources but are used
simply to represent a connection or a link between the events are known as
dummies. It is shown in the network by a dotted line.
 In a network diagram, a looping error is also known as cycling error. Drawing
an endless loop in a network is known as error of looping. A loop can be
formed if an activity is represented as going back in time.
 After the network is drawn in a logical sequence, every event is assigned a
number. The number sequence must be such so as to reflect the flow of the
network.
Self-Instructional
Material 261
Project Scheduling and  Once the network of a project is constructed, the time analysis of the network
PERT-CPM
becomes essential for planning various activities of the project. Activity time
is a forecast of the time an activity is expected to take from its starting point
to its completion (under normal conditions).
NOTES
 The basic difference between the slack and float is that slack is used for
events only; whereas float is used for activities.
 The sequence of critical activities in a network is called the critical path. It is
the longest path in the network, from the starting event to the ending event
and defines the minimum time required to complete the project.
13.7 KEY WORDS
 Network: It is the graphic representation of logically and sequentially

connected arrows and nodes representing activities and events in a project.
 Activity: An activity represents some action and is a time consuming effort,
necessary to complete a particular part of the overall project.
 Event: The beginning and end points of an activity are called events or
nodes.
 Dummy activity: Certain activities, which neither consume time nor
resources but are used simply to represent a connection or a link between
the events, are known as dummies.
 Float: It is defined as the difference between the latest and the earliest
activity time.
 Slack: It is defined as the difference between the latest and the earliest
event time.
 Critical activity: An activity is said to be critical if a delay in its start will
cause a further delay in the completion of the entire project.
 Critical path: The sequence of critical activities in a network is called the
critical path.
 Optimistic time estimate (to): It is the smallest time taken to complete
the activity if everything goes well.
 Most likely time estimate (tm): It refers to the estimate of the normal
time the activity would take.
 Pessimistic time estimate (tp): It is the longest time that an activity would
take if everything goes wrong.
Self-Instructional
262 Material
13.8 SELF ASSESSMENT QUESTIONS AND PERT-CPM
EXERCISES
1. What do you understand by merge and burst events?

2. What are the common errors in a network construction?
3. State the rules that need to be followed in numbering the events.
4. What is the difference between free float and independent float?
5. What is critical path? Mention the conditions that must be satisfied for the
activities to lie on the critical path.
1. The following table gives the activities and duration of a construction project.
Activity 1–2 1–3 2–3 2–4 3–4 4–5
Duration (days) 20 25 10 12 6 10
(i) Draw the network for the project

(ii) Find the critical path.
2. A small project consists of 11 activities A, B, C ... K. According to the
precedence relationship A and B can start simultaneously, given A < C, D,
I; B < G, F; D < G, F; F < H, K; G, H < J; I, J, K < E. The duration of
the activities are as follows.
Duration (days) 5 3 10 2 8 4 5 6 12 8 9
Draw the network of the project. Summarise the CPM calculations in a

tabular form computing the total and free floats of activities as well as
determine the critical path.
3. Draw the network and determine the critical path for the given data. Also
calculate all the floats involved in CPM.
Jobs 1–2 1–3 2–4 3–4 3–5 4–5 4–6 5–6
Duration 6 5 10 3 4 6 2 9
Self-Instructional
Material 263
Project Scheduling and 4. A small maintenance project consists of the following 12 jobs.
PERT-CPM
Jobs 1–2 2–3 2–4 3–4 3–5 4–6 5–8 6–7 6–10 7–9 8–9 9–10
Duration (days) 2 7 3 3 5 3 5 8 4 4 1 7
NOTES Draw the arrow network of the project. Summarise CPM calculations in a
tabular form, calculating the three types of floats and hence determine the
critical path.
5. Consider the following data for activities in a given project.
6. The data for a small PERT project is as given below, where a represents
optimistic time, m the most likely time and b the pessimistic time. Estimates
(in days) of the activities A, B ... J, K are given in the table.
a 3 2 6 2 5 3 3 1 4 1 2
m 6 5 12 5 11 6 9 4 19 2 4
b 5 14 30 8 17 15 27 7 28 9 12
A, B and C can start simultaneously; A £ D, I; B < G, F; D < G, F; C < E;

E < H, K; F < H, K; G, H < J.
(i) Draw the arrow network of the project.
(ii) Calculate the earliest and the latest expected times to each event
and find the critical path.
(iii) What is the probability that the project will be completed 2 days
later than expected?
7. The three estimates for the activities of a project are given below:
Estimate duration (days)
Activity
a m b
1–2 5 6 7
1–3 1 1 7
1–4 2 4 12
2–5 3 6 15
3–5 1 1 1
4–6 2 2 8
5–6 1 4 7
Draw the project network. Find out the critical path and duration of the
project. What is the probability that the project will be completed at least 5
days earlier than expected?
What is the probability that the project will be completed by 22 days?
Self-Instructional
264 Material
8. Consider the network shown in the figure below. The estimate to, tm and tp Project Scheduling and
PERT-CPM
are shown in this order for each of the activities, on top of the arcs denoting
the respective activities.
Find the probability of completing the project in 25days.
NOTES
9. A project is represented by the network shown below and has the following
table:
Task A B C D E F G H I
Least time 5 18 26 16 15 6 7 7 3
Greatest time 10 22 40 20 25 12 12 9 5
Most likely time 8 20 33 18 20 9 10 8 4
Determine the following:

(i) Expected time of tasks and their variance.
(ii) The earliest and the latest expected time to reach each mode.
(iii) The critical path.
(iv) The probability of completing the project within 41.5 weeks.
10. Consider a project having the following activities and their time estimates.
Draw an arrow diagram for the project. Identify the critical path and compute
the expected completion time. What is the probability that the project will
require at least 75 days?
to tm tp
Activity Predecessor
days
A – 2 4 6
B A 8 12 16
C A 14 16 30
D B 4 10 16
E C, B 6 12 18
F E 6 8 22
G D 18 18 30
H F, G 8 14 32
Self-Instructional
Material 265
PERT-CPM 13.9 FURTHER READINGS

NOTES Singh, J. K. 2009. Business Mathematics. New Delhi: Himalaya Publishing
House.
Self-Instructional
266 Material
Game Theory:
UNIT 14 GAME THEORY: Competitive Situations
COMPETITIVE
NOTES
SITUATIONS
Structure
14.0 Introduction
14.1 Objectives
14.2 Basic Concepts of Game Theory
14.2.2 The Maximin-Minimax Principle
14.2.3 Games without Saddle Points (Mixed Strategies)
14.3 Dominance Property
14.5 Summary
14.6 Key Words
14.0 INTRODUCTION
Competition is the watch word of modern life. We say that a competitive solution
exists, if two or more individuals make decisions in a situation that involves
conflicting interests, and in which the outcome is controlled by the decision of all
the concerned parties. A competitive situation is called a game. The term game
represents a conflict between two or more parties. A situation is termed a game
when it possesses the following properties.
(i) The number of competitors is finite.
(ii) There is a conflict in interests between the participants.
(iii) Each of the participants has a finite set of possible courses of action.
(iv) The rules governing these choices are specified and known to all players.
The game begins when each player chooses a single course of action from
the list of courses available to him.
(v) The outcome of the game is affected by the choices made by all the players.
(vi) The outcome for all specific set of choices, by all the players, is known in
advance and numerically defined.
The outcome of a game consists of a particular set of courses of action
undertaken by the competitors. Each outcome determines a set of payments
(positive, negative or zero), one to each competitor.
Self-Instructional
Material 267
Game Theory:
Competitive Situations 14.1 OBJECTIVES

NOTES  Discuss the characteristics of game theory and competitive situations
 Explain the maximin-minimax principle
 Describe the dominance theory
14.2 BASIC CONCEPTS OF GAME THEORY
The term ‘strategy’ is defined as a complete set of plans of action specifying

precisely what the player will do under every possible future contigency that might
occur during the play of the game, i.e., strategy of a player is the decision rule he
uses for making a choice, from his list of courses of action. Strategy can be classified
as:
(i) Pure strategy
(ii) Mixed strategy
A strategy is called pure if one knows in advance of the play that it is certain
to be adopted, irrespective of the strategy the other players might choose.
The optimal strategy mixture for each player may be determined by assigning
to each strategy, its probability of being chosen. The strategy so determined is
called mixed strategy because it is a probabilistic combination of the available
choices of strategy. Mixed strategy is denoted by the set, S = {X1, X2...Xn} where,
Xj is the probability of choosing the course j such that Xj > 0, j = 1, 2...n and
X1 + X2 + ... Xn= 1. It is evident that a pure strategy is a special case of mixed
strategy.
In the case where all but one Xj is zero, a player may be able to choose only
n pure strategy, but he has an infinite number of mixed strategies to choose them.
Pay-off
Pay-off is the outcome of playing the game. A pay-off matrix is a table showing the
amount received by the player named at the left hand side after all possible plays
of the game. The payment is made by the player named at the top of the table.
If a player A has m courses of action and player B has n courses, then a payoff
matrix may be constructed using the following steps.
(i) Row designations for each matrix are the courses of action available to A.
(ii) Column designations for each matrix are the courses of action available
to B.
Self-Instructional
268 Material
(iii) With a two-person zero-sum game, the cell entries in B’s payoff matrix Game Theory:
Competitive Situations
will be the negative of the corresponding entries in A’s payoff matrix and
the matrices will be as shown below.
Player B NOTES
 1 2 3 ... j ... n 
a a1n 
 11 a12 a13 ... a1 j ...
a a22 a23 ... a2 j ... a2n 
 21 
Player A  a31 a32 a33 ... a3 j ... a3n 
 
 .. ... ... 
 . 
 m am 2 am3 ... amj ... amn 

A’s payoff matrix.
Player B
1 2 3 ... j ... n
1   a11 a12 a13   a1 j   a1 

 
2   a21  a22 a23   a2 j  a2 
 
Player A   
i   ai1  ai2  ai3   a1 j  ain 
 
  
m   am1  am 2  am 3   amj  amn 


(i) Two-person games and n-person games In two-person games, the
players may have many possible choices open to them for each play of
the game but the number of players remain only two. Hence, it is called a
two-person game. In case of more than two persons, the game is generally
called n-person game.
(ii) Zero-sum game A zero-sum game is one in which the sum of the payment
to all the competitors is zero, for every possible outcome of the game is in
a game if the sum of the points won, equals the sum of the points lost.
(iii) Two-person zero-sum game A game with two players, where the gain
of one player equals the loss of the other, is known as a two-person zero-
sum game. It is also called a rectangular game because their payoff
matrix is in the rectangular form. The characteristics of such a game are:
(a) Only two players participate in the game.
(b) Each player has a finite number of strategies to use.
(c) Each specific strategy results in a payoff.
(d) Total payoff to the two players at the end of each play is zero.
Self-Instructional
Material 269
Game Theory: 14.2.2 The Maximin-Minimax Principle
This principle is used for the selection of optimal strategies by two players. Consider
two players A and B. A is a player who wishes to maximize his gains, while player
B wishes to minimize his losses. Since A would like to maximize his minimum gain,
NOTES
we obtain for player A, the value called maximin value and the corresponding
strategy is called the maximin strategy.
On the other hand, since player B wishes to minimize his losses, a value called
the minimax value, which is the minimum of the maximum losses is found. The
corresponding strategy is called the minimax strategy. When these two are equal
(maximin value = minimax value), the corresponding strategies are called optimal
strategies and the game is said to have a saddle point. The value of the game is
given by the saddle point.
The selection of maximin and minimax strategies by A and B is based upon the
so-called maximin-minimax principle, which guarantees the best of the worst results.
Saddle point A saddle point is a position in the payoff matrix where, the maximum
of row minima coincides with the minimum of column maxima. The payoff at the
saddle point is called the value of the game.
We shall denote the maximin value by  , the minimax value of the game by 
and the value of the game by .
Note:
(i) A game is said to be fair if,
maximin value = minimax value = 0, i.e., if  =  = 0
(ii) A game is said to be strictly determinable if,
maximin value = minimax value  0.  =  =  .
Example 14.1: Solve the game whose payoff matrix is given by,
Player B
B1 B2 B3
A1  1 3 1
Player A
A2  0  4  3 
A3  1 5  1 
Solution:
Player B
B1 B2 B3 Row minima
A1  1 3 1  1
Player A
A2  0 4 3  4
A3  1 5 1  1
Self-Instructional
270 Material
Column maxima 1 5 1 Game Theory:
Maxi (minimum) = Max (1, – 4, – 1) = 1
Mini (maximum) = Min (1, 5, 1) = 1.
i.e., Maximin value  = 1 = Minimax value  NOTES
 Saddle point exists. The value of the game is the saddle point, which is 1.
The optimal strategy is the position of the saddle point and is given by, (A1, B1).
Example 14.2: For what value of 1, is the game with the following matrix strictly
determinable?
Player B
B1 B2 B3
A1   6 2 
Player A A2  1  7 
A3  2 4  
Solution Ignoring the value of l, the payoff matrix is given by,
Player B
B1 B2 B3 Row minima
A1   6 2 2
 1  7  7
Player A A2  
A3  2 4   2
Column maxima –1 6 2
The game is strictly determinable if,
 =  =  . Hence,  = 2,  = – 1
 – 1    2.
Example 14.3: Determine which of the following two-person zero-sum games

are strictly determinable and fair. Given the optimum strategy for each player in
the case of strictly determinable games.
(i) Player B (ii) Player B

B1 B2 B1 B2
A1  5 2  A1  1 1 
Player A Player A
A2  7 4  A2  4 3 
Maxi (minimum) =  = Max (– 5, – 7) = – 5

Mini (maximum) =  = Min (– 5, 2) = – 5
Self-Instructional
Material 271
Game Theory: Solution:
(i) Player B
B1 B2 Row minima
A1  5 2  5
NOTES Player A
A2  7 4  7
Column maxima –5 2
Since  =  = – 5  0, the game is strictly determinable. There exists a
saddle point = – 5. Hence, the value of the game is –5. The optimal strategy
is the position of the saddle point given by, (A1, B1).
(ii) Player B
B1 B2 Row minimum
Player A A1  1 1  1
A2  4 3  3
Column maximum 4 1
Maxi (minimum) =  = Max (1, – 3) = 1.
Mini (maximum) =  = Min (4, 1) = 1.
Since  =  = 1  0, the game is strictly determinable. Value of the game
is 1. The optimal strategy is, (A1, B2).
Example 14.4: Solve the game whose payoff matrix is given below.
 2 0 0 5 3
 3 2 1 2 2 

 4 3 0 2 6 
 
 5 3 4 2 6 
Solution:
B1 B2 B3 B4 B5 Row minima
A1  2 0 0 5 3  2

A2  3 2 1 2 2  1
Player A
A3  4 3 0 2 6  4
 
A4  5 3 4 2 6  6
Column maxima 5 3 1 5 6
Maxi (minimum) =  = Max (– 2, 1, – 4, – 6) = 1.
Mini (maximum) =  = Min (5, 3, 1, 5, 6) = 1.
Since,  =  = 1, there exists a saddle point. Value of the game is 1. The
position of the saddle point is the optimal strategy and is given by, [A2, B3].
Self-Instructional
272 Material
14.2.3 Games without Saddle Points (Mixed Strategies) Game Theory:
A game without saddle point can be solved by various solution methods.
2 × 2 Games Without Saddle Point
NOTES
Consider a 2 × 2 two-person zero-sum game without any saddle point, having the
payoff matrix for player A as,
B1 B2
A1  a11 a12 
A2  a21 a22 
The optimum mixed strategies,
A A  B B 
SA =  1 2  and SB  1 2 
 p1 p2   q1 q2 
a22  a12
where, p1 = , p1 + p2 = 1  p2 = 1 – p1
(a11  a22 )  (a12  a21 )
a22  a21
q1 = , q1 + q2 = 1  q2 = 1 – q1
(a11  a22 )  (a12  a21 )
a11 a22  a12 a21
The value of the game () = .
(a11  a22 )  (a12  a21 )
Example 14.5: Solve the following payoff matrix. Also determine the optimal
strategies and value of the game.
B
5 1
A 
3 4
Solution:
B
 5 1  . Let this be,
A 
3 4
B1 B2
A1  a11 a12  . The optimum mixed strategies,
A2  a21 a22 
 A A  B B 
SA =  1 2  and SB =  1 2 
 p1 p2   q1 q2 
a22  a21 43 1
where, p1 = = 
(a11  a22)  (a12  a21) (5  4)  (1  3) 5
1 4
p2 = 1 – p1  p2 = 1 – =
5 5
Self-Instructional
Material 273
Game Theory:
a22  a12 4 1 3
Competitive Situations q1 = = 
(a11  a22)  (a12  a21) (5  4)  (1  3) 5
3 2
q2 = 1 – q1  q2 = 1 – =
NOTES 5 5
(5  4)  (1  3) 17
value of game, = =
(5  4)  (1  3) 5
 The optimum mixed strategies,
SA =  ,  ; SB = 3
1 4 2
 , 
5 5
  5 5
17
Value of game = .
5
Example 14.6: Solve the following game and determine its value.
B
 4 4 
A 
 4 4 
Solution: It is clear that the payoff matrix does not possess any saddle point.
The players will use mixed strategies. The optimum mixed strategy for the players
are,
 A A 
SA =  1 2  , p1 + p2 = 1
 p1 p2 
B B 
SB =  1 2  , q1 + q2 = 1.
 q1 q2 
a22  a21 4  ( 4) 8 1
p1 = = = 
(a11  a22 )  (a12  a21 ) 4  4  ( 4  4) 16 2
1 1
p2 = 1 – p1  p2 = 1 – =
2 2
a22  a12 4  ( 4) 8 1
q1 = = = 
(a11  a22 )  (a12  a21 ) 4  4  (  4  4) 16 2
1 1
q2 = 1 – q1 = 1 – = .
2 2
The optimum strategy is, SA =  ,  1

1 1 1
 ; SB =  
2 2 2 2
a22 a11  a12 a21
The value of the game is,  =
(a22  a11 )  (a12  a21 )
(4  4)  [ 4  (  4)]
= = 0.
(4  4)  [ 4  ( 4)]
Self-Instructional
274 Material
Example 14.7: In a game of matching coins with two players, suppose A wins Game Theory:
one unit of value when there are two heads, wins nothing when there are two tails
and losses 1/2 unit of value when there are one head and one tail. Determine the
payoff matrix, the best strategies for each player and the value of the game to A.
Solution: The payoff matrix for the player A is given by,
NOTES
Player B
H T
 1
H 1  
Player A 2
 
 1
T  0
 2 
Let this be,
B1 B2
A1  a11 a12 .
 
A2  a21 a22 
 A A 
The optimum mixed strategies,SA =  1 2  , p1 + p2 = 1
 p1 p2 
 B1 B2 
SB =  q q  , q1 + q2 = 1.
 1 2 
 1
0   
a22  a21  2
p1 = =
(a11  a22)  (a12  a21)  1 1
1 0    
 2 2
1
1
= 2 =
2 4
1 3
p2 = 1 – p1 = 1 – = .
4 4
 1
0   
a22  a12  2 1
q1 = = =
(a11  a22)  (a12  a21)  1 1  4
1 0    
 2 2
1 3
q2 = 1 – q1 = 1 – =
4 4
 1 1
1  0        1
 2 2  4
Value of the game =
 1 1 2
1 0    
 2 2 
1
= 
8
Self-Instructional
Material 275
Game Theory:  The optimum mixed strategy is given by,
SA =  ,  ; SB =  , 
1 3 1 3
4 4 4 4
and the value of game is –1/8.
NOTES
Graphical Method for 2 × n or m × 2 Games

Consider the following 2 × n games.
B
 B1 B2 ... Bn 
A1 
A a a ... a1n 
A2  11 12
 a21 a22 ... a2 n 
Let the mixed strategy for player A be given by,
A A 
SA =  1 2  such that, p1 + p2 = 1, p1, p2  0
 p1 p2 
Now, for each of the pure strategies available to B, expected payoff for player
A would be as follows.
B’s pure move A’s expected payoff E(p)
B1 E1 (p) = a11 p1 + a21 p2
B2 E2 (p) = a12 p1 + a22 p2
Bn En (p) = a1n p1 + a2n p2
The player B would like to choose that pure move Bj against SA for which Ej (p)
is a minimum for j = 1, 2 ... n. Let us denote this minimum expected payoff for A
by,
 = Min (Ej(p)) j = 1, 2 ... n.
The objective of player A is to select P1 and P2 in such a way that, is as large
as possible. This may be done by plotting the straight lines,
Ej(p) = a1j p1– a2j p2 = (a1j – a2j) p1+ a2j
j = 1, 2.... n.
as linear functions of P1.
The highest point on the lower boundary of these lines will give the maximum
value among the minimum expected payoffs on the lower boundary (lower
envelope) as well as the optimum value of probability P1 and P2.
Now the two strategies of player B corresponding to the lines that pass through
the maximin point can be determined. It helps in reducing the size of the game to
(2 × 2).
Similarly, we can treat m × 2 games in the same way and get the minimax point,
which will be the lowest point on the upper boundary (upper envelope).
Self-Instructional
276 Material
Example 14.8: Solve the following 2 × 3 game graphically. Game Theory:
Player B
1 3 11
Player A  
8 5 2  NOTES
Solution: Since the problem does not possess any saddle point, let player A
play by the mixed strategy
A A 
SA =  1 2  with, p2 = 1 – p1
 p1 p2 
against player B.
A’s expected payoff against B’s pure move is given by,
B’s pure move A’s expected payoff
E(p1)
B1 E1(p1) = p1 + 8(1 – p1) = – 7p1 + 8
B2 E2(p1) = 3p1 + 5(1 – p1) = 2p1 + 5
B3 E3(p1) = 11P1+ 2(1 – P1) = 9p1 + 2.
These expected payoff equations are then plotted as functions of P1 as shown
in the Fig. 14.1, which shows the payoffs of each column represented as points on
two vertical axes 1 and 2, of unit distance apart.
Fig. 14.1
Now, since player A wishes to maximize his minimum expected payoff, we

consider the highest point of intersection H on the lower envelope of A’s expected
payoff equation. The lines B2 and B3 passing through H define the relevant moves
that B2 and B3 alone need to play. The solution to the original 2 × 3 game
reduces to,
B2 B3
A1  3 11 
A2  5 2 
Self-Instructional
Material 277
Game Theory: The optimum strategy for A and B is given by,
A A 
SA =  1 2  p1 + p2 = 1
 p1 p2 
B B B 
NOTES SB =  1 2 3  q1  q2 = 1
 0 q1 q2 
25 3 3
p1 =  
3  2  (11  5) 11 11
3 3
p2 = 1 – p1 = 1  
11 11
2  11 9 9
q1 =  
11 11 11
9 2
q2 = 1  
11 11
A1 A2 B1 B2 B3
SA = 
3 8  9 2
 and S B  0 
 11 11 11 11 
6  11 49
Value of the game, =  .
11 11
Example 14.9: Solve the game problem graphically.

 1 3 
 3 5 
 
 1 6 
 
 4 1 
 2 2 
 
 5 0 
Solution: The given problem does not possess any saddle point. Therefore, let
player B play by the mixed strategy
B B 
SB =  1 2  with, q2 = 1 – q1
 q1 q2 
against player A.
The expected payoff equations are plotted in the Fig. 14.2 with two axes I and
II vertically at unit distance apart.
Self-Instructional
278 Material
Game Theory:
NOTES
Fig. 14.2
Since player B wishes to minimize his maximum expected payoff, we consider

the lowest point of the upper boundary of B’s expected payoff equation. The point
H (intersection of lines A2 and A4) represents the minimax expected value of the
game for player B. Hence, the solution to the original 6 × 2 game reduces to the
2 × 2 payoff matrix.
Player B
B1 B2
A2  3 5 
Player A  
A4  4 1 
The optimal mixed strategy for A and B is given by,
A A A3 A4 A5 A6 
SA =  1 2  with, p1 + p2 = 1
 0 p1 0 p2 0 0 
B B 
SB =  1 2  q1  q2 = 1
 q1 q2 
1 4 3 3
P1 =  
3  1  (5  4) 5 5
3 2
P2 = 1 – 
5 5
1  5 4 4
q1 =  
5 5 5
4 1
q2 = 1 – q1 = 1  
5 5
 1A , A 2 , A3 , A4 , A5 , A6 
SA =  
 0 3 2
0 0 0 
 5 5 
Self-Instructional
Material 279
Game Theory:
 B1 B2 
SB =  4 1 

 
5 5 
3 1  5  4 3  20 17
NOTES Value of the game,  = = =
(3  1)  (5  4) 5 5
17
 =
5
14.3 DOMINANCE PROPERTY

Sometimes it is observed that one of the pure strategies of either player is always
inferior to at least one of the remaining. The superior strategies are said to dominate
the inferior ones. In such cases of dominance, we reduce the size of the payoff
matrix by deleting those strategies, which are dominated by others. The general
rules for dominance are:
(i) If all the elements of a row, say kth row, are less than or equal to the
corresponding elements of any other row, say rth row, then kth row is
dominated by the rth row.
(ii) If all the elements of a column, say kth column, are greater than or equal to
the corresponding elements of any other column, say rth column, then the
kth column is dominated by the rth column.
(iii) Dominated rows and columns may be deleted to reduce the size of the
payoff matrix as the optimal strategies will remain unaffected.
(iv) If some linear combinations of some rows dominate ith row, then the ith
row will be deleted. Similar arguments follow for columns.
Example 14.10: Using the principle of dominance, solve the following game.
Player B
 3 2 4 
 
Player A  1 4 2 
 2 2 6 
Solution: In the given payoff matrix, all the elements in the third column are
greater than or equal to the corresponding elements in the first column. Therefore,
column three is dominated by first column. Delete column three. The reduced
payoff matrix is given by,
Player B
 3 2 
Player A  1 4 
 2 2 
Since no row (or column) dominates another row (or column). The 3 × 2 game
can now be solved by the graphical method. Since player B wishes to minimize his
Self-Instructional
280 Material
maximum loss, we find the lowest point of the upper boundary. The expected Game Theory:
payoff equations are then plotted as shown in Fig. 14.3.
NOTES
Fig. 14.3
The lowest point in the upper boundary is given by the intersection of lines A1
and A2. The solution in the original game is reduced to a 2 × 2 matrix.
B1 B2
A1  3 2 
A3  2 2 
The optimum strategy for A and B is given by,
A A2 A3 
SA =  1  p  p2 =1
 p1 0 p2  1
B B 
SB =  1 2  q1  q2 = 1
 q1 q2 
22
p1 = =0
3  2  (2  2)
p2 = 1 – p1 = 1 – 0 = 1
22 4
q1 = =
3  2  (2  2) 5
4 1
q2 = 1  q1  1 
5 5
 B1 B2 
A A2 A3 
SA =  1  and SB =  4 1 

0 0 1  
5 5 
3  2  (2)  (2) 2
Value of the game,  =  .
3  2  (2  2) 5
Self-Instructional
Material 281
Game Theory: Example 14.11: Solve the following game.
Player B
1 7 2
Player A 6 2 7
NOTES  
 5 1 6 
Solution: Since all the elements in the third row are less than or equal to the
corresponding elements of the second row, therefore, the third row is dominated
by the second row. Delete this dominated row. The reduced payoff matrix is given
by,
Player B
1 7 2 
Player A  
6 2 7 
The elements of the third column are greater than or equal to the corresponding
elements of the first column, which give that column third is dominated by column
one. This dominated column is deleted and the reduced payoff matrix is given by,
Player B
Player A 1 7 
6 2
 
The reduced payoff matrix is a 2 × 2 matrix. The optimal strategy for players A
and B is given by,
A A A3 
SA =  1 2  p  p2 = 1
 p1 p2 0 1
B B B3 
SB =  1 2  q  q2 = 1
 q1 q2 0 1
26 4 2
p1 =  
2  1  (7  6) 10 5
2 3
p2 = 1  
5 5
26 5 1
q1 =  
2  1  (7  6) 10 2
1 1
q2 = 1  q1  1  
2 2
2 1  7  6 40
Value of the game, =  = 4.
2  1  (7  6) 10
The optimal strategy is given by,
 A1 A2 A3 
SA =  2 3

 0 
5 5 
 B1 B2 B3 
SB =  
 1 1
0 
2 2 
Value of the game is,  = 4.
Self-Instructional
282 Material
Example 14.12: Is the following two-person zero-sum game stable? Solve the Game Theory:
game.
Player B
 5 10 9 0 
6 7 8 1  NOTES
Player A 
8 7 15 1 
 
3 4 1 4 
Solution Since the game has no saddle point, it is not stable. All the elements of
the first row and the second row are  to the corresponding elements of the third
row. Hence, these two rows are dominated rows. Deleting these two rows from
the payoff matrix, the reduced payoff matrix is given by,
Player B
 8 7 15 1 
Player A  3 4 1 4 
 
In this modified payoff matrix, we observe that all the elements of the second
column are  to the corresponding elements of the fourth column. Hence, this
dominated column (2nd column) is deleted from the payoff matrix. The reduced
payoff matrix is given by,
 8 15 1
Player A  
 3 1 4 
Now we observe that no row or column dominates another row or column.
However, we note that a convex combination of 2nd and 3rd columns is given by,
1 1
15   1 = 8  8
2 2
1 1 3
1   4  = 3
2 2 2
and hence the elements of the first column are greater than or equalto the
corresponding elements of this combination. Deleting this dominated column, the
reduced payoff matrix is given by,
Player B
 15 1 
Player A  
 1 4 
A B 
SA =  3 4  , p1  p2 = 1
 p1 p2 
4  (1) 5
p1 = 
15  4  (1  1  1) 19
5 14
p2 = 1  
19 19
B B 
SB =  3 4  q1  q2 = 1
 q1 q2 
Self-Instructional
Material 283
Game Theory: 4 1 3
Competitive Situations q1 = 
19 19
3 19
q2 = 1  
19 16
NOTES The optimum strategy of the given payoff matrix is given by,
 A1 A2 A3 A4   B1 B2 B3 B4 
SA =  5
 S 
14  B  3

16 
0 0  0 0 
 19 19   19 19 
(4  15)  (1   1) 61
and the value of game is,  .
19 19
Check Your Progress

1. How is the optimal strategy mixture determined by each player?
2. What are the characteristics of two-person zero-sum game?
3. What do you understand by the value of game?

QUESTIONS
1. The optimal strategy mixture for each player may be determined by assigning
to each strategy, its probability of being chosen. The strategy so determined
is called mixed strategy because it is a probabilistic combination of the
available choices of strategy.
2. The characteristics of two-person zero-sum game are the following:
(a) Only two players participate in the game.
(b) Each player has a finite number of strategies to use.
(c) Each specific strategy results in a payoff.
(d) Total payoff to the two players at the end of each play is zero.
3. The payoff at the saddle point is called the value of the game.
14.5 SUMMARY
 The term ‘strategy’ is defined as a complete set of plans of action specifying

precisely what the player will do under every possible future contigency
that might occur during the play of the game, i.e., strategy of a player is the
decision rule he uses for making a choice, from his list of courses of action.
Strategy can be classified as pure strategy and mixed strategy.
Self-Instructional
284 Material
 Pay-off is the outcome of playing the game. A pay-off matrix is a table Game Theory:
showing the amount received by the player named at the left hand side after
all possible plays of the game. The payment is made by the player named at
the top of the table.
NOTES
 In two-person games, the players may have many possible choices open to
them for each play of the game but the number of players remains only two.
Hence, it is called a two-person game. In case of more than two persons,
the game is generally called n-person game.
 A zero sum game is one in which the sum of the payment to all the competitors
is zero for every possible outcome of the game is in a game if the sum of the
points won equals the sum of the points lost.
 A game with two players, where the gain of one player equals the loss of the
other is known as a two-person zero sum game. It is also called a rectangular
game because the pay-off matrix is in a rectangular form.
 A saddle point is a position in the pay-off matrix, where the maximum of
row minima coincides with the minimum of column maxima. The pay-off at
the saddle point is called the ‘value of the game’.
 It is observed that one of the pure strategies of either player is always inferior
to at least one of those remaining. The superior strategies are said to dominate
the inferior ones. In such cases of dominance, we reduce the size of the
pay-off matrix by deleting those strategies which are dominated by others.
 The maximin-minimax principle is used for the selection of optimal strategies
by two players.
14.6 KEY WORDS
 Pure strategy: A strategy is called pure if one knows in advance of the

play that it is certain to be adopted, irrespective of the strategy the other
players might choose.
 Zero-sum game: A zero-sum game is one in which the sum of the payment
to all the competitors is zero, for every possible outcome of the game is in a
game if the sum of the points won, equals the sum of the points lost.
 Two-person zero-sum game: A game with two players, where the gain of
one player equals the loss of the other, is known as a two-person zero-sum
game.
 Saddle point: A saddle point is a position in the payoff matrix where, the
maximum of row minima coincides with the minimum of column maxima.
Self-Instructional
Material 285
Game Theory:
Competitive Situations 14.7 SELF ASSESSMENT QUESTIONS AND
EXERCISES
NOTES Short Answer Questions

1. Determine which of the following two-person zero-sum games are strictly
determinable and fair. Give the optimum strategies for each player in the
case of strictly determinable games.
(i) Player B (ii) Player B
0 2 0 2
Player A Player A
–1 4 –1 4
2. For a game with the following payoff matrix,

Player A
–1 2 –2
Player B 6 4 –6
determine the best strategies as well as the value of the game for players A
and B. Is this game (i) fair, (ii) strictly determinable?
3. Determine the optimal minimax strategies for each player in the following
game.
B1 B2 B3 B4
A1 –5 2 0 7
A2 5 6 4 8
A3 4 0 2 –3

1. For a game with the following payoff matrix, determine the optimal strategy
and the value of the game.
(i) B
6 –3
A
-3 3
(ii) B
2 5
A
4 1
2. Two players, A and B match coins. If the coins match, then A wins two units
of value. If coins do not match, then B wins two units of value.
Determine the optimum strategies for the players and the value of the game.
Self-Instructional
286 Material
3. Consider a ‘modified form’ of matching-based coins game problem. The Game Theory:
matching player is paid 800 if both the coins turn heads and Re. 1 if both
the coins turn tails. The non-matching player is paid 300 when the two
coins do not match. Given the choice of being the matching or non-matching
player, which one would you choose and what would be your strategy? NOTES
4. Solve the following problems graphically.
Player B
3 –3 4
(i) Player A
–1 1 –3
Player B
1 3 –3 7
(ii) Player A
2 5 4 –6
Player B
1 2
5 4
(iii) Player A –7 9
–4 –3
2 1
B1 B2
73
SB = 5 12 ; =
, 17
17 17
Player B
–6 7
4 –5
(iv) Player A –1 –2
–2 5
7 6
B1 B2
23
SB = 11 9 ;γ=
30
20 20
5. The companies A and B are competing for the same product. Their different
strategies are given in the following payoff matrix.
Company B
4 –3 3
Company A
–3 1 –1
Determine the best strategies for the two companies.
6. Using dominance, solve the payoff matrix, given by,
Self-Instructional
Material 287
Game Theory: Player B
Competitive Situations 2 2 4 1
6 1 12 3
(i) Player A
3 2 0 6
NOTES 2 3 7 7
Player B
1 7 3 4 
 
(ii) Player A 5 6 4 5
7 2 0 3 
B1 B2 B3 B4 B5
A1  4 4 2 4 6 
 6 8 4 0 
(iii) A2  8
A3  10 2 4 0 12 
7. The following matrix represents the payoff to P1 in a rectangular game

between two persons P1 and P2.
P2
 8 15 4 –2 
P1 19 15 17 16 
 0 20 15 5
By the notion of dominance, reduce the game to a 2 × 4 game and solve it

graphically.

House.
Self-Instructional
288 Material

1 - M - Com - (F&C) - 335 21 - Operations Research PDF

Uploaded by

Copyright:

Available Formats

You might also like

1 - M - Com - (F&C) - 335 21 - Operations Research PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

1 - M - Com - (F&C) - 335 21 - Operations Research PDF

Uploaded by

Copyright:

Available Formats

ALAGAPPA UNIVERSITY

[Accredited with ‘A+’ Grade by NAAC (CGPA:3.64) in the Third Cycle

M.Com. [Finance & Control]

"The copyright shall be vested with Alagappa University"

Vikas® is the registered trademark of Vikas® Publishing House Pvt. Ltd.

BLOCK - I: OPERATIONS RESEARCH AND LINEAR PROGRAMMING

Unit 6: Transportation Problem: Formulation of Transportation Problem Unit 6: Transportation Problem

BLOCK - II: INTEGER PROGRAMMING AND QUEUING CONCEPTS

BLOCK - III: SIMULATION GAME THEORY CONCEPTS

BLOCK II: INTEGER PROGRAMMING AND QUEUING CONCEPTS

BLOCK III: SIMULATION GAME THEORY CONCEPTS

Operational Research, or simply OR, originated in the context of military operations,

UNIT 1 OPERATIONS RESEARCH

Modern technological progress is accompanied by growth in scientific techniques.

1.2 HISTORICAL BACKGROUND

The term, ‘Operations Research’ was first coined by well-known authors J. F.

Check Your Progress

1.3 SCIENTIFIC METHODS IN OPERATIONS

The methodology of operations research generally involves the following steps:

opportunity for learning.

Facts, opinions Define the

Develop objective Maximize OR

Fig. 1.1 Flow Chart Showing OR Approach

Check Your Progress

Check Your Progress

1.5 TYPES OF MODELS IN OPERATIONS

After problem formulation, modelling plays an important part in OR. In fact, it is

teaching purposes are scaled up. Similarly, maps, photographs, drawings

production, Yi is equal to Y1 + Y2. This is nothing but an accounting statement.

project. Modelling, thus, helps in finding avenues for research and

Check Your Progress

1.6 APPLICATION AREAS OF OPERATIONS

patients, can be reached through OR. Similarly, for operating a transportation

involves the use of mathematical models, equations and similar other

problem is characterized by the need for designing a procedure to

Check Your Progress

1.7 ANSWERS TO CHECK YOUR PROGRESS

1. The essential characteristics of operations research are as follows:

1.9 KEY WORDS

 Gantt chart: It is a type of bar chart, adapted by Karol Adamiecki in 1896

1.10 SELF ASSESSMENT QUESTIONS AND

1.11 FURTHER READINGS

Sharma, A. 2009. Operations Research. New Delhi: Himalaya Publishing House.

UNIT 2 LINEAR PROGRAMMING

2.2 FORMULATION OF LINEAR PROGRAMMING

problems. Production planning and scheduling, transportation, sales and advertising,

Subject to the constraints,

form. This model can be interpreted as a profit maximization situation, where n

a11 x1  a12 x2    a1n xn (, , )b1 

Here, the constraints can be inequality  or  or even in the form of an

2.2.5 Case Studies

3.00 and on model M2 is 4.00. Whatever is produced in a week is sold in the

Subject to, 10 X 1  2 X 2  2100

Example 2.3: A person requires 10, 12 and 12 units of chemicals A, B and C,

minimize the cost.

Hence, the complete LPP is,

Non-negative restriction X1, X2  0

100  X 1  150, 200  X 2  0,50  X 3  0