MAM-056

Indira Gandhi National Open University

Qualitative and Quantitative
School of Agriculture
Analysis for Agribusiness

Block

3
OPERATIONS RESEARCH
UNIT 9
Overview of Operations Research
UNIT 10
Decision Theory
UNIT 11
Transportation Model and Assignment Problems
UNIT 12
Inventory Control
UNIT 13
Game Theory and Network Analysis
PROGRAMME DESIGN COMMITTEE
Prof. R. P. Das, PVC, IGNOU Dr. Leena Singh, Assistant Professor,
SOMS, IGNOU
Prof. S.K. Yadav, Director, SoA, IGNOU
Prof. Sunil Gupta, SOMS, IGNOU
Dr. B.K. Sikka, Former Dean, College of
Agribusiness Management, GBPUAT Dr. P. Vijayakumar, Associate Professor,
SoA, IGNOU
Dr. V.C. Mathur, Former Professor and
Head, Div. of Agri. Econ., IARI Dr. Mita Sinhamahapatra, Associate
Professor, SoA, IGNOU
Dr. Pramod Kumar, Principal Scientist
(Agri. Econ.), IARI Dr. Mukesh Kumar, Assistant Professor,
SoA, IGNOU
Prof. M. K. Salooja, School of Agriculture,
IGNOU Dr. P. K. Jain, Associate Professor and
Programme Coordinator, SoA, IGNOU
Dr. Anjali Ramtake, Associate Professor,
SOMS, IGNOU

Programme Coordinator: Dr. Praveen Kumar Jain

Block Preparation Team

Unit Writers Editors
Dr. Girish Kumar Jha,
Units 9, 10, 11, 12, & 13
Principal Scientist,
Dr. Sachin S Kamble, Division of Agricultural Economics,
NITIE, Mumbai IARI, New Delhi

Dr. Praveen Kumar Jain

Course Coordinator: Dr. Praveen Kumar Jain

MATERIAL PRODUCTION
October, 2022
© Indira Gandhi National Open University
ISBN:
All rights reserved. No part of this work may be reproduced in any form, by mimeograph or any other means,
without permission in writing from the copyright holder.
The University does not warrant or assume any legal liability or responsibility for the academic content of this
course provided by the authors as far as the copyright issues are concerned.
Further information on the Indira Gandhi National Open University courses may be obtained from the University’s
office at Maidan Garhi, New Delhi-110 068 or the official website of IGNOU at www.ignou.ac.in.
Printed and published on behalf of Indira Gandhi National Open University, New Delhi by the Director, School of
Agriculture.
Laser typeset by
Printed at
BLOCK 3 OPERATION RESEARCH
Operations Research (OR) is a scientific method of providing an analytical and objective basis
for decision-making for executives, usually under scarce resources. In the present context, the
business situation has become complex owing to an increased number of factors influencing the
decision, the situation of uncertainty, knowledge explosion, etc., making OR more relevant.
Decision theory has become an important technique for decision-makers in two different
environments namely, decision-making under uncertainty and under risk. Decision trees are used
to analyze sequential decision-making situations. The most prominent OR technique is linear
programming (LP). It is designed for models with linear objective and constraint functions. The
transportation model which deals with the objective of minimization of total transportation costs
in a business is an application of an LP. The assignment problem is a special type of
transportation problem to obtain a minimum assignment cost subject to demand and supply
restrictions. Inventory management is the planning and controlling of inventories to minimize
inventory costs. Many operations research situations can be modeled and solved as networks
(nodes connected by branches). CPM (Critical Path Method) and PERT (Programme Evaluation
and Review Technique) are network-based methods designed to assist in the planning,
scheduling, and control of projects.

Game theory deals with decision situations in which two intelligent opponents with conflicting
objectives are vying to outdo one another. This block of Operations Research comprises five
units.

Unit 9 introduces the meaning and importance of operations research, including the scope and
techniques of OR, interaction between management and research, and phases and limitations of
OR. It stresses that formulating the problem correctly is the most important phase of practicing
OR.

Unit 10 provides the meaning and types of decision rules that are available to decision-makers. It
covers the decision theory procedures designed to solve problems with a limited number of
decision alternatives and a finite list of possible states of nature.

Unit 11 deals with the meaning, assumptions, and formulation of Transportation Models,
including initial basic solutions, and the case of unbalanced and transshipment problems. It
explains the meaning and formulation of the assignment problem, including the unbalanced
situation.

Unit 12 explains the concept of inventory management, including the types of inventory and its
costs. It describes the various inventory management techniques.

Unit 13 introduces the assumptions, basic terminology, concept, and applicability of the game
theory. It explains the Two Person Zero Sum Games and its solutions by dominance. It discusses
the critical path method and programme evaluation and review technique in agribusiness
projects.

The material provided in this block is supplemented with various examples and activities to
make the learning process simple and interesting. We have also provided Check Your Progress
questions for the self-test at a few places in these units which invariably lead to possible answers
to the questions set in those exercises. What perhaps you ought to do, is to go through units and
jot down important points as you read, in the space provided in the margin. This will help you in
assimilating the content. A list of reference books has been provided at the end of each unit for
further detailed reading.
UNIT 9 OVERVIEW OF OPERATIONS RESEARCH
Structure

9.0 Objectives
9.1 Introduction
9.2 Meaning of Operations Research
9.3 Importance of Operations Research
9.4 Scope of Operations Research
9.4.1 Techniques of Operations Research
9.5 Interactions between Management and Operations Research
9.6 Phases of Operation Research
9.7 Limitations of Operations research
9.8 Let Us Sum Up
9.9 Keywords
9.10 Suggested Further Readings / References
9.11 Answers to Check Your Progress
9.12 Unit End Questions

9.0 OBJECTIVES
After studying this unit, you should be able to:
 discuss the meaning and importance of operations research;
 explain the interaction between management and operations research;
 identify the phases of operations research; and
 highlight the limitations of operations research.

9.1 INTRODUCTION
A project manager is working on the installation of a sulphur recovery unit in one of his
client refineries. The client requires him to complete the installation before the
scheduled time because of the mounting pressure from the market to supply the
product. The project manager is constrained by the following factors;
 Availability of the skilled manpower
 Availability of the equipment for installing the plant machinery

1
If in case additional resources are deployed in terms of additional manpower and
equipment deployment for the project, the project may be completed before its
scheduled completion date but the project costs are going to rise considerably. Dealing
with the situation, is it possible to finish the project before the completion time within
the constraints? Operations research deals with such types of situations and provides us
with the best possible solutions on how to have maximum utilization of the available
resources.
The term operations research was first coined in 1940 by McClosky and Trefthen in the
small town Bowdsey in the United Kingdom. This new science came into existence in a
military context. It was during World War II, that scientists were called from various
disciplines by military management and organized into teams to assist in solving
strategic and tactical problems in the field of warfare management. They aimed to
arrive at optimal utilization of scarce military resources and efforts while providing
effective decisions. This operation of optimization of the resources using mathematical
and scientific techniques was called Operations Research (OR) or operational research.
OR provides a quantitative analysis of the problem from which the management can
make an objective decision. OR has drawn upon skills from mathematics, engineering,
business, computer science, economics, and statistics to contribute to a wide variety of
applications in business, industry, government, and military. OR methodologies and
their applications continue to grow and flourish in a number of decision-making fields.
In this Unit, you will understand the meaning of Operations Research, its importance in
the field of management, its applications, and various limitations.

9.2 MEANING OF OPERATIONS RESEARCH

From the historical summary of developments that have been taken in the field of
operations research, it is apparent that the term “operations research” has a number of
quite distinct variations of meaning. OR had been defined so far in various ways and it
is perhaps still too young to be defined in some authoritative way. It is not possible to
give uniformly acceptable definitions of OR. A few opinions about the definition of OR
are given below.
OR is a scientific method of providing executive departments with a quantitative basis
for decisions regarding the operations under their control. [Morse and Kimbal (1946)]
OR is the scientific method of providing the executive with an analytical and objective
basis for decisions. [P.M.S Blackett (1948)]
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 & M. J Netzorg]
OR is an aid for the executive in making his decisions by providing him with
quantitative information based on the scientific method of analysis. [C. Kittee]

2
OR is the scientific knowledge through interdisciplinary team effort to determine the
best utilization of limited resources. [H.A. Taha]
As per The Operational Research Society of Great Britain “Operational research 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 the industry,
business, government, and defense”. 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 decisions, strategies, or
controls. The purpose is to help management determine its policy and actions
scientifically.
The Operations Research Society of America too has adopted a shorter, but similar,
description: Operations research is concerned with scientifically deciding how to best
design and operate man-machine systems, usually under conditions requiring the
allocation of scarce resources.
In general, most of the definitions of OR emphasize its methodology, namely its unique
approach to problem-solving, which may be due to the use of interdisciplinary teams or
due to the application of scientific and mathematical models. In other words, each
problem may be analyzed differently, though the same basic approach of operations
research is employed.
As more research went into the development of OR, the researchers were able to
classify to some extent many of the important management problems that arise in
practice. Examples of such problems are those relating to allocation, inventory,
network, queuing, replacement, scheduling, and so on.
The theoretical research in OR concentrated on developing appropriate mathematical
models and techniques for analyzing these problems under different conditions. Thus,
whenever a management problem is identified as belonging to a particular class, all the
models and techniques available for that class can be used to study that problem. In this
context, one could view OR as a collection of mathematical models and techniques to
solve complex management problems. Hence, it is very common to find OR courses in
universities emphasizing different mathematical techniques of operations research such
as mathematical programming, queuing theory, network analysis, dynamic
programming, inventory models, simulation, and so on.
Hence to summarise, it can be observed that the various definitions given above, bring
out the following essential characteristics of operations research:
(i) System Orientation
(ii) Use of Interdisciplinary terms
(iii) Application of the scientific method
(iv) Uncovering new problems
(v) Quantitative solutions

3
 (vi) Human factors

Check Your Progress 9.1

Note:

1. The main objective of OR is to provide a ……….. basis for decision-makers.

2. OR imbibes …………….. team approach

3. OR increases the effectiveness of ……………….. ability.

4. OR gives a qualitative solution (True / False)

…………………………………………………………………

5. OR methodology consists of definition, solution, and validation only (True /

False)

…………………………………………………………………

6. Operations research is

a) Multi-disciplinary

b) Scientific

c) Intuitive

d) All of the above

7. Managerial decisions are based on

a) An evaluation of quantitative data

b) The use of quantitative factors

c) Numbers produced by formals models

d) All of the above

8. OR approach is typically based on the use of

a) Physical model

b) Mathematical model

c) Iconic model

d) Descriptive model

9. The scientific method in OR study generally involves

4
 a) Judgement Phase

b) Research Phase

c) Action Phase

d) All of the above

10. The qualitative approach to decision analysis relies on

a) Experience

b) Judgment

c) Intuition

d) All of the above

11. Operations research practitioners do not

a) Take responsibility for solution implementation

b) Collect essential data

c) Predict future actions/operations

d) Build more than one model

9.3 IMPORTANCE OF OPERATIONS RESEARCH

After having studied what is operations research, we shall now try to understand why is
Operations Research important. The following is the list of factors that makes the use
of operation research techniques important to the decision-makers in the industry.
i. Complexity: The number of factors influencing a decision has increased thus
making the situations complex. Also, the interactions between these variables
become complicated to understand. For instance, consider a factory production
schedule which has to take into account the following factors while making the best
decision
 Market demand,
 Raw materials requirements,
 Equipment availability
 Availability of labor
ii. Scattered responsibility and authority: In a big industry, responsibility and
authority of decision-making are scattered throughout the organization and thus the

5
 organization, if it is not conscious, may follow inconsistent goals. Mathematical
quantification of OR overcomes this difficulty to a great extent.
iii. Uncertainty: With the increasing dynamism in the business environment, decision-
makers are required to deal with the situation of uncertainty. OR helps to validate
the decisions before their implementation.
iv. Knowledge explosion: OR techniques rely on quality data. Decision-makers are
required to collect the latest information for analysis purposes which is quite useful
for the industries. The collected data adds to the knowledge repository of the
organization which can be later used for any other purposes.

Exhibit 4.1. Application of OR in day-to-day activities

In Higher Education Administration
OR addresses many problems in higher education administration, which include resource
allocation accreditation, number of academic staff; distribution of academic staff; the
number of non-academic staff; the number of graduate research assistants, salary increases,
and total payroll budget.
Financial planning has become an area that lends itself naturally to OR techniques. The
importance of decisions and future repercussions which can occur as a result of inefficient,
irrational decisions has invited much research into the construction of OR models to assist
decision-makers in this area.
In Public Utilities
OR approach is directly applicable to business and society. It is also equally applicable to
big and small organizations.
In Defence
OR has a wide scope for application in defense operations. All the defense operations are
carried out by different agencies, namely the air force, army, and navy. Operations
research is helpful for achieving the desired goals of different agencies.
In Planning
OR is helpful for the planning of various activities of the organization. Planning is an
important function of management. Without effective planning, we cannot achieve the
desired goals.
Research and Development
OR is very helpful for determining areas of thrust for research and development, it helps in
selecting the criteria for a specific project.
Retail Management
Modern and other forms of Retail need planning for merchandise optimization, demand
planning, product mix, distribution, etc. as in this format on a day-to-day bases the

6
 organization has to deal with a customer. OR is widely used in online retail and digital
marketing also.
Airlines Industry
OR is used in Airlines Industry in form of optimization techniques and decision support
systems applications in air transportation. It covers major operations in the air
transportation industry for example Demand forecasting, Network design, Revenue
management, Route planning, Airline schedule planning, Irregular operations aircraft
routing, Integrated scheduling, Real-time crew management, Crew pairing optimization,
Air traffic flow management, Crew rostering, Airport traffic simulation and control, etc.

9.4 SCOPE OF OPERATIONS RESEARCH

OR is useful in the various important fields like:
 Agriculture
 Accounting & Finance
 Marketing
 Personnel Management
 Production Management
 Supply chain management
 Material management
 Research and Development etc.

9.4.1 Techniques of Operations Research

As mentioned earlier, operations research finds extensive application in various areas.
Gupta & Hira (2007) and Sharma (2010) have outlined the following OR techniques
used for problem-solving in the industry.
i. Linear Programming Models: When decision-making pertains to profits, cost,
etc. and these parameters have a linear relationship of several variables, the model
is known as a Linear Programming Model, having constraints or limitations on
various resources and also as a linear function of the decision variables or
parameters. The constraints in linear form can be expressed either as equalities or
inequalities.
ii. When the decision variable values are required to be integers, such as the number
of men, machines, etc. the constraint of such a nature can be inbuilt into the
model, and this technique is called Integer Programming.
iii. The problems, which have multiple, conflicting, and incommensurable objective
functions but under linear constraints, model is termed as Goal Programming

7
 Model. When the decision variables are not definite or deterministic but depend
on chance, the problem becomes a stochastic programming problem.
iv. For working out the cost and time minimization based on deterministic
information available for unit cost or cost per unit distance, the model formulation
is called Transportation Problem. When definite resources are allocated to
perform certain assigned activities, such problems are called Assignment
problems, and the models so used are called Assignment Models.
v. Sequencing Models: Instead of assigning the jobs in a definite activity system,
when we have to determine in what sequence the activities should be performed
out of given resources in the most cost/time effective manner, the models are
called Sequencing Models.
vi. Waiting Line and Queuing Models: These models are used to establish a
tradeoff between the cost of waiting for a customer and that of providing service
following a queue system. In this case, we have to describe various components
of the system such as traffic intensity, the average waiting time of the customer in
the queue, average queue length, etc.
vii. Game Models: These models are formulated and utilized to describe the behavior
of two or more opponents or players who are performing the functions to achieve
certain objectives or goals and in the bargain, would gain or lose in the business
process. Such models are very effectively used for optimizing the strategies of the
players with respect to the anticipated strategies of the competing players.
viii. Dynamic Programming Models: These models are the offshoots of
mathematical programming for optimizing multistage decision processes. The
problems are solved by first dividing the problem into sub-problems or stages and
solving them sequentially till the original problem has been solved.
ix. Inventory Models: These models are primarily meant for working out the
optimal level of stocking and ordering of items for a given situation. The main
objective is to optimize the cost under conflicting requirements of ordering,
holding, and shortages. Quantity discounts and selective inventory controls are
also useful derivations.
x. Replacement Models: These models are utilized when we have to decide the
replacement policy for a piece of equipment for one reason or the other. The
deterioration of efficiency of the equipment with use and time is the reason for
such replacement whether partial or full. The differing performance parameters
create variations in the form of varied replacement policies with the help of
Replacement Models.
xi. Simulation Models: These models are utilized when we want to evaluate the
merit of an alternate course of action by experimenting with a mathematical
model of the problem and the variables in the problem are random. Thus,
repetition of the process by using a simulation model indicates the merit of the
alternative course of action concerning the decision variables.

8
xii. Network Models: These are basically project management models utilized in
planning, monitoring, and controlling various projects, where the utilization of
human and non-human resources has to be optimized with reference to the time
and cost available for the project.
xiii. CPM/PERT as a basic network model helps in the identification of important
bottlenecks or potential trouble areas. The techniques improve the project
coordination by working out effective trade-off analysis for the resource
allocations.
xiv. Decision Analysis Models (Decision Theory): These models are used for the
section of an optimal strategy of operation given the possible payoffs and their
associated probabilities of occurrence. The models are used for the decision-
making process under uncertainty or risk conditions.

Check Your Progress 9.2

Note:

1. OR techniques mainly deal with:

a. Physical models

b. Analogue models

c. Analytical models

d. Iconic models

2. Queuing theory is a prescriptive technique that gives us the ideal configuration of a

service system (True / False)

…………………………………………………………………

3. Inventory models deal with:

a. When to replace equipment

b. Scheduling of activities

c. Most economic quantity of stock to order and hold

d. Allocation of scarce resources

4. One of the properties of the Linear Programming Model is:

a. It will not have constraints

b. It should be easy to solve

9
 c. It must be able to adapt to solve any type of problem

d. The relationship between problem variables and constraints must be linear.

5. Decision theory always guarantees optimal solutions.

a. True

b. False

6. Dynamic programming is a quantitative analysis applicable for large complex

problems that have a sequence of decisions to be made.

a. True

b. False

7. Simulation modelling is generally one of the least expensive approaches to use in

generating solutions to a problem.

a. True

b. False

8. PERT is a tool for planning and control of time

a. True

b. False

9. The CPM is used for completing the projects that involve activities of repetitive
nature

a. True

b. False

10. OR methodology consists of definition, solution, and validation only.

a. True

b. False

9.5 INTERACTIONS BETWEEN MANAGEMENT AND

OPERATIONS RESEARCH
Almost all the fields of management use operations research for decision-making. The
techniques used in Operations Research have wide application in various fields of
business/ industrial/ government/ social sector. A few areas of applications are
mentioned below:

10
Marketing Management
a. Product Selection
b. Competitive Strategies
c. Advertising strategy
Production Management
a. Production scheduling
b. Project scheduling
c. Allocation of resources
d. Location of factories and their sizes
e. Equipment replacement and maintenance
f. Inventory policy
Finance Management
a. Cash flow analysis
b. Capital requirement
c. Credit policies
d. Credit risks
Personal Management
a. Recruitment policies
b. Assignment of jobs
Purchasing and Procurement
a. Rules of purchasing
b. Determining the quality
c. Determining the time of purchases
Distribution
a. Location of warehouses
b. Size of the warehouses
c. Rental outlets
d. Transportation strategies

Check Your Progress 9.3

Note:

11
 1. The interaction between the OR team and Management reaches a peak level in
the implementation phase.
a. True
b. False

Activity 9.1:
Visit any organization of your choice and interview the manager on the use of
different techniques for decision-making.

9.6 PHASES OF OPERATIONS RESEARCH

The procedure to be followed in the study of OR generally involves the following
major phases. I would like you to go through these steps carefully as it will help you to
understand the succeeding units on linear programming problems, transportation,
assignment, decision theory, and game theory.
i. Observe the Problem environment: The first step in the process of OR
development is the problem environment observation. This step includes different
activities; they are conferences, site visits, research, observations, etc. These
activities provide sufficient information to the OR specialists to formulate the
problem.
ii. Formulating the problem: This step is analyzing and defining the problem. In
this step, in addition to the problem definition the objectives, uses and limitations
of OR study of the problem are also defined. The outputs of this step are a clear
grasp of the need for a solution and its nature of understanding.
iii. Constructing a mathematical model: This step develops a model; a model is a
representation of some abstract or real situation. The models are basically
mathematical models, which describe systems, and processes in the form of
equations, formulas/relationships. The different activities in this step are variables
definition, formulating equations, etc.
iv. Derivation of the solution from the model: This step is to get a solution with
the help of the model and input data. This solution is not implemented
immediately, instead, the solution is used to test the model and to find if there are
any limitations.
v. Testing the model and its solutions (updating the model): Suppose the solution
is not reasonable or the behaviour of the model is not proper, the model is updated

12
 and modified at this stage. The output of this stage is the solution(s) that supports
the current organizational objectives.
vi. Implementation: At this step, the solution obtained from the previous step is
implemented. The implementation of the solution involves many behavioral
issues. Therefore, before implementation, the implementation authority has to
resolve the issues. A properly implemented solution results in the quality of work
and gains support from the management.

Check Your Progress 9.4

Note:

1. The first step in solving problems through OR is:

a. Need analysis

b. Model formulation

c. Problem identification

d. Sensitivity analysis

2. One of the OR phases is the Action phase

a. True

b. False

3. OR gives a qualitative solution

a. True

b. False

9.7 LIMITATIONS OF OPERATIONS RESEARCH

No doubt that operations research offers a high level of benefits to the industry but it
does come with its own limitations. Based on the experience and exposure to applying
operation research in the industry, the following limitations are observed:
i. The results may be misleading in case of incorrect assumptions while assigning
quantitative values to the factors involved in the model
ii. Linear Programming Problems (LPP) highly rely on quantitative data and hence
major intangible factors that may affect real situations which are in form of
qualitative data are not considered.

13
 iii. Complex problems may be difficult to be solved using LPP. Simple problems
may economically not justify the use of the LPP technique.
iv. The decision-makers are required to have knowledge of mathematics and
statistics to use LPP.
v. OR only supplements the decision-making and does not replace the entire
process of decision-making. Interpretations based on the findings of the LPP
should be drawn with care.
However, these limitations may be overcome by a proper understanding of the term OR
and taking care while applying the OR techniques for different applications.

Check Your Progress 9.5

Note:
1. The quantitative methods are expensive
a. True
b. False
2. Operations research is a substitute for decision making
a. True
b. False
3. All the quantitative operation research models are based on assumptions
a. True
b. False
4. For adoption of quantitative analysis by the managers ___________ is required.
5. OR employs a team of __________ from ___________.

Activity 9.2:
Suggest a suitable OR model, studied in the unit for the following situations:
1. Deciding the time duration and faculty teaching three subjects per week in a
management institute.
2. Opening another ‘cash counter’ in the retail store.
3. Storing minimum stock level of ‘life-saving drugs’ in hospital.
4. Adapting the exponential smoothing forecast model for fluctuations in
demand for ‘diesel car’.
5. Deciding the recruitment pattern of faculties in a management institute on the
expansion of the number of intake of students and courses offered.

14
9.8 LET US SUM UP
OR is a scientific method of providing an analytical and objective basis for decision-
making for executives. Various definitions of OR bring out the following essentials.
Characteristics of OR
1. System Orientation
2. Interdisciplinary
3. Application of Scientific method
4. Uncovering new problems
5. Quantitative solution
6. Human factors
The present scenario makes the business situations complex, and uncertain and
knowledge abundance increases the use and importance of OR. OR is used in almost all
the fields of business and management functions. The step used while solving OR
model includes.
1. Observing the problem environment
2. Formulating the problem
3. Constructing mathematical model
4. Deriving the solution
5. Testing the model and its solution
6. Final implementation

9.9 KEYWORDS
Operations Research : A method of mathematically based analysis for providing
a quantitative basis for management decisions.

9.10 SUGGESTED FURTHER READINGS /REFERENCES

1. Frederick S Hillier and Gerald J Lieberman. (2009), Introduction to Operations
Research, Ninth Edition, McGraw-Hill Higher Education.
2. Hira, D. S. and Gupta. P.K. (2007), Operations Research, S Chand and
Company, New Delhi.

15
 3. Kalavathy, S. (2002), Operations Research, Fourth Edition, Vikas Publishing
House Private Ltd.
4. Mishra, D.N. Agarwal, S.K. (2009), Operation Research, Lucknow, India,
Global Media.
5. Panneerselvam, R. (2006), Operation Research, Second Edition, PHI Learning
Private Ltd, New Delhi, India.
6. Sharma, A. (2009). Operations Research. Global Media, Himalaya Publishing
House.
7. Sharma, J.K. (2010), Operations Research – Problems and Solutions, Third
Edition, Macmillan Publishers India Ltd.
8. Shenoy, G.V., Srivastava, U.K. and Sharma, S.C. (2005), Operations Research
for Management, Second Edition, New Age International (P) Ltd.
9. Srinivasan, G. (2010), Operations Research – Principles and Applications,
Second Edition, PHI Learning Private Ltd, New Delhi, India.
10. Taha, H. A. (2008), Operations Research – An Introduction, Eight Edition,
Prentice – Hall of India Private Ltd.

9.11 ANSWERS TO CHECK YOUR PROGRESS

Check Your Progress 9.1
1. Scientific
2. Interdisciplinary
3. Decision making
4. True
5. False
6. (d)
7. (d)
8. (b)
9. (d)
10. (d)
11. (d)

Check Your Progress 9.2

1. (c)

16
 2. False
3. (c)
4. (d)
5. (b)
6. (a)
7. (b)
8. (a)
9. (a)
10. (b)

Check Your Progress 9.3

1. (b)

Check Your Progress 9.4

1. (c)
2. True
3. True

Check Your Progress 9.5

1. True
2. False
3. False
4. Prerequisite Knowledge
5. Scientists, different disciplines

Activity 9.1:
Points to be covered
 Decision problems a manager faces in his organization
 How does he solve them? What techniques does he use?

17
  Is he aware of operations research? If so which techniques he uses?
 What challenges does he face while using these techniques?

Activity 9.2:
1. Allocation
2. Queuing
3. Inventory
4. Inventory Control
5. Replacement

9.12 UNIT END QUESTIONS

1. What is Operations Research?
2. Give any three useful definitions of operations research and explain them.
3. Give reasons why most definitions of operations research are not satisfactory.
4. Discuss the origin and development of operations research with a suitable
classification.
5. Discuss the scope of OR.
6. List the different OR techniques.
7. Describe management problem situations where OR techniques can be used.
8. Describe the different phases/steps involved in OR.
9. OR is a decision-making tool, justify using an example.
10. Discuss the limitations of operations research.

18
UNIT 10 DECISION THEORY
Structure

10.0 Objectives
10.1 Introduction
10.2 Decision Making under Uncertainty
10.2.1 Maximax Criterion
10.2.2 Maximin Criterion
10.2.3 The Minimax Regrets Criterion
10.2.4 Laplace Criterion
10.3 Decision Making Under Risk
10.4 Decision Tree Analysis
10.5 Let Us Sum Up
10.6 Keywords
10.7 Suggested Further Readings / References
10.8 Answer to Check Your Progress
10.9 Unit End Questions
10.10 Assignment Questions
10.11 Solution to Assignment Questions

10.0 OBJECTIVES
After studying this unit, you should be able to,
 discuss the meaning and types of decision theory;
 explain the decision-making under uncertainty;
 describe the decision-making under risk situations; and
 explain the decision tree analysis.

10.1 INTRODUCTION
As an individual, we make many decisions every day. Sometimes these decisions are
highly important and may have a long-term impact on our future. Decisions about the
selection of a vehicle, purchase of a plot, renting a farm, investment in shares/stocks,
etc, are all important decisions and one would like to make a correct choice out of the

1
available alternatives. Decision theory is defined as a body of several methods which
facilitate the decision-maker to select wisely the best course of action from amongst
several alternatives.
In this unit, we will study the various decision rules that are available to decision-
makers under two different decision-making environments namely,
1. Decision-making under uncertainty, and
2. Decision-making under risk.
The decision under uncertainty are those problems in which each course of action can
result in alternative outcomes, the probabilities of which cannot be assigned as a
decision-maker does not know which outcomes can or will occur. For example, a
farmer in a village may know the possible outcomes for the monsoon this year as heavy
rainfall, moderate rainfall, and low rainfall but does not know the probability of its
occurrence, as he may not have past information with him that he can use for
prediction. For decision-making under risks, the decision-maker would be having the
additional information in terms of probabilities or the chances of having heavy,
moderate, and low rainfall based on past knowledge and records.

10.2 DECISION MAKING UNDER UNCERTAINTY

Under conditions of uncertainty, a farmer would know that the yield of the crop may be
high, moderate, or low but would not have enough information to be able to use
probabilities for the possible states that will occur. Further, these yields are dependent
on the type of crops used by the farmer. Presently he has to decide amongst the three
crops (A, B, and C), he knows that the possible yield (High, Moderate and Low) will
yield the following payoff (profits in 1000’s Rs) as shown in Table 10.1 below,
Table 10.1: Payoff matrix (profits in 1000’s Rs)
Alternate States of Nature
Strategies High Moderate Low
A Rs. 60 Rs. 42 Rs. – 10
Crops

B Rs. 85 Rs. 60 Rs. – 20

C Rs. 50 Rs. 25 Rs. – 12.5
The above table is called the payout table, we will try to help the farmer in decision
making by demonstrating the use of the following decision rules:
1. The Maximax Criterion
2. The Maximin Criterion
3. The Minimax regret Criterion
4. Laplace’s Criterion

2
10.2.1 Maximax Criterion
The maximax criterion is based on an optimistic rule assuming that the best situation
will occur in the future. Considering the payoff in Table 10.2 described earlier we
demonstrate how to use the maximax criterion.
Table 10.2. Payoff matrix
Rs. 60 Rs. 42 Rs. – 10
Rs. 85 Rs. 60 Rs. – 20
Rs. 50 Rs. 25 Rs. – 12.5
The following steps are to be followed,
Step 1: Select the maximum payoff for each alternative/strategy. We observe the result
in following Table 10.3
Table 10.3: Payoff table
Crops Payoff
A Rs. 60
B Rs. 85
C Rs. 50
Step 2:Select the maximum of these maximum payoffs: we observe crop B with Rs. 85
as the maximum payoff. So we have selected the strategy with a maximum of
maximums.

10.2.2 Maximin Criterion

The Maximin criterion is based on a pessimistic rule assuming that the worst situation
is likely to occur in the future and we could like to maximize the profits. Using the
payoff matrix given in Table 10.2 referred to earlier we explain the rule as follows:
Table 10.2: Payoff matrix
Rs. 60 Rs. 42 Rs. – 10
Rs. 85 Rs. 60 Rs. – 20
Rs. 50 Rs. 25 Rs. – 12.5
Step 1: Select the minimum payoff for each strategy from following Table 10.4.
Table 10.4: Payoff matrix
Crops Payoff
A Rs. – 10
B Rs. – 20
C Rs. – 12.5

3
 Step 2: Select the maximum of the minimum payoff; we observe that crop A leads to
the maximum payoff and hence using crop A as the strategy is recommended.
Note that in case, we have a payoff matrix for costs and not for the profit we will be
using the minimax criterion wherein we will be minimizing the maximum costs likely
to be incurred. The steps to be followed in such a situation will be as follows.
Step 1: Select the maximum cost from each strategy.
Step 2: Select the minimum of the maximum costs selected in step 1.

10.2.3 The Minimax Regrets Criterion

This technique calculates the opportunity loss or regrets for not taking the best decision
under each state of nature and the use of the minimax regret criterion is discussed
below for our payoff matrix shown in Table 10.2;
Table 10.2: Payoff matrix
Rs. 60 Rs. 42 Rs. – 10
Rs. 85 Rs. 60 Rs. – 20
Rs. 50 Rs. 25 Rs. – 12.5
Step 1:Develop a regret matrix from the given payoff matrix, on the following
principle:
i. When payoff represents profit
ith regret = (maximum payoff - ith pay off) for the jth event

ii. When payoff represents costs

ith regret = (maximum payoff - ith pay off) for the jth event

The payoff matrix for the problem we have selected for illustration represents a profit
payoff matrix. Therefore the regret Table 10.5 is prepared as follows;
Table 10.5: Profit payoff matrix
Pay off amounts Regret Pay off
Maximum
Alternative amounts
Regret
High Medium Low High Medium Low
A Rs. 60 Rs. 42 Rs. - 10 25 18 0 25
B Rs. 85 Rs. 60 Rs. - 20 0 0 0 10
C Rs. 50 Rs. 25 Rs. – 12.5 35 35 2.5 35
From the regret payoff matrix, the minimum regret is observed for crops B (Regret =
10) and therefore the selection of Crop B will be recommended.

4
10.2.4 Laplace Criterion
Also known as the equally likely decision criterion, this rule assumes that the states of
nature (Low, Moderate, and High yield) will occur with equal probabilities. The
expected value for each alternative is then calculated. The calculations for farmers'
payoff matrix are shown below;
1 1 1
Crop A = 3 (60) + (42) + (−10) = 30.66
3 3

1 1 1
Crop B = 3 (85) + (60) + (−20) = 41.66
3 3

1 1 1
Crop C = 3 (50) + (25) + (−12.5) = 20.83
3 3

It is observed that crop B has the highest expected value and hence will be selected by
the farmer.
Once we have understood the different rules available for decision-makers under
uncertain environments, the next logical question is which one is the best technique to
be used in practice, as it is observed that the different techniques may produce different
results (selection of alternatives). I would like to mention here that the use of a
particular technique depends on the decision maker’s style, risk-taking ability, past
behaviour, whether he knows the rule adopted by his competitor, etc. Therefore it is
expected that different decision-makers may deploy different rules to take the decision
in a given situation.

Check Your Progress 10.1

Note:

1. The decision taken under uncertain conditions results in alternative outcomes

a. True

b. False

2. The decision taken under risk needs additional information.

a. True

b. False

3. The maximax criterion is based on ___________ rule assuming that the best
situation will occur in the future.

4. The Maximin criterion is based on ______________ rule assuming that the

worst situation is likely to occur in the future.

5
5. The Minimax Regrets Criterion technique calculates loss and profit.

a. True

b. False

6. In the Laplace Criterion technique, the outcome occurs with equal probabilities.

a. True

b. False

7. Which is the best technique to be used in practice?

a. Maximax

b. Maximin

c. Minimax

d. Laplace Criterion

e. All the above

8. The maximin criteria is a (n) __________ criteria.

a. Optimistic

b. Neutral

c. Pessimistic

d. None of the above

9. Table 10.6 indicates 12 possible payoffs for the manufacturing company’s

expansion decision.

Table 10.6: Payoff matrix

States of nature (product demand)

Alternatives
High Moderate Low Nil
Expand Rs. 50,000 Rs. 25,000 Rs.  25,000 Rs.  25,000
Contract Rs. 70,000 Rs. 30,000 Rs.  40,000 Rs.  80,000
Subcontract Rs. 30,000 Rs. 15,000 Rs.  1,000 Rs.  10,000
Select optimum decision criterion using Maximax criterion

10. A company has to choose one of the three types of biscuits: Creams, Coconut,
and Glucose, Sales expected during next year are highly uncertain. The

6
 marketing department estimates the profits considering manufacturing cost,
promotional efforts, distribution setup, etc. as given in Table 10.7 below.

Table 10.7: Payoff matrix

Profits on the estimated level of sales (in lakhs) for

Type of Biscuits quantities
5,000 10,000 20,000
Cream (C) 15 25 45
Coconut (Co) 20 55 65
Glucose (G) 25 40 70
Select optimum decision criterion using Maximin criterion

10.3 DECISION MAKING UNDER RISK

As mentioned in the introductory section of this unit, this risk environment assumes
that the decision-maker has sufficient knowledge about the different states of nature
based on which he can assign the probability of occurrences for them. In our illustration
example, the farmer is of the view that a high yield is likely to occur with a probability
of 0.65, a moderate yield with a probability of 0.20, and a low yield with a probability
of 0.15. We can then compute the expected value for each alternative as shown in Table
10.8.
Table 10.8: Payoff matrix
States of Nature
Alternate High Moderate Low Expected Value
Strategies
yield yield yield
Probabilities 0.65 0.20 0.15
Crop A Rs. 60 Rs. 42 Rs. – 10 45.90
Crop B Rs. 85 Rs. 60 Rs. – 20 64.25
Crop C Rs. 50 Rs. 25 Rs. – 12.5 35.62
Expected value for crop A = 0.65  Rs. 60  0.20  Rs. 42  0.15  Rs. (10) = 45.90
Expected value for crop B = 0.65  Rs. 85 + 0.20  Rs. 60 + 0.15  Rs. (20) = 64.25
Expected value for crop C = 0.65  Rs. 50 + 0.20  Rs. 25 + 0.15  Rs. (12.5) = 35.62
The farmer should now opt for crop B to gain a higher expected value.
Now let us consider a typical situation where decisions are taken daily. Naresh is a
newspaper vendor who sells newspapers in a small complex. He buys the newspaper
from a local dealer at Rs. 1 per unit and sells it for Rs. 1.2 per paper. Any unsold paper
at the end of the day reduces its value to Rs. 0.30 and is sold to a scrap dealer. Naresh’s
demand fluctuated between 60 to 65 customers a day. Further, he loses customers if he

7
does not have sufficient stock. Naresh has been maintaining data on his customers. His
data reveals that the newspaper demand in the past 300 days was as given in Table
10.9.
Table 10.9: Demand Table for 300 days.
Demand in units 60 61 62 63 64 65
No. of days 45 60 90 60 30 15
Probability of demand 0.15 0.20 0.30 0.20 0.10 0.05
(i.e. no. of days / 300)
The probability that the demand may be 60, 61, 62, and so on can be computed by
considering the relative frequency. The following payoff Table 10.10 is constructed.
Table 10.10: Payoff matrix
Stocking Policy 60 61 62 63 64 65
Probability of demand 0.15 0.20 0.30 0.20 0.10 0.05
60 12 12 12 12 12 12
61 11.1 12.2 12.2 12.2 12.2 12.2
62 10.2 11.3 12.4 12.4 12.4 12.4
63 9.3 10.4 11.5 12.6 12.6 12.6
64 8.4 9.5 10.6 11.7 12.8 12.8
65 7.5 8.6 9.7 10.8 11.9 13.0
In the above payoff table it is observed that, as the demand varies between 60 to 65,
Naresh's profit also varies depending on how many units he has stocked for that day. It
can be seen that If Naresh stocks 63 units and the demand for that day turns out to be
63 units, the total profit earned by him will be;
63 units  0.20 (profit margin) = 12.6
However, if the demand that day is below his stocking level of 63 units and only 60
units are sold then his profit will be;
(profit from the sold unit)  loss from unsold units (i.e. Rs. 0.90 per unit)
= (60 units  0.2 profit/unit)  (3 units  0.90 / unit)
= Rs. 12  Rs. 2.7
= Rs. 9.3
If Naresh stocks 64 units and the demand for that day is 65 he will be able to make a
maximum profit of Rs.12.8 as all the units are sold. Table 10.10 is also referred to as
the conditional table.
The conditional table (refer to Table 10.10) for this problem tells how much profits
Naresh will be able to make under different stocking and demand conditions, but it

8
does not tell how much Naresh is required to stock. Proceeding further we would like
to make use of the assigned probabilities for the occurrence of demand and calculate
the expected profits for different stocking situations as shown in Table 10.11
Table 10.11: Expected Profit
Demand
Stocking Policy 60 61 62 63 64 65 Expected
Probability 0.15 0.20 0.30 0.20 0.10 0.05 Value
60 12 12 12 12 12 12 12
61 11.1 12.2 12.2 12.2 12.2 12.2 12.035
62 10.2 11.3 12.4 12.4 12.4 12.4 11.3
63 9.3 10.4 11.5 12.6 12.6 12.6 11.335
64 8.4 9.5 10.6 11.7 12.8 12.8 10.6
65 7.5 8.6 9.7 10.8 11.9 13.0 9.755
Expected values for;
60 units stocking policy = 12  0.15 + 12  0.2 + 12  0.3 + 12  0.2 + 12  0.1 + 12 
0.05
= 12
61 unit stocking policy = 11.1  0.15 + 12.2  0.2 + 12.2  0.3 + 12.2  0.2 + 12.2 
0.1 + 12.2  0.05
= 12.035 and so on
62 unit stocking policy = 11.3
63 unit stocking policy = 11.335
64 unit stocking policy = 10.6
65 unit stocking policy = 9.7555
From the above situation expected values, Naresh should stock 61 newspapers per day
as this stock level gives him the maximum expected value. The next concept we will
discuss now is the Expected Value of Perfect Information (EVPI).
EVPI as the name refers is a situation wherein the decision-maker spends money to get
additional information so that he gets perfect information for what is going to be the
demand on that day so that he can stock only the required quantity and maximize his
profits. If Naresh can remove all uncertainty from his problem by obtaining more
information on the probability of occurrences of demand he will not be required to
incur a loss of Rs. 0.9 on every unsold newspaper. With perfect information available
to Naresh the demand will still vary between 60 to 65 units per day. Table 10.12 shows
the conditional profit table for Naresh if he has perfect information. It shows the
maximum profit that he can expect.

9
Table 10.12: Conditional profit table
Demand
Stocking Policy
60 61 62 63 64 65 Expected
Probability
0.15 0.20 0.30 0.20 0.10 0.05 Value
60 12 1.8
61 12.2 2.44
62 12.4 3.72
63 12.6 2.52
64 12.8 1.28
65 13.0 0.65
Total 12.41
The maximum profit that Naresh can expect with perfect information is Rs.12.41. In
absence of this information, we expected Naresh’s profit to be Rs. 12.035. The value of
perfect information is Rs. 12.41 – Rs. 12.035 = Rs. 0.375. This means that Naresh
should therefore not spend more than Rs. 0.375 every day on gathering perfect
information else he will not attain the maximum profit of Rs. 12.41.

Check Your Progress 10.2

Note:

1. In Decision making under risk, decision-maker must have __________

knowledge.

2. When decisions are at risk, the outcome received is calculated as the expected
monetary value.

a. True

b. False

3. Decision-making under risk is a situation where more than one state of nature
exists and a decision-maker has sufficient information to assign probability
values to the likelihood of occurrence of each of these states.

a. True

b. False

4. EVPI Stands for

a. Expected Value of Perfect Information (EVPI).

b. Expected Value of Pure Information (EVPI).

10
 c. Expected Value of Present Information (EVPI).

d. Expected Value of Projected Information (EVPI).

5. The probability of demand for cool cab services for hiring on any day in a given
district is given in following Table 10.13.

No. of cab demanded 0 1 2 3 4

Probability 0.1 0.2 0.3 0.2 0.2
Cab services have a fixed cost of Rs. 90 each day to keep the daily hire charges
(net of variable costs of running) Rs. 200. If the cab services company owns 4
cabs, what is its daily expectation? If the company is about to go into business
and currently has no cab how many cabs should it buy?

10.4 DECISION TREE ANALYSIS

The above decision environments deal with decision-making reference to the
availability of probabilities or not. Further, both of the environments do not take the
time sequence into account in which various actions and consequent events would
occur. Decision tree analysis is a technique of decision-making that considers the time
sequence.
A tree formation listing of all the possible events and resultant outcomes is presented in
a decision tree analysis as we know that decision-making involves several stages and at
each stage, each of the alternatives will result in different outcomes and payoff.
The various components of a decision tree are shown in Figure 10.1

Chance
node
Chance Chance
node node
Decision Chance
Points node Chance
node Outcome of
Chance Pay-off
decision branches

node
Courses open

Chance
Strategies

Branches
Chance

node

Fig. 10.1: Decision Tree

Decisions Points: The Square indicates a decision point. The branches leading from
this square represent various courses of action available to the decision-maker and are
called as decision branches.

11
Node: The node is denoted by a circle. At the end of each branch, there is a node.
Various outcomes emerge out of the node with their associated probabilities estimates.
Branches also emerge out of the nodes and are called as chance branches. These chance
branches are indicative of the likelihood that particular action would be assumed within
the given solution. The payoff written in the above figure against each chance branch
can be positive or negative depending on the nature of the event. Profits or revenues are
represented as positive payoffs whereas expenditure and losses as negative payoffs.
Now let us solve a decision problem using a decision tree with the following illustrative
example
Illustrative Example 10.1
1. A farm equipment manufacturing company wants to select between two
alternatives as follows:
Alternative I: Distribution through dealers
Alternative II: Direct selling of equipment (owned showroom).
There is a 0.7 probability of having high sales when distributed through a dealer
with a payoff of Rs. 70 lakhs. The possible payoff when the sales are low is Rs.
30 lakhs. If the company decides to go for its own distribution (direct-selling)
the probability of having high sales is 0.60 with a payoff of Rs. 90 lakhs and Rs.
25 lakhs when sales are low.
Now let us draw the decision tree to understand the problem. We know that there is a
choice of two alternatives and hence this can be shown as a decision point in following
Figure 10.2.

Fig. 10.2: Decision Tree

Next, we know that there is the possibility of two chances for each alternative, high
sales or low sales with some expected payoff. The same is represented as follows in
Figure 10.3;

12
 Rs. 70 lakhs
High Sales (0.7)

Dealership Low Sales (0.3)

Rs. 30 lakhs
D
High Sales (0.6) Rs. 90 lakhs
Direct
Selling
Low Sales (0.4) Rs. 25 lakhs
Figure 10.3: Decision Tree
In the above problem, the probability for high sales is given as 0.7 for the dealership,
therefore the probability of low sales will be 0.3 (1 – 0.7 = 0.3), similarly for direct
selling the probability for low sales is calculated as 0.4.
Now we calculate the expected payoffs to decide on the profitable strategy.
Expected payoff for dealership = (0.7  70) + (0.3  30) = Rs. 58 lakhs
Expected payoff for direct selling = (0.6  90) + (0.4  25) = Rs. 64 lakhs
Based on the above two payoffs it is observed that the direct selling strategy is more
attractive with higher expected returns.
So this is how a decision tree can be put into practice. Decision trees can be used to
solve complex problems as it offers to present the problem in a graphical representation
and can be approached in stages.

Check Your Progress 10.3

1. Decision tree analysis is a technique of decision-making that considers the

series sequence.

a. True

b. False

2. Decision tree analysis uses which types of formation to represent possible

events and resultant outcomes.

a. Star

b. Tree

c. Ring

13
 d. Mesh

3. A decision point is indicated by

a. Circle

b. Point

c. Node

d. Square

4. In the decision tree at the end of each branch there is a ___________

a. Circle

b. Point

c. Node

d. Square

5. In the Decision tree branches that emerge out of the nodes are called as
_________ branches.

6. A firm is planning to develop and market a new soft drink. The cost of
extensive research to develop the soft drink has been estimated at Rs. 1,00,000.
The manager of the research programme has found that there is a 60% chance
that the soft drink will be developed successfully. The market potential has been
assessed in the following Table 10.14.

Table 10.14: Market Potential

Market Condition Probability Present value of

profit (Rs.)
Large Market Potential 0.1 50,000
Moderate Market potential 0.6 25,000
Low Market potential 0.3 10,000
The present values figures do not include the cost of research. While the firm is
considering this proposal, a second proposal almost similar comes up for
consideration. The second one also requires an investment of Rs. 1,00,000 but
the present value of all profits is 12,000. The return on investment in the second
proposal is certain.

1. Draw a decision tree indicating all events and choices of the firm

2. What decision the firm should take regarding the investment of Rs.
1,00,000?

14
10.5 LET US SUM UP
Decision theory has become an important technique for decision-making in the face of
uncertainty. This unit in detail discussed the decision theory procedures designed to
solve problems with a limited number of decision alternatives and a finite list of
possible states of nature. The goal of the decision theory approach was to identify the
best decision alternatives given an uncertain or risk-filled pattern of future events (i.e.,
states of nature).
In Decision making under uncertainty, we discussed the application of various decision
techniques like maximin, maximax, minimax regret, and the Laplace criterion. When
the problems have probabilities known to the decision-maker based on the past data we
apply the technique of decision-making under risk and identify the best decision and
calculate the expected value of perfect information. The EVPI is the amount the
decision maker can pay to get perfect information regarding the occurrence of an event.
Decision trees are used to analyse sequential decision-making situations. The decision
alternatives and possible events are structured in the form of a tree diagram. The tree is
analysed in a right-to-left manner. Expected value computations are performed at each
event node and the optimal decision alternatives are identified at each decision point.

10.6 KEYWORDS
Decision Theory : It is a method or framework of logical and
mathematical concepts, aimed at helping decision-
makers to formulate rules and choose among a set of
alternatives.
Decision Tree Analysis : A schematic method of alternatives available to the
decision-maker, to analysis the circumstances along
with their possible consequences.”
Laplace Criterion : This method explicitly uses the probability of
assessments regarding the likelihood of occurrence of
the states of nature.
Maximax Criterion : This method looks at the best that could happen
under each action and then chooses the action with
the largest value.
Maximin Criterion : This method involves selecting the alternative that
maximises the minimum pay-off achievable.
Minimax Regrets : This method minimizes regret which is highest when
Criterion one decision has been made instead of another.

15
13.7 SUGGESTED FURTHER READINGS / REFERENCES
1. Frederick S Hillier and Gerald J Lieberman. (2009), Introduction to Operations
Research, Ninth Edition, McGraw-Hill Higher Education.
2. Hira, D. S. and Gupta. P.K. (2007), Operation Research, S Chand and
Company, New Delhi.
3. Kalavathy, S. (2002), Operation Research, Fourth Edition, Vikas Publishing
House Private Ltd.
4. Panneerselvam, R. (2006), Operation Research, Second Edition, PHI Learning
Private Ltd, New Delhi, India.
5. Sharma, A. (2009). Operations Research. Global Media, Himalaya Publishing
House.
6. Sharma, J.K. (2010), Operations Research – Problems and Solutions, Third
Edition, Macmillan Publishers India Ltd.
7. Shenoy, G.V., Srivastava, U.K. and Sharma, S.C. (2005), Operations Research
for Management, Second Edition, New Age International (P) Ltd.
8. Srinivasan, G. (2010), Operations Research – Principles and Applications,
Second Edition, PHI Learning Private Ltd, New Delhi, India.
9. Taha, H. A. (2008), Operations Research – An Introduction, Eight Edition,
Prentice – Hall of India Private Ltd.

10.8 ANSWER TO CHECK YOUR PROGRESS

Check Your Progress 10.1
1. a
2. a
3. Optimistic
4. Pessimistic
5. False
6. a
7. e
8. c
9. The maximax payoff is Rs. 70,000 corresponding to the alternatives ‘construct’.
States of nature (product demand) Maximum of
Alternatives row
High Moderate Low Nil Rs.

16
 Expand Rs. Rs. 25,000 Rs.  25,000 Rs.  25,000 50,000
50,000
Contract Rs. Rs. 30,000 Rs.  40,000 Rs.  80,000 70,000 Maximax
70,000
Subcontract Rs. Rs. 15,000 Rs.  1,000 Rs.  10,000 30,000
30,000
10.
Profits on estimated level of sales (in lakhs) for
Type of Biscuits quantities
5,000 10,000 20,000
Cream (C) 15 25 45
Coconut (Co) 20 55 65
Glucose (G) 25 40 70
Select the minimum payoff for each strategy
Profits on the estimated level of sales (in lakhs)
Types of Biscuits for quantities
5,000
Cream (C) 15
Coconut (Co) 20
Glucose (G) 25
In using the maximin criterion, the decision-maker adopts a pessimistic approach and
tries to maximise his security in the face of a highly uncertain situation. For the worst
situation, the payoffs are 15, 20, and 25 for C, Co and G respectively. Even at the
pessimistic level, the manager tries to make the best of the situation reaching the
decision as Glucose, pay-off being the best amongst the worst. This maximises the
minimum pay-off. Hence, the company will launch Glucose biscuits (G).

Check Your Progress 10.2

1. Sufficient
2. False
3. True
4. a
5. Fixed Cost = Rs. 90
Variable cost = Rs. 200

17
 Now, we have
No. of cab demanded: 0 1 2 3 4
Pay-offs: 360 160 40 240 440
and
Daily expectation = (360  0.1) + (160  0.2) + (40  0.3) + (240  0.2) +
(440  0.2) = 80

Demand of Probability Conditional pay-off

Cabs 0 1 2 3 4
0 0.1 0 90 180 270 360
1 0.2 0 110 20 70 160
2 0.3 0 110 220 130 40
3 0.2 0 110 220 330 240
4 0.2 0 110 220 330 440

Demand of Probability Conditional pay-off

Cabs 0 1 2 3 4
0 0.1 0 9 18 27 36
1 0.2 0 22 4 14 32
2 0.3 0 33 66 39 12
3 0.2 0 22 44 66 48
4 0.2 0 22 44 66 88
Expected Value 0 90 140 130 80
As the expected value of the course of action is 2 is 140, which is higher than the
rest of all, the company should purchase two cabs.

Check Your Progress 10.3

1. False
2. Tree
3. Square
4. Node
5. Chance

18
 6. The following figure is the decision tree
PV of Profit
Large Market Rs. 25,000
Develop and Developed successfully Medium Market Rs. 25,000
D1
Market New Soft
Drink D2 Low Market

Accepted Proposal II Not Successfully Rs. 10,000

Developed
Stop Stop
Do not enter Market
At Point D2
Decision are (1) Enter the market (2) Do not enter the market
(a) Enter market
EMV = Expected P V
= (50,000  0.1)  (25,000  0.6) + (10,000  0.3)
= 5000 + 15000 + 3000
= 23,000
(b) Do not enter the market
EMV = Expected P V = 0  1 = 0
Decision: Enter the market since EMV is more.

At D1, Decisions are (1) Develop a new soft drink (2) Accept proposal II
(a) Develop new soft drink
EMV = Expected P V = (23,000  0.6) + (0  0.4) = (13,800 + 0) = 13,800
(b) Accept proposal II
EMV = Expected P V = 12,000  1 = 12,000

Using the EMV criterion, the optimal decision at D1 is to develop and market
the new soft drink.

10.9 UNIT END QUESTIONS

1. What do you understand by the term decision theory?
2. Describe some methods which are useful for decision-making under uncertainty.
3. What techniques are used to solve decision-making problems under uncertainty?
4. Which decision theory techniques result in an optimistic decision? Which
technique results in a pessimistic decision?

19
 5. What do you understand by decision-making under risk?
6. Write a note on the value of perfect information.
7. What are the main steps associated with decision tree analysis.

10.10 ASSIGNMENT QUESTIONS

1. Mr. Naresh wants to invest Rs. 10,000 in one of the three options A, B, and C.
The payoff for his investment depends on the nature of the benefits he gets from
the investment (share market, gold market, fixed deposit). The possible returns
under each situation are given below;
Strategy Nature of Benefits from the investment
Share Market (S1) Gold Market (S2) Fixed Deposit (S3)
A 2000 1200 1500
B 3000 800 1000
C 2500 1000 1800
What course of action has he to take according to,
i. Maximin
ii. Maximax
iii. Laplace
iv. Minimax (the regret criterion)
2. At a cake shop, special cookies are sold, it has the following probability of
selling cookies.
No. of cookies sold Probability
10 0.10
11 0.15
12 0.20
13 0.25
14 0.30
The cost of a cookie is 30 paise and the sale price is 50 paise. At end of the day,
unsold cookies have to be thrown. How many cookies should the shop owner
order to avoid wastage?
3. A glass factory specializing in crystal is developing a substantial backlog and
the firm’s management is considering three courses of action: Arrange for sub-
contracting (S1) begin overtime production (S2), and construct new facilities
(S3). The correct choice depends largely upon future demand which may be

20
 low, medium, or high. By consensus, management ranks the respective
probabilities as 0.10, 0.50, and 0.40. Cost analysis reveals the effect on the
profits that are shown in the table below;

Demand Probability Course of Action

S1 S2 S3
(Subcontracting) (Begin (Construct
Overtime) Facilities)
Low (L) 0.10 10  20  150
Medium 0.50 50 60 20
(M)
High (H) 0.40 50 100 200
Show this decision situation in the form of a decision tree and indicate the most
preferred decision and corresponding expected value.

10.11 SOLUTION TO ASSIGNMENT QUESTIONS

1. Solution:
i. Maximin criterion
Strategy Nature of Benefits from the investment
(S1) (S2) (S3) Min
A 2000 1200 1500 1200 Maximin
B 3000 800 1000 800
C 2500 1000 1800 1000
Maximum among the minimum pay-offs is 1200. Hence the corresponding
option A is to be chosen
ii. Maximax criterion
Strategy Nature of Benefits from the investment
(S1) (S2) (S3) Man
A 2000 1200 1500 2000
B 3000 800 1000 3000 Maximax
C 2500 1000 1800 2500
Maximum among the maximum pay-offs is 3000. Therefore option B is to be
chosen.

21
 iii. Laplace criterion
Strategy Nature of Benefits from the
investment
(S1) (S2) (S3)
A 2000 1200 1500 1 1 1
( × 2000) + ( × 1200) + ( × 1500) = 1566
3 3 3
B 3000 800 1000 1 1 1
( × 3000) + ( × 800) + ( × 1000) = 1600
3 3 3
C 2500 1000 1800 1 1 1
( × 2500) + ( × 1000) + ( × 1800) = 1766
3 3 3
Alternative C has the maximum expected payoff. Hence alternative C is to be
chosen.
iv. Regret criterion
Strategy Nature of Benefits from Regret pay-off Maximum
the investment regret
(S1) (S2) (S3) (S1) (S2) (S3)
A 2000 1200 1500 1000 0 300 1000
B 3000 800 1000 0 400 800 800
C 2500 1000 1800 500 200 0 500 Minimax
Max 3000 1200 1800
Among the maximum regrets, we find that 500 is the minimum. Therefore
alternative C is to be chosen.
2. We can apply either the EMV criterion or the EOL criterion. Let us apply the EMV
criterion for which we have to calculate the payoff. The numbers of cookies ordered are
the different courses of action. The cookies ordered maybe 10, 11, 12, 13, 14. Denote
them by A1, A2, A3, A4, A5.
Similarly, the number of cookies demanded maybe 10, 11, 12, 13 or 14. These
demands may be D1, D2, D3, D4, D5. These are events. The pay-off values are
calculated below;
The selling price of each item = 50 Ps. and cost of a cookie = 30 Ps.
A1D1 = (10  50) – (10  30) = 200
A2D1 = (10  50) – (11  30) = 170
A3D1 = (10  50) – (12  30) = 140
A4D1 = (10  50) – (13  30) = 110 etc.

22
 A1 A2 A3 A4 A5
D1 200 170 140 110 80
D2 200 220 190 160 130
D3 200 220 240 210 180
D4 200 220 240 260 230
D5 200 220 240 260 280
Given probabilities are 0.10, 0.15, 0.20, 0.25, 0.30
Calculation of EMV for all the Acts
A1 A2 A3 A4 A5
Pay off  Prob Pay off  Prob Pay off  Prob Pay off  Prob Pay off  Prob
0.10  200 = 20 0.10  170 = 17 0.10  140 = 20 0.10  110 = 20 0.10  80 = 8
0.15  200 = 30 0.15  220 = 33 0.15  190 = 208.5 0.15  160 = 30 0.15  130 = 19.5
0.20  200 = 40 0.20  220 = 44 0.20  240 = 48 0.20  210 = 40 0.20  180 = 36
0.25  200 = 50 0.25  220 = 55 0.25  240 = 60 0.25  260 = 50 0.25  230 = 55
0.30  200 = 60 0.30  220 = 66 0.30  240 = 72 0.30  260 = 50 0.30  280 = 84
= 200 = 215 = 222.5 = 220 = 205
 EMV for Acts A1, A2, A3, A4, and A5 are respectively 200, 215, 222.5, 220,
and 205
EMV for A3 is greater and therefore A3 is an optimal act.
 No. of cookies to be ordered = 12.
3. A decision tree that represents possible courses of action and states of nature is
shown in the following figure. In order to analyze the tree, we start working backward
from the end branches.

23
The most preferred decision at decision node 0 is found by calculating the expected
value of each decision branch and selecting the path (course of action) with a high
value. 4
0.10  10 = 1 (in Rs. 1000)
L, (p = 0.10)

M, (p = 0.50) 5 0.50  50 = 25
1

EMV = 46 H, (p = 0.40)
S1: Subcontracting 6 0.40  50 = 20
46
L, (p = 0.10) 7 0.10   20 = 2
S2: Begin overtime
0 2
M, (p = 0.50)
EMV = 68 8 0.50  60 = 30
H, (p = 0.40)
S3: Construct facilities
9 0.40  100 = 40
68

3
L, (p = 0.10) 10 0.10  150 = 15

EMV = 75 M, (p = 0.50)

H, (p = 0.40) 11 0.50  20 = 10

Figure: Decision Tree 12 0.40  200 = 80

75
Since node 3 has the highest EMV, therefore the decision at node o will be to choose
the course of action S3, i.e. construct new facilities.

24
UNIT 11 TRANSPORTATION MODEL AND
ASSIGNMENT PROBLEMS
Structure

11.0 Objectives
11.1 Introduction
11.2 Assumptions in the Transportation Model
11.3 Formulation and Solution of Transportation Models
11.4 Solution to Transportation Problem
11.4.1 North-West Corner Rule or North-West Corner Method (NWCM)
11.4.2 Least Cost Method
11.4.3 Vogel Approximation Method (VAM)
11.4.4 Row Minima Method
11.4.5 Column Minima Method
11.5 Case of Unbalanced Problem
11.6 Transshipment Problem
11.7 Assignment Problem
11.7.1 Formulation of Assignment Problem
11.7.2 Solution to Assignment Problem (Hungarian Method)
11.7.3 Unbalanced Assignment Problem
11.8 Let Us Sum Up
11.9 Keywords
11.10 Suggested Further Readings / References
11.11 Answer to Check Your Progress
11.12 Unit End Questions

11.0 OBJECTIVES
After studying this unit, you should be able to,
 discuss the meaning, assumption, and formulation of Transportation Models;
 explain the initial basic solution to Transportation Problem;
 discuss the case of Unbalanced Problem and Transshipment Problem;

1
 explain the meaning and formulation of the Assignment Problem; and
 discuss the unbalanced Assignment Problem.

11.1 INTRODUCTION
The Transportation model is an application of a Linear programming problem that is
used to make transportation decisions. We know that transportation is an activity that
deals with the movement of goods/products from one place to another. In a business
scenario, the products are to be moved from the source of origin (factory,
manufacturing plant, etc.) to a destination source (market, warehouse, etc.). This
movement incurs transportation costs and hence in a transportation model our
objectives would be to minimize those costs. Apart from the transportation decisions,
the Transportation Problem Model can be also used in taking scheduling, investment,
plant location, and inventory management decisions.
Before we discuss the structure of a Transportation Problem let us briefly consider the
various assumption used for Transportation problems.

11.2 ASSUMPTIONS IN THE TRANSPORTATION MODEL

The following are the assumptions made in Transportation Problem
1. The total supply from all the sources is equal to the total demand from all the
destinations.
2. There are no restrictions on the items being transported from all sources to
destinations unless stated in the problem.
3. The unit transportation cost of the item from all sources to destinations is
certain and known beforehand.
4. The transportation cost on a given route is directly proportional to the number
of units shipped on that route.
5. The objective function of a transportation problem is to minimize the total
transportation cost.

11.3 FORMULATION AND SOLUTION OF TRANSPORTATION

MODELS
The transportation problem is formulated in form of a matrix structure. The rows
represent the different sources of supply and column market or the demand centers. The
unit costs of transportation are placed in each cell which means the transportation cost
associated with a single unit. The maximum supply units (capacity) of a source are
shown against each row and the demand units (requirements) are shown against each
column.

2
Let us now try to formulate the problem in Table 11.1
Table 11.1: Transportation Problem

Markets Retail Agency Output

M1 M2 M3 M4
Factory A c11 c12 c13 c14 b1
B c21 c22 c23 c24 b2
C c31 c32 c33 c34 b3
Demand d1 d2 d3 d4 ∑ 𝑏𝑗 ⁄∑ 𝑑
𝑗

For the data given above, the mathematical model will be:
Objective Function
Minimize Z =
𝑐11 𝑥11 +𝑐12 𝑥12 +𝑐13 𝑥13 +𝑐14 𝑥14 +𝑐21 𝑥21 +𝑐22 𝑥22 +𝑐23 𝑥23 +𝑐24 𝑥24 +𝑐31 𝑥31 +𝑐32 𝑥32 +𝑐33 𝑥33
+𝑐34 𝑥34
Constraints
𝑎11 𝑥11 +𝑎12 𝑥12 +𝑎13 𝑥13 +𝑎14 𝑥14 ≤ b1 (because the sum must be less than or equal to the
available capacity).
𝑎21 𝑥21 +𝑎22 𝑥22 +𝑎23 𝑥23 +𝑎24 𝑥24 ≤ b2
𝑎31 𝑥31 +𝑎32 𝑥32 +𝑎33 𝑥33 +𝑎34 𝑥34 ≤ b3

Mixed Structural Constraints

𝑎11 𝑥11 +𝑎21 𝑥21 +𝑎31 𝑥31 ≥ d1 (because the sum must be greater than or equal to the
demand of the market, we cannot send less than what is required).
𝑎12 𝑥12 +𝑎22 𝑥22 +𝑎32 𝑥32 ≥ d2
𝑎13 𝑥13 +𝑎23 𝑥23 +𝑎33 𝑥33 ≥ d3
𝑎14 𝑥14 +𝑎24 𝑥24 +𝑎34 𝑥34 ≥ d4
All 𝑥𝑖𝑗 and 𝑥𝑗𝑖 are ≥ 0 where i = 1, 2, 3 and j = 1, 2, 3, 4. (This is because we cannot
supply negative elements).
Let us try to understand the structure with the help of an illustration. The matrix in
Table 11.1 shows a transportation formulation. It can be seen that a retail agency is
dealing with three suppliers (factories A, B, and C) and four markets (M1, M2, M3, and
M4). The maximum supply from factory A is 200 units, factory B is 300 units and
factory C is 300 units. The maximum demands from markets M1, M2, M3, and M4 are
200 each. The cells represent the unit transportation cost between the source and

3
destination. The transportation cost between factories A to market M1 is 9 Rs per unit
(represented by cell (A, M1)).
Our objectives in solving the transportation problem will be to minimize the total
transportation cost by making a suitable allocation of units between the supply source
and the destination.
Table 11.2: Retail Agency
Market M1 M2 M3 M4 Output
A 9 6 7 5 200
Factory

B 7 7 6 7 300
C 8 7 6 10 300
Demand 200 200 200 200 800 (Total)

The above problem has got the following properties:

1. It has an objective function.
2. It has structural constraints
3. It has non-negativity constraints
4. The relationship between the variables and the constraints is linear.

11.4 SOLUTION TO TRANSPORTATION PROBLEM

The solution to the Transportation Model involves making a transportation model (in
the form of a matrix), finding a basic feasible solution, performing an optimality test,
and iterating towards an optimal solution if required. The scope of the present Unit is
restricted to obtaining the initial basic feasible solution and hence the discussions on
checking for optimality and iterating towards the optimal solutions are restricted.
Following are the methods that can be used for solving transportation problems;
1. North – West Corner Rule (N.W. Corner Rule)
2. Least Cost Method (LCM)
3. Vogel’s Approximation Method (VAM)

11.4.1 North-West Corner Rule or North-West Corner Method

(NWCM)
Step 1 : Balance the problem. That is see whether ∑ 𝑏𝑖 =∑ 𝑑𝑖 . If not open a dummy
column or dummy row as the case may be and balance the problem.
Step 2 : Start from the left-hand side top corner or cell and make allocations
depending on the availability and requirement constraint. If the availability

4
 constraint is less than the requirement constraint, then that cell makes an
allocation in units that is equal to the availability constraint.
In general, verify which is the smallest among the availability and
requirement and allocate the smallest one to the cell under question. Then
proceed to allocate either sidewise or downward to satisfy the rim
requirement. Continue this until all the allocations are over.
Step 3 : Once all the allocations are over, i.e. both rim requirements (column and row
i.e., availability and requirement constraints) are satisfied, write allocations
and calculate the cost of transportation.
The technique is illustrated as follows;
Retail Agency
Market M1 M2 M3 M4 Output
A 9 6 7 5 200
Factory

B 7 7 6 7 300
C 8 7 6 10 300
Demand 200 200 200 200 800
(Total)
We select the topmost N-W corner cell (A, M1). The maximum allocation we can make
for this cell is 200 units (minimum of the demand and supply value for the respective
cell). This cancels out the further supply from factory A to other markets as the source
has exhausted its supply. Also, the demand of market M1 is met.
Retail Agency
Market M1 M2 M3 M4 Output
A 9 10 7 5 200 0
Factory

(200)
B 13 7 6 7 300
C 8 11 12 10 300
Demand 200 200 200 200 800
0 (Total)
Now select the next N-W corner (i.e. B, M2) and repeat the procedure.
Retail Agency
Market M1 M2 M3 M4 Output
A 9 10 7 5 200
(200) 0
Factory

B 13 7 6 7 300
(200) 100
C 8 11 12 10 300

5
 Demand 200 200 200 200 800 (Total)
0 0
Finally, we get the allocation as shown
Retail Agency
Market M1 M2 M3 M4 Output
A 9 10 7 5 200
(200) 0
Factory

B 13 7 6 7 300
(200) (100) 100 0
C 8 11 12 10 300
(100) (200) 200 0
Demand 200 200 200 200 800
0 0 100 0 (Total)
0
The initial basic solution is given as
xAM1= 200, xBM2 = 200, xBM3 = 100, xCM3 = 100 and xCM4 = 200
The corresponding transportation cost is
(9200) + (7200) + (6100) + (12100) + (10200) = Rs. 7000

11.4.2 Least Cost Method

Step 1 : Identify the lowest cost cell in the given matrix. When more than one cell has
the same cost, then both the cells are competing for allocation. This situation
in transportation problems is known as a tie.
Step 2 : To break the tie select any cell for allocation. Make allocation to this cell
either to satisfy availability constraint or requirement constraint.
Step 3 : Once one of these is satisfied then mark crosses (×) in all the cells in the row
or column whichever has been completely allocated.
Step 4 : Next search for the lowest cost cell.
Retail Agency
Market M1 M2 M3 M4 Output
A 9 10 7 5 200
Factory

B 13 7 6 7 300
C 8 11 12 10 300
Demand 200 200 200 200 800
(Total)
The least-cost cell in the above matrix is (A, M4) with an associated cost of 5 Rs/unit.

6
We proceed with the cell selection and make the allocation as studied earlier.

Retail Agency
Market M1 M2 M3 M4 Output
A 9 10 7 5 200
Factory (200) 0
B 13 7 6 7 300
C 8 11 12 10 300
Demand 200 200 200 200 800
0 (Total)

The next matrix minimum value is Rs. 6, i.e. cell (B, M3)
Retail Agency
Market M1 M2 M3 M4 Output
A 9 10 7 5 200
(200) 0
Factory

B 13 7 6 7 300
(200) (100)
C 8 11 12 10 300
Demand 200 200 200 200 800
0 0 (Total)

Repeating the above procedure we get the following allocations;

Retail Agency
Market M1 M2 M3 M4 Output
A 9 10 7 5 200
(200) 0
Factory

B 13 7 6 7 300
(100) (200) 100
C 8 11 12 10 300
(200) (100) 100 0
Demand 200 200 200 200 800
0 100 0 0 (Total)
0
The initial basic solution is given as
xAM4= 200, xBM2 = 100, xBM3 = 200, xCM1 = 200 and xCM2 = 100

7
The corresponding transportation cost is
(5200) + (7100) + (6200) + (8200) + (11100) = Rs. 5600
It can be observed that the minimum transportation cost obtained by the least cost
method is much lower than the corresponding cost of the solution developed by using
the northwest corner method.

11.4.3 Vogel Approximation Method (VAM)

VAM is an improved version of the least-cost method that generally produces better
solutions. The steps involved in this method are:
Step 1 : For each row (column) with strictly positive capacity (requirement),
determine a penalty by subtracting the smallest unit cost element in the row
(column) from the next smallest unit cost element in the same row (column).
Step 2 : Identify the row or column with the largest penalty among all the rows and
columns. If the penalties corresponding to two or more rows or columns are
equal we select the topmost rows and the extreme left column.
Step 3: We select Xij as a basic variable if Cij is the minimum cost in the row or
column with the largest penalty. We choose the numerical value of Xij as high
as possible subject to the row and the column constraints. Depending upon
whether ai or bj is the smaller of the two ith row or jth column is crossed out.
Step 4 : Step 2 is now performed on the uncrossed-out rows and columns until all the
basic variables have been satisfied.
The technique is demonstrated as follows;
Retail Agency
Market M1 M2 M3 M4 Output
A 9 10 7 5 200
Factory

B 13 7 6 7 300
C 8 11 12 10 300
Demand 200 200 200 200 800
(Total)

The penalty is determined as follows;

For the columns: M1= 9 - 8 = 1, M2 = 10 - 7 = 3, M3 = 7 – 6 = 1 and M4 = 7 – 5 = 2
For the rows: A = 7 - 5 = 2, B = 7 – 6 = 1 and C = 10 – 8 = 2
We find that the highest penalty is 3 for column M2. Next, we see that the least cost in
that column is 7 i.e. cell (B, M2). The allocation is made for this cell as shown below.

8
 Retail Agency
Market M1 M2 M3 M4 Output Penalty
A 9 10 7 5 200 2
Factory

B 13 7 6 7 300 1
(200) 100
C 8 11 12 10 300 2
Demand 200 200 200 200 800
(Total)
Penalty 1 3 1 2

Now we get the revised penalties by not considering the canceled cells.
Retail Agency
Market M1 M2 M3 M4 Output Penalty
A 9 10 7 5 200 2
2
Factory

B 13 7 6 7 300 1 1
(200) 100
C 8 11 12 10 300 2 2
Demand 200 200 200 200 800
0 (Total)
Penalty 1 3 1 2
1 - 1 2
As there is a tie with three equal penalties of 2. We consider the topmost row and
proceed with the allocation and then get the revised penalties.
Retail Agency
Market M1 M2 M3 M4 Output Penalty
A 9 10 7 5 200 2 2 -
(200) 0
Factory

B 13 7 6 7 300 1 1 7
(200) 100
C 8 11 12 10 300 2 2 4
Demand 200 200 200 200 800
0 0 (Total)

Penalty 1 3 1 2

9
 1 - 1 2
5 - 6 -
Repeating the procedure we get the final allocation as;
Retail Agency
Market M1 M2 M3 M4 Output Penalty
A 9 10 7 5 200 2 2 -
(200) 0
Factory

B 13 7 6 7 300 1 1 7
(200) (100) 100 0
C 8 11 12 10 300 2 2 4
(200) (100) 100 0
Demand 200 200 200 200 800
0 100 0 (Total)

Penalty 1 3 1 2
1 - 1 2
5 - 6 -

The initial basic solution is given as

xAM4= 200, xBM2 = 100, xBM3 = 200, xCM1 = 200 and xCM3 = 100
The corresponding transportation cost is
(5200) + (7200) + (6100) + (8200) + (12100) = Rs. 5800
Apart from the three discussed methodologies for getting the initial basic feasible
solution we also have some less popular techniques like row minima and column
minima. The same is discussed below.

11.4.4 Row Minima Method

This method consists in allocating as much as possible in the lowest-cost cell of the
first center so that either the capacity of the first plant is exhausted or the requirement
at the jth distribution center is satisfied or both. In case of a tie among the cost, select
arbitrarily.
Three cases arise:
i. if the capacity of the first plant is completely exhausted, cross off the first row
and proceed to the second row.
ii. if the requirement at the jth distribution center is satisfied, cross off the jth
column and reconsider the first row with the remaining capacity.

10
 iii. if the capacity of the first plant as well as the requirement at the jth distribution
center is completely satisfied, make a zero allocation in the second-lowest-cost
cell of the first row. Cross off the row as well as the jth column and move down
to the second row.
Continue the process for the resulting reduced transportation table until all the
conditions (supply and requirement conditions) are satisfied.
Distribution centers
1 2 3 4 Output
1 2 3 11 7 6/0
(6)
Plants

2 1 0 6 1 1/0
(1)
3 5 8 15 9 10/9/5/2/0
(1) (4) (3) (2)
Demand 7/1/0 5/4/0 3/0 2/0 17

In this problem, we first allocate to cell (1, 1) in the first row as it contains the
minimum cost 2. We allocate min. (6, 7) (6) in this cell. This exhausts the supply
capacity of plant I and thus the first row is crossed off. The next allocation, in the
resulting reduced matrix, is made in cell (2, 2) of row 2 as it contains the minimum cost
0 in that row. We allocate min. (1, 5) (1) in this cell. This exhausts the supply capacity
of plant 2 and thus the second row is crossed off. The next allocation, in the resulting
reduced matrix, is made in cell (3, 1) of row 3 as it contains the minimum cost of 5
in that row. We allocate min. (1, 10) (1) in this cell. This exhausts the required
condition of a distribution center 1 and hence the first column is crossed off.
Proceeding in this way we allocate (4), (2) and (3) units to cells (3, 2), (3, 4), and (3, 3)
till all the conditions are met. The resulting matrix is shown in the Table above.
The transportation cost associated with this solution is
Z = Rs. [2  6 + 0  1 + 5  1 + 8  4 + 15  3 + 9  2]  100 = Rs. 11, 200

11.4.5 Column Minima Method

This method consists of allocating as much as possible in the lowest-cost cell of the
first column so that either the demand of the first distribution center is satisfied or the
capacity of the ith plant is exhausted or both. In case of a tie among the lowest cost
cells in the column, select arbitrarily. Three cases arise:
i. if the requirement of the first distribution center is satisfied, cross off the
first column and move right to the second column.
ii. if the capacity of the ith plant is satisfied, cross off the ith row and reconsider
the first column with the remaining requirement.

11
 i. if the requirement of the first distribution center, as well as the capacity of
the ith plant, are completely satisfied, make a zero allocation in the second-
lowest-cost cell of the first column. Cross off the column as well as the ith
row and move right to the second column.
Continue the process for the resulting reduced transportation table until all the
conditions are satisfied.
Distribution centers
1 2 3 4 Output
1 2 3 11 7 6/0
(6)
Plants

2 1 0 6 1 1/0
(1)
3 5 8 15 9 10/9/5/2/0
(5) (3) (2)
Demand 7/6/0 5/4/0 3/0 2/0 17

In the given problem we allocate first to cell (2, 1) in the first column as it contains
the minimum cost 1. We allocate min. (1, 7) = (1) in this cell. This exhausts the
supply capacity of plant 2 and thus the second row is crossed off.
The next allocation in the resulting reduced matrix is made in cell (1, 1) of column 1 as
it contains the second-lowest-cost 2 in that column. We allocate min. (6, 6) = (6) in
this cell. This exhausts the supply capacity of plant 1 as well as the requirement of
distribution centre 1. Therefore, we allocate zero in cell (3, 1) of the first column,
cross off the first row and first column and move on to the second column. Proceeding
in this way we allocate (5), (3), and (2) to cells (3, 2), (3, 3), and (3, 4) till all the rim
conditions are met with. The resulting matrix is shown in the table above.
The transportation cost associated with the above solution is
Z= Rs. [2  6 + 1  1 + 8  5 + 15  3 + 9  2]  100 = Rs. 11,600

11.5 CASE OF UNBALANCED PROBLEM

In the above problems discussed so far, we had the total supply matching the total
demand. The case is called a balanced transportation problem. However, in certain
scenarios, we may come across situations wherein the total supply and demand may not
match. This arising case is referred to as an unbalanced transportation problem. In such
cases, we will be required to add a dummy row or a dummy column with the cell
values representing zero costs of transportation and the column or the row totals
representing the units to balance the supply and demand.
We now discuss how to solve an unbalanced transportation problem

12
Consider Cement bags are to be dispatched to the 4 different markets (1, 2, 3, and 4)
from three manufacturing locations (A, B, C). The details of the supply and demand
requirements with the unit transportation cost are given in the following matrix.

1 2 3 4 Output
A 20 21 16 18 10
B 17 28 14 16 9
C 29 23 19 20 7
Demand 6 10 4 5 Unbalanced

It can be seen from the above matrix that it is not a balanced problem as the supply and
demand figures are not matching which is the basic condition to be fulfilled before
solving a transportation problem. It is therefore required to add a dummy row or a
column with 0 transportation cost so that the matrix is balanced and we proceed to get
the initial basic feasible solution. For the above problem we add a dummy column to
balance the problem as shown below:
1 2 3 4 5 Output
A 20 21 16 18 0 10
B 17 28 14 16 0 9
C 29 23 19 20 0 7
Demand 6 10 4 5 1 26 (Balanced)

Solving the above problem with the VAM technique gives the following solution:

1 2 3 4 5 Output Penalty
A 20 21 16 18 0 10/6/2/0 16 2 2 3
(4) (4) (2)
B 17 28 14 16 0 9/3/0 14 2 2 12
(6) (3)
C 29 23 19 20 0 7/6/0 19 1 1 3
(6) (1)
Demand 6/0 10/4/0 4/0 5/2/0 1/0 26
Penalty 3 2 2 2 0
3 2 2 2 -
- 2 2 2 -
- 2 - 2 -
- 2 - 2 -

13
This is the initial feasible solution with the total transportation cost as below:
214 + 164 + 182 + 176 + 163 + 236 = Rs. 472

11.6 TRANSSHIPMENT PROBLEM

We may come across a certain situation, in which a company (or Companies) may be
producing the product to their capacity, but the demand arises for these products during
a certain period in the year or the demand may reach the peak point in a certain period
of the year. This is particularly true that products like Cool drinks, Textbooks,
Notebooks, crackers, etc. The normal demand for such products will exist, throughout
the year, but the demand may reach peak points during certain months of the year. It
may not possible for all the companies to put together to satisfy the demand during
peak months. It is not possible to produce beyond the capacity of the plant. Hence
many companies have their regular production throughout the year, and after satisfying
the existing demand, they stock the excess production in a warehouse and satisfy the
peak demand during the peak period by releasing the stock from the warehouse. This is
quite common in the business world. The only thing that we have to observe the
inventory carrying charges of the goods for the months for which it is stocked is to be
charged to the consumer.
For example, crackers, though their production cost is very much less, they are sold at
very high prices because of inventory carrying charges. When a company stocks its
goods in a warehouse and then sends the goods from the warehouse to the market, the
problem is known as
Transshipment problem.
Transshipment is a special case of transportation problem in which shipments may be
required to be allowed between supply points or between demand points and not
directly from the source point to the destinations. These in-between points are referred
to as transshipment points.
In some transshipment problems, it is economical if the shipment passes through some
transient nodes between sources and destinations.
Types of the transshipment problem
1. Transshipment Problem with Sources
2. Destinations Acting Transient Nodes
Following Figure 11.1 show a simple form of transshipment problem in which the
sources and destinations act as transient nodes:

14
 S1 D1

S2 D2

S3 S3

Si Si

Sn Sn

Fig. 11.1: Transshipment Model.

In the figure, consider the shipment of items from source 1 to destination 2. The
shipment from source 1 can pass through source 2 and destination I before it reaches
the specified destination 2. Since, in this case, the shipment passes through some
transient nodes, the arrangement is termed as the transshipment model.
In the transshipment model, the main objective is to find the optimal shipping pattern
so that the total cost of transportation is minimized.

Check Your Progress 11.1

1. The transportation problem is formulated in form of a _________ structure.

2. Transportation Model involves

a. Finding a basic feasible solution

b. Performing optimality test

c. Iterating toward an optimal solution

d. All the above

3. The least-cost method is also known as the matrix maximum method

a. True

b. False

15
4. VAM is an improved version of the North West corner method that generally
produces better solutions.

a. True

b. False

5. The transportation problem is basically a

a. Maximization model

b. Minimization model

c. Transhipment problem

d. Iconic model

6. A transportation model must have the same number of rows as columns.

a. True

b. False

7. In the transportation model, we must always make the total supply equal to the
total demand.

a. True

b. False

8. In a transportation problem, we must make the number of __________ and

__________ equal.

a. destinations; sources

b. units supplied; units demanded

c. columns; rows

d. positive cost coefficients; negative cost coefficients

e. warehouses; suppliers

9. ______________ is a point that can both receive goods from other points and
send goods to other points.

a. Supply point

b. Demand point

c. Transshipment point

16
 d. None of the above

10. The _____________ solution to a transhipment problem can be found by

solving a transportation problem.

a. Optimal

b. Feasible

c. Initial

d. All the above

11. XYZ Inc. manufactures modular kitchen and dining tables in all its four
furniture manufacturing plants. It has 5 warehouses across the country. One
transportation problem can be used to determine how to ship modular kitchen
and dining tables.

a. True

b. False

12. Solve the given problem by North West Corner Method

Demand centers

1 2 3 Output
Source 6 4 1 50
1
2 3 8 7 40
3 4 4 2 60
Demand 20 95 35 150
13. Solve the problem by Least Cost Method

1 2 3 Output
Source 6 4 1 50
1
2 3 8 7 40
3 4 4 2 60
Demand 20 95 35 150

14. Solve the problem by VAM

17
 1 2 3 Output
Source 6 4 1 50
1
2 3 8 7 40
3 4 4 2 60
Demand 20 95 35 150

15. A company manufacturing agricultural equipment has two plants located at

Pune and Nagpur with a weekly capacity of 200 units and 100 units
respectively. The company supplies this equipment to its 4 showrooms situated
in Mumbai, Gujarat, Jaipur, and Bihar which have a demand of 75, 100, 100,
and 30 units, respectively. The cost per unit (in Rs.) is shown in the following
table:

Mumbai Gujarat Jaipur Bihar

Pune 90 90 100 100
Nagpur 50 70 130 85

Design the production programme to minimize the total cost of transportation.

11.7 ASSIGNMENT PROBLEM

The transportation problem helps the manager to minimize the cost of transportation
incurred while moving products from supply sources to demand destinations. An
assignment is a special type of transportation problem which is used when managers
are faced with problems of what jobs should they assign to what machines. What
person they should assign to what duties? (one job to one machine). In Assignment
problems, the demand and supply units are constrained to 1. Furthermore, only one
assignment is to be made in each row or column as opposed to a transportation problem
wherein we can have multiple assignments in a given row or column.
The assignment becomes a problem because people possess varying abilities for
performing different jobs and therefore the cost of performing the jobs by different
people is different. The objective function would therefore the minimization of cost or
time.

11.7.1 Formulation of Assignment Problem

Similar to a Transportation problem, the Assignment problem is also formulated in a
matrix form. A sample format is shown in the following table:

Jobs Cost in Rs.

18
 Machines
𝑴𝟏 𝑴𝟐 𝑴𝟑 𝑴𝟒
𝐽1 5 7 11 6
𝐽2 8 5 9 6
𝐽3 4 7 10 7
𝐽4 10 4 8 3
In the above table, the rows represent the four jobs (J1, J2, J3, and J4) which are assigned
to the four machines (M1, M2, M3, and M4) and the cells represent the cost of allocation
i.e. if job J1 is to be assigned to machine M1 it will cost Rs. 5. Now the managers'
concern is to make the allocations of the jobs to the machines such that it will result in
minimum cost. Unlike in the Transportation problem where we have the supply and the
demand constraints, the constraints in assignment problems are set equal to one which
means that we can have only one assignment for each column and row.
Step 1 : Key decision is what to whom i.e. which jobs be assigned to which machine
or what are the ‘n’ optimum assignments on a 1-1 basis.
Step 2 : Feasible alternatives are n! possible arrangements for the n × n assignment
situation. In the given situation there are 4 different arrangements.
Step 3 : Let cij be the cost of assigning ith person to the jth job. The assignment problem
can be stated in the form of n × n cost matrix or effectiveness matrix cij as
shown in the Table.

19
 Jobs
1 2 3 … j … n Supply
1 c11 c12 c13 … c1j … c1n 1
2 c21 c22 c23 … c2j … c2n 1
Persons 3 c31 c32 c33 … c3j … c3n 1
: : : : : : : : :
: : : : : : : : :
i ci1 ci2 ci3 … cij … cin 1
: : : : : : : : :
: : : : : : : : :
n cn1 cn2 cn3 … cnj … cnn 1
Demand 1 1 1 … 1 … 1 N
Let xij denote the assignment of person i to job j such that,
1 𝑖𝑓 𝑝𝑒𝑟𝑠𝑜𝑛 𝑖 𝑖𝑠 𝑎𝑠𝑠𝑖𝑔𝑛𝑒𝑑 𝑡𝑜 𝑗𝑜𝑏 𝑗
xij = {
0 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
The assignment problem is simply the following LPP:
Minimize z = ∑𝑛𝑖=1 ∑𝑛𝑗=1 𝑐𝑖𝑗 𝑥𝑖𝑗
Subject to the constraints
∑𝑛𝑗=1 𝑥𝑖𝑗 = 1, for i = 1, 2………n

and, ∑𝑛𝑗=1 𝑥𝑖𝑗 = 1, for j = 1, 2………n

with 𝑥𝑖𝑗 = 0 or , for j = 1, 2………n
Note that an assignment problem is an n × n transportation problem with each ai = bj=1.

11.7.2 Solution to Assignment Problem (Hungarian Method)

After formulation, the next logical step is to obtain the solution or make the optimal
allocation. We use the Hungarian Method as suggested by Mr. Koning of Hungary. It
involves a rapid reduction of the original matrix and finding of a set of n independent
zero’s, one in each row and column, which results in an optimal solution. The method
consists of the following steps:
Step 1 : Locate the smallest cost element row of the given cost matrix. Then subtract
this smallest element from each element in the row including itself. As a
result, there shall be at least one zero in each of the new matrices.
Step 2 : Now consider each column of the reduced cost matrix from step 1 and locate
the smallest element in it. Subtract a value from each element of the column

20
 including itself. As a result, there shall be at least one zero in each of the new
matrices.
Step 3 : Draw the minimum number of horizontal and vertical lines to cover all zero
elements. If the number of lines drawn is equal to the number of
rows/columns (n), the solution is optimal. Proceed to step 6.
Step 4 : Now if in the matrix, the number of lines is less than the number of
rows/columns, go to step 5.
Step 5 : Select the smallest uncovered cost element of the modified matrix from step
3. Subtract this element from all uncovered elements and add it to each value
located at the interactions of any two lines.
Step 5 : Repeat steps 4 and 5 till an optimal solution of obtained i.e. number of lines
drawn equals the number of columns/rows.
Step 6 : Make feasible job assignments on zero elements.
Let us now examine the application of the Hungarian method enumerated above with
help of an illustrative example. Suppose an agriculturist has 5 seeders that are used to
spread seeds in the soil. He has to make a decision on assigning the seeders to five
different fields based on the cost incurred on using the seeders. The details of the cost
(in Rs. 1000’s) incurred. Using the following cost matrix let's determine (a) optimal
seeder assignment (b) the cost of assignments.
Fields
1 2 3 4 5
A 10 3 3 2 8
B 9 7 8 2 7
Seeder C 7 5 6 2 4
D 3 5 8 2 4
E [ 9 10 9 6 10]

Now to obtain the solution to this problem;

Select the smallest element in each row and subtract this smallest element from all the
elements in its row.
Fields
1 2 3 4 5
A 8 1 1 0 6
B 7 5 6 0 5
Seeder

C 5 3 4 0 2
D 1 3 6 0 2
E 3 4 3 0 4

21
Subtract the minimum element from each column and subtract this element from all the
elements in its column. With this, we get the first modified matrix.
Fields
1 2 3 4 5
A 7 0 0 0 4
B 6 4 5 0 3
Seeder

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 zero
(horizontal or vertical).
Fields
1 2 3 4 5
A 7 0 0 0 4
B 6 4 5 0 3
Seeder

C 4 2 3 0 0
D 0 2 5 0 0
E 2 3 2 0 2

The number of lines drawn to cover all zeros is N = 4

The order of matrix is n = 5 (5  5 square matrix)
Hence, N < n.
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. The smallest uncovered element in the above matrix is 2. With the
new modified we can check with lines that the no. of lines required to cover all zeros is
= 5, which is N = n and hence optimal assignment can be made which is discussed
below:

22
 Fields
1 2 3 4 5
A 9 0 0 2 6
B 6 2 3 0 3

Seeder
C 4 0 1 0 0
D 0 0 3 0 0
E 2 1 0 0 2
Now we determine the optimum assignment
The first row contains more than one zero. So proceed to the 2 nd row. It has exactly one
zero. The corresponding cell is (B, 4). Circle this zero thus, making an assignment.
Mark (x) for all other zeros in its column. Showing that they can’t be used for making
other assignments. Now row 5 as a single zero in the cell (E, 3). Make an assignment in
this cell and cross the 2nd zero in the 3rd column.
Fields
1 2 3 4 5
A 9 0 0 2 6

B 6 2 3 0 3
Seeder

C 4 0 1 0 0
D 0 0 3 0 0
E 2 1 0 0 2

Now row 1 has a single zero in column 2. i.e. in the cell (A, 2) make an assignment in
this cell and cross the other zero in the 2nd column. This leads to a single zero in
column 1 of the cell (D, 1), make an assignment in this cell. 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.

23
 Optimal assignment and optimum cost of the assignment
Field Seeder Machine Cost (in 1000’s Rs.)
1 D 3
2 A 3
3 E 9
4 B 2
5 C 4
Rs 21

Therefore, 1 → 𝐷, 2 → 𝐴, 3 → 𝐸, 4 → 𝐵, 5 → 𝐶 with minimum cost Rs. 21,000 /-

Example 1:
A farmer intends to plant five different crops in each of five equal-sized paddocks. The
paddocks vary in soil fertility and the crops vary in their nutrient requirements so the
fertilizers which must be applied depending on which crop is grown in which paddock.
The associated cost matrix showing fertilizer costs for each crop in each paddock is
given in the table below.
Since the paddocks are of equal size it may be assumed that the total revenue will be
the same whatever assignment is adopted. The farmers' objective in this decision
problem is, therefore, to assign paddocks to crops in a manner that minimizes the total
fertilizer cost. Mixed cropping of the paddocks is not permitted, nor may the same crop
be grown in two paddocks.
Table11.1: Initial Table of Fertiliser Costs (Rs.)

Paddocks
Crops A B C D E
1 1300 800 1600 1800 1900
2 900 1500 2400 900 1200
3 1300 900 400 400 400
4 600 1200 100 800 1300
5 1500 1700 1800 1200 2000
Solution:
We form the first modified matrix by subtracting the minimum element from all the
elements' respective rows and the same with the respective column. The resulting
matrix is;

24
 Paddocks
A B C D E
1 500 0 800 1000 1100
2 0 600 1500 0 300

Crops
3 800 500 0 0 0
4 0 600 400 200 700
5 300 500 600 0 800

Since each column has a minimum element of 0. We have the first modified matrix.
Now we draw the minimum number of lines to cover all zeros.
Paddocks
A B C D E
1 500 0 800 1000 1100
2 0 600 1500 0 300
Crops

3 800 500 0 0 0
4 0 600 400 200 700
5 300 500 600 0 800

The number of lines drawn to cover zero is N = 4 which is less than the order of the
matrix (5  5 matrix). Therefore, we proceed with the further modification that is; we
find the second modified matrix by subtracting the smallest uncovered element from
the entire uncovered element and adding to the element which is at the point of
intersection of lines. We get the resulting matrix.
Paddocks
A B C D E
1 500 0 500 1000 800
2 0 600 1200 0 0
Crops

3 1100 800 0 300 0

4 0 600 100 200 400
5 300 500 300 0 500

Numbers of lines drawn to cover all zeros = 5, which is equal to the order of the matrix.
Hence we can proceed with the assignment as follows;

25
 Paddocks
A B C D E
1 500 0 500 1000 800

2 0 600 1200 0 0
Crops 3 1100 800 300 0
0

4 0 600 100 200 400

5 300 500 300 0 500

All five crops have been assigned to 5 different paddocks.

Here the optimal assignment is.
Crops Paddocks
1 B
2 E
3 C
4 A
5 D

Minimum (Total cost) = 800 + 1200 + 400 + 600 + 1200 = Rs. 4200

Check Your Progress 11.2

1. An assignment problem:

a. Will always have an integer solution.

b. Has all supplies and demands equal to 0.

c. Always has the demand greater than the supply.

d. All of the above.

e. None of the above.

2. Which of the following is an assumption of assignment problems?

a. The number of assignees and the number of tasks are the same

b. The objective is to minimize the number of assignments not made.

c. Each task is to be performed by exactly one assignee.

26
 d. a. and c. only.

e. None of the above.

3. The method used for solving an assignment problem is called as:

a. Reduced matrix method

b. MODI method

c. Hungarian method

d. None of the above

4. Assignment Problem is basically a

a. Maximization Problem.

b. Minimization Problem

c. Dual Problem

d. Primal Problem

5. The Assignment Problem is solved by

a. Simplex method

b. Graphical method

c. Vector method

d. Hungarian method

6. The objective of the Assignment problem is to minimize cost or minimize

working hours

a. True

b. False

7. The assignment problem

a. Requires that only one activity be assigned to each resource

b. Is a special case of a transportation problem

c. Can be used to maximize resources

d. All the above

8. Hungarian Method is suggested by Mr. Koning of Hungary.

27
 a. True

b. False

9. The Hungarian method for solving an assignment problem can also be used to
solve

a. a transportation problem

b. a traveling salesman problem

c. both (a) and (b)

d. only (b)

10. 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 III IV V
A 30 25 33 35 36
Machines

B 23 29 38 23 26
C 30 27 22 22 22
D 25 31 29 27 32
E 27 29 30 24 32
[

11.7.3 Unbalanced Assignment Problem

An assignment problem is said to be unbalanced when the matrix is not a square
matrix. In such cases to balance it, we add a dummy row or dummy column with all the
entries as zero. Let’s consider the following example with an unbalanced matrix.
Illustrative
There are four farms to be assigned to the daily wage contract worker. Only one farm
could be assigned to one worker. The amount of time in hours required for the activities
on a farm is given in the following matrix.

28
 Contract Worker (Daily Wages)
Farm 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
Find an optimum assignment of workers to the farms to minimize the total activity time
and also find for which farm no worker is assigned. What is the total time to complete
all the activities?
Solution:
Since the cost matrix is not a square matrix, the problem is unbalanced. We add a
dummy farm 5 with corresponding entries equal to zero.
Modified matrix
Contract Worker (Daily Wages)
A B C D E
1 4 3 6 2 7
2 10 12 11 14 16
Farm

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 row.
Contract Worker (Daily Wages)
A B C D E
1 2 1 4 0 5
2 0 2 1 4 6
Farm

3 3 2 1 0 4
4 2 1 0 3 0
5 0 0 0 0 0

Since each column has a minimum element as zero, we draw the minimum number of
lines to cover all zeros.

29
 Contract Worker (Daily Wages)
A B C D E
1 2 0 4 0 5
2 0 2 1 4 6

Farm
3 3 2 1 0 4
4 3 1 0 3 0
5 0 0 0 1 0

The number of lines to cover all zeros N = 4 < the order of the 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. Now
repeat the same procedure of covering zeros with lines we get.
Contract Worker (Daily Wages)
A B C D E
1 2 1 4 0 5
2 0 2 1 4 6
Farm

3 3 2 1 0 4
4 2 1 0 3 0
5 0 0 0 0 0
Here the number of lines drawn to cover all zero = 5 = Order of matrix. Therefore we
can make the assignment as follows;
Contract Worker (Daily Wages)
A B C D E
1 2 0 3 0 4

2 0 2 1 0 6
Farm

3 3 2 1 0 4
4 2 1 0 3 0

5 0 0 0 0 0

Optimum assignment
Farm Worker
1 A
2 C
3 D
4 E

30
For Worker E no job is assigned and will remain idle.
Optimum (minimum) time = 4 + 11 + 1 + 6 = 22 hours.

Check Your Progress 11.3

1. An assignment problem is said to be unbalanced when the matrix is a square

matrix

a. True

b. False

2. The assignment problem is called as _________ problem when the cost matrix
is not a square matrix.

3. The Hungarian method can be used when a cost matrix is a square matrix.

a. True

b. False

4. To balance the assignment matrix we have to:

a. Open a dummy row

b. Open a dummy column

c. Open either a dummy row or column depending on the situation

d. None of the above.

5. A company has 4 machines to do 3 jobs. Each job can be assigned to one and
only one machine. The cost of each job on each machine is given below.
Determine the job assignments which will minimize the total cost.

Machines

W X Y `Z
A 18 24 28 32
Job B 8 13 17 18
C 10 15 19 22

11.8 LET US SUM UP

In this unit we have discussed the following points:
 Transportation problem deals with the objective of minimization of transportation
costs.

31
 The different methods available for obtaining an initial basic feasible solution are
North West Corner Method, Least Cost Matrix Method, Row Minima Method,
Colum Minima Method, and Vogel’s Approximation method.
 There are cases of unbalanced transportation problems wherein the demand does
not match the supply quantities.
 The assignment problem is a special type of Transportation problem with the
objective of obtaining a minimum assignment cost subject to demand and supply-
constrained to 1.
 The Hungarian method is used to get the optimal solution for the Assignment
problem.
 There exists a case of an unbalanced assignment problem where the matrix is not a
square matrix.
11.9 KEYWORDS
Assignment Problem : It is a special case of transportation problem, use to
minimize the cost or time of completing a number of
jobs by a number of persons.
Least cost method : This is another method that takes into account the
minimum unit cost of transportation for obtaining an
initial basic solution.
North West corner rule : It’s an easy and efficient method to get an initial
solution. North West corner rule does not take into
account the cost of transportation on any route of
transportation.
Transhipment Problem : It is defined as the shipment of goods (raw material,
semi-finished, finished goods) or containers to an
intermediate destination, then to yet another
destination.
Transportation Problem/ : It is concerned with shipping a commodity between
Model a set of sources (e.g. manufacturers) and a set of
destinations (e.g. retailers).
Unbalanced Assignment : If the number of jobs is not equal to the number of
Problem operators then the assignment problem is known as
an unbalanced assignment problem.
Unbalanced Problem : A Transportation Problem is said to be an
unbalanced transportation problem if the total
number of supply is not the same as total number of
demand. This is the case of an assignment problem
where the number of persons is not equal to the
number of jobs.
Vogel Approximation : It is also called as penalty and regret method, this
Method method is a heuristic and is preferred mostly.

32
11.10 SUGGESTED FURTHER READINGS/REFERENCES
1. Bronson, R. (1997), Schaum’s Outline of Operations Research, McGraw Hill
Professional.
2. Carter, M. W. and Price, C. C. (2001), Operations Research: A Practical
Introduction, Taylor and Francis.
3. Eiselt, H. A. and Sandblom, C. L. (2012), Operations Research: A Model-based
Approach, Second Edition, Springer.
4. Gupta, P. K. and Hira, D. S. (1992), Operations Research, S. Chand and Company,
New Delhi.
5. Hillier, F. S. and Lieberman, G. J. (2008), Introduction To Operations Research –
Concepts and Cases, Eight Edition, Tata McGraw-Hill Higher Education.
6. Hillier, F. S. and Lieberman, G. J. (2009), Introduction to Operations Research,
Ninth Edition, McGraw-Hill Higher Education.
7. Iyer, P.S. (2008), Operations Research, Tata McGraw-Hill Higher Education.
8. Kalavathy, S. (2002), Operation Research, Fourth Edition, Vikas Publishing House
Private Ltd.
9. Kothari, C. R. (2009), An Introduction to Operational Research, Third Edition,
Vikas Publishing House Private Ltd.
10. Mishra, D.N. Agarwal, S.K. (2009), Operation Research, Lucknow, India, Global
Media.
11. Morse, P. M. and Kimball, G. E. (2003), Methods of Operations Research, Dover
Publications Inc.
12. Mustafi, C. K. (2006), Operations Research – Methods and Practice, Third Edition,
New Age International (P) Ltd.
13. Panneerselvam, R. (2006), Operation Research, Second Edition, PHI Learning
Private Ltd, New Delhi, India.
14. R. Veerachamy and V. Ravi Kumar. (2010), Operations Research, I.K. International
Publishing House Pvt. Limited.
15. Ramamurthy, P. (2007), Operations Research, Daryaganj, Delhi, India, New Age
International (P) Ltd.
16. Shah, N. H, RAVI M. Gor, R. M. and Soni, H. (2007), Operations Research,
Prentice – Hall of India Private Ltd.
17. Sharma, A. (2009). Operations Research. Global Media, Himalaya Publishing
House.
18. Sharma, J.K. (2009), Operation Research – Theory and Application, Fourth Edition,
Macmillan Publishers India Ltd.

33
19. Sharma, J.K. (2010), Operations Research – Problems and Solutions, Third Edition,
Macmillan Publishers India Ltd.
20. Sharma, S.C. (2006), Introductory Operation Research, Discovery Publishing
House.
21. Shenoy, G.V., Srivastava, U.K. and Sharma, S.C. (2005), Operations Research for
Management, Second Edition, New Age International (P) Ltd.
22. Srinivasan, G. (2010), Operations Research – Principles and Applications, Second
Edition, PHI Learning Private Ltd, New Delhi, India.
23. Taha, H. A. (2008), Operations Research – An Introduction, Eight Edition, Prentice
– Hall of India Private Ltd.
24. Veerachamy, R. and Kumar, V. R. (2010), Operations Research, I.K. International
Publishing House Private Ltd.

11.11 ANSWERS TO CHECK YOUR PROGRESS

Check Your Progress 11.1
1. Matrix
2. d
3. False
4. False
5. b
6. b
7. a
8. b
9. c
10. a
11. b

34
Problem 12
This is a balanced Transportation Problem
Demand centers
1 2 3 Output
Source 6 4 1 50/30/0
1 (20) (30)
2 3 8 7 40/0
(40)
3 4 4 2 60/0
(25) (35)
Demand 20/0 95/65/25/0 35/0 150

Total transportation cost is as shown below;

620 + 430 + 840 + 425 +235 = Rs. 730

Problem 13
This is a balanced Transportation Problem
1 2 3 Output
Source 6 4 1 50/15/0
1 (15) (35)
2 3 8 7 40/20/0
(20) (20)
3 4 4 2 60/0
(60)
Demand 20/0 95/80/20/0 35/0 150

Total transportation cost is as shown below;

415 +135 +320 + 820 + 460 = Rs.555

35
 Problem 14
This is a balanced Transportation Problem
1 2 3 Output Penalty
1 6 4 (15) 1 50/15/0 3 3 4 4 -
(35)
Source

2 3 8 (20) 7 40/20/0 4 1 8 - -
(20)
3 4 4 (60) 2 60 2 2 4 4 4
Demand 20/0 95/75/60/0 35/0 150
Penalty 1 0 1
- 0 1
- 0 -
- 0 -
- 4 -
Total transportation cost is as shown below;
415 + 135 + 320 + 820 + 460 = Rs. 555
Problem 15
This is an unbalanced Transportation Problem
Delhi Kolkata Bangalore Hyderabad
Mumbai 90 90 100 100
Chennai 50 70 130 85
Demand  Capacity, Insert a dummy row Z.
Delhi Kolkata Bangalore Hyderabad Capacity Penalty
Mumbai 90 90 (75) 100 (95) 100 (30) 200/105/30/0 0 0 10 10
Chennai 50 70 (25) 130 85 100/25/0 20 20 15 15
(75)
Z 0 0 0 (5) 0 5/0 0 - - -
Demand 75/0 100/75/0 100/95/0 30/0 305
Penalty 50 70 100 85
40 20 30 15
- 20 30 15
- 20 - 15
- - - -
- - - -

36
Total transportation cost is as shown below;
9075 + 10095 + 10030 + 5075 + 7025 + 05 = Rs. 24,750 /-

Check Your Progress 11.2

1. a
2. d
3. c
4. b
5. d
6. False
7. d
8. True
9. b
Problem 10
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.
Operators
I II III IV V
A 5 0 8 10 11
Machines

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.
Operators
I II III IV V
A 5 0 5 10 8
Machines

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 number of lines drawn to cover all zeros is N = 4 < the order of matrix = 5 (5  5
matrix). Hence, we form the second modify matrix by subtracting the smallest

37
uncovered element from the remaining uncovered element and add to the element
which is at the point of intersection of lines.
N = 5. i.e. the number of lines drawn to cover all zero = order of the matrix. Hence we
can make the assignment.
I II III IV V
A 5 0 5 10 8

B 0 6 12 0 0
Machines

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 = Rs. 122.

Check Your Progress 11.3

1. False
2. Unbalanced
3. True
4. c
Problem 5
Since the cost matrix is not a square matrix we add a dummy row D with all the
elements 0.

38
 Machines
W X Y Z
A 18 24 28 32
B 8 13 17 18

Jobs
C 10 15 19 22
D 0 0 0 0

Subtract the minimum element in each row form all the elements in its row.
Machines
W X Y Z
A 0 6 10 14
B 0 5 9 10
Jobs

C 0 5 9 12
D 0 0 0 0

Since each column has minimum element we draw minimum number of lines to cover
all zeros.
Machines
W X Y Z
A 0 6 10 14
B 0 5 9 10
Jobs

C 0 5 9 12
D 0 0 0 0

∴ The number of lines drawn to cover all zero = 2 < the order of matrix. We form a
second modified matrix.
Machines
W X Y Z
A 0 1 5 9
B 0 0 4 5
Jobs

C 0 0 4 7
D 5 0 0 0

39
Here N = 3 < n =4.
Again we subtract the smallest uncovered element form all the uncovered elements and
add to the element at the point of intersection.
Machines
W X Y Z
A 0 1 1 4
B 0 0 0 1
Jobs C 0 0 0 3
D 9 4 0 0

Here N = 4 = n. Hence, we make an assignment.

Machines Machines
W X Y Z W X Y Z
A 0 1 1 4 A → W A 0 1 1 4 A → W
→ →
Jobs
Jobs

B 0 0 0 1 D Z o B 0 0 0 1 D Z
C 0 0 0 3 B → X r 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 has been assigned no job. Therefore optimum cost
= 18 + 13 + 19 = Rs.50

11.11 UNIT END QUESTIONS

1) What do you mean by transportation problems?
2) Explain the different solutions to transportation problems.
3) What are the various assumptions about transportation problems?
4) What are the different techniques available to get the initial feasible solution to the
transportation problem? Discuss in brief.
5) What is meant by an unbalanced transportation problem? How do you balance it?
6) Explain the procedure of getting a basic feasible solution by using VAM
7) Explain the meaning of assignment problems.
8) Explain briefly the procedure adopted in the assignment algorithm
9) Explain the difference and similarities between the Assignment problem and the
Transportation problem.
10) What do you mean by an unbalanced assignment problem? How do you balance it?

40
UNIT 12 INVENTORY CONTROL
Structure

12.0 Objectives
12.1 Introduction
12.2 Inventory costs
12.2.1 Ordering Cost or Setup cost
12.2.2 Carrying cost
12.2.3 Unit Purchasing Cost
12.2.4 Stock-out or Storage Cost
12.3 Types of Inventory
12.3.1 Raw Materials
12.3.2 Work-in-Process
12.3.3 Finished Goods
12.3.4 Other Types of Inventory
12.4 Economic Order Quantity (EOQ) Model
12.5 Fixed Order Quantity System (Q - System)
12.6 Periodic Review (P) System
12.7 Let Us Sum Up
12.8 Keywords
12.9 Suggested Further Readings / References
12.10 Answers to Check Your Progress
12.11 Unit End Questions

12.0 OBJECTIVES
After studying this unit, you should be able to:
 explain the concept of inventory management:
 discuss the types of inventory and inventory costs; and
 describe the various inventory management techniques such as Economic Order
Quantity (EOQ), Fixed Order Quantity System (Q – System), and Periodic Review
(P) System.

1
12.1 INTRODUCTION
Krishi Bhandar is a cooperative society carrying 150 different items. The society has
four departments- seeds, fertilisers, pesticides, and farm equipment. The society has an
annual turnover of Rs. 1 crore. Mr. Rakesh who is the store manager is a commerce
graduate working with this society for the last 10 years and is very concerned with the
increasing costs of maintaining the stock. He is curious to know if there are some
management techniques available so that he can reduce the stock levels and maintain
optimum levels so that he never faces stock-outs. He is also interested to know how he
can reduce the cost of ordering and procurement. Stocks in the above situations are
referred to as inventory in business settings.
Inventory management is the planning and controlling of inventories in order to meet
the competitive priorities of the organization. The objective of any organization is to
minimize the inventory costs and therefore requires the inventory manager to control
the movement of goods. Whenever stocks are received by a supplier the inventory gets
created and depletes on their disbursement. It is very important that optimal levels are
maintained so that there is no case of overstocking or stock-outs/shortages.
This Unit deals with the concepts in Inventory Management, which include
understanding cost structure, techniques of inventory management like EOQ, fixed
order quantity, and periodic review inventory system.

12.2 INVENTORY COSTS

If we want to proceed further on the topic of inventory management it is essential to
understand the costs associated with the inventory as the basic objective of inventory
management is to minimize the inventory costs. There are two basic inventory costs-
ordering costs and carrying (or holding) costs.

12.2.1 Ordering Cost or Setup Cost

These are basically the costs of getting an item into the inventory of the firm. They are
incurred each time an order is placed. Placing an order involves not just the cost of
stationery and clerical effort but also may require the tendering action, cost of
advertising tenders, shipment costs, costs of transit insurance, and so on. The costs are
calculated in terms of Rs. per order, that is incremental costs are calculated.

12.2.2 Carrying Cost

Also referred to as holding costs, are basically the cost incurred because the firm owns
or maintains the inventory. They include interest on the money invested in the
inventory, storage costs, costs of deterioration and obsolescence, security, losses and
pilferage, insurance, accounting, and so on.
These costs are normally expressed as a percentage of average inventory value or as a
cost per unit per time period. Expressing them as a percentage is convenient as the
same percentage can be applied irrespective of the unit cost of the item. If we assume
2
that the usage of the inventory follows a constant steady rate and if we start with a
certain inventory Q at the beginning of a period then at some stage the inventory will
become zero and need to be replenished. The ideal inventory system assumes that
inventory is used at a constant steady rate and is replenished to the original level
whenever it reaches zero. The various categories of costs normally incurred in business
situations are listed in Table 12.1 below.
Table 12.1: Categories of Cost
Cost (and range) as a
Category
Percent of Inventory Value
Housing costs (building rent or depreciation operating
6 % (3 – 10 %)
costs, taxes, insurance)
Material handling costs (equipment lease or
3% (1 - 3.5%)
depreciation, power, operating cost)
Labor cost 3% (3 – 5%)
Investment costs (borrowing costs, taxes, and
11% (6 – 24%)
insurance on inventory)
Pilferage, space, and obsolescence 3% (2 - 5%)
Overall carrying cost 26%
Source: Krajewski, L. J., Ritzman, L. P. and Malhotra, M. K. (2010), Operations
Management, Ninth Edition, Prentice-Hall, India Private Ltd.
Apart from the two inventory costs, it is also very important to understand the meaning
of unit cost and stock-out cost. The same is defined below.

12.2.3 Unit Purchasing Cost

It is defined as the variable cost associated with purchasing a single unit. For items
purchased from external vendors, it includes the purchase price along with the shipping
cost. For the units manufactured/produced in-house the Variable labor cost, Variable
overhead cost, and Raw material cost are included.

12.2.4 Stock-out or Shortage Cost

Any unfulfilled customer demand or demand not met on time leads to stock-out or
shortage. It is difficult to estimate the stock-out costs as it relies heavily on the concept
of opportunity costs.

Check Your Progress 12.1

1. Which of the following is not a reason for carrying inventory?

a) To maintain the independence of operations

b) To take advantage of economic purchase-order size

c) To make the system less productive

3
 d) To meet variation in product demand

e) To allow flexibility in production scheduling

2. Counting items to ensure an order is correct is an example of:

a) Setup cost

b) Ordering cost

c) Carrying cost

d) Stockout cost

e) Holding cost

3. Inventory cost structures incorporate four types of costs. Which of the following
is not one of those four?

a) Crash (expedite) cost

b) Item (acquisition) cost

c) Order (set up) cost

d) Carrying (holding) cost

e) Stock-out cost

4. An example of an ordering cost is

a) Freight costs.

b) Spoilage.

c) Purchase price.

d) Inspecting goods received.

12.3 TYPES OF INVENTORY

After knowing the different costs associated with the inventory, the next question is to
define the scope of inventory management. This includes answering the basic question
as to what are the different types of inventory. This section identifies three main types
of inventory. The same is discussed below.

12.3.1 Raw Materials

Goods, components, or items that go into the production process eg: gypsum,
limestone, coal, etc. are all procured from outside vendors that are required for
producing cement.

4
12.3.2 Work-in-Process
Work-in-process inventory items are those materials and parts that are semi-finished
and are waiting for a few more operations to be performed on them, to form a complete
product. Even finished goods that have not yet been packaged and inspected, as well as
raw materials that have moved from storage to a preassembly area will be referred to as
work-in-process items.

12.3.3 Finished Goods

Finished goods are any products that are ready to be moved to the market for the end
customer’s consumption. Finished goods may lie in a warehouse or on a shop floor.
For example, packaged cement bags in the despatch section or the warehouse is termed
as finished goods inventory.

12.3.4 Other Types of Inventory

Maintenance, Repair, and Operating (MRO) inventory are all the items an
organization needs in order to operate, such as office stationery, lubricating oils,
maintenance items, and tools to repair equipment. Machine lubricants, bearings, oil,
etc. form the MRO inventory.
In the literature we also find the following types of inventories mentioned;
Transit inventory: these are items in transit, such as items in trucks that are moving
from a warehouse to a factory
Buffer inventory: these are extra items maintained as part of inventory to take care
of external and internal demand variations that may arise due to supply problems,
such as poor quality or slow delivery of raw materials, etc.
Anticipation inventory: These are items that a company maintains as a part of
inventory to take care of excess demand. eg. In anticipation of Diwali shopping.
Pipeline inventory, also known as pipeline stock is used to refer to those goods that
have left the firm's warehouse but are still in the company's distribution chain as they
are yet to be bought by ultimate consumers. This concept is similar to work-in-progress
inventory where the product is still under production whereas in pipeline inventory the
finished good is still in the process of delivery. E.g. The inventory with a transport
company carrying the parcel goods for the client's destination which is still to be
delivered but has left the company’s warehouse is considered as Pipeline inventory.
Source: http://www.w-e.com.au/blog

Check Your Progress 12.2

Note:
1. Which of the following is NOT a type of inventory?
a) MRP

5
 b) Raw material
c) Finished goods
d) Work-in-process
2. Work-in-process inventory is devoted to maintenance, repair, and operations
(True/False)
3. Pipeline Inventory - (In transit) inventory represents products moving from
point to point in the materials flow system. This consists of orders that have
been placed but not yet received. Reasons: Time and distance, Work-In-Process
inventory (True/False)

12.4 ECONOMIC ORDER QUANTITY (EOQ) MODEL

The economic order-quantity model aims at minimizing the total inventory costs. EOQ
identifies two major components of inventory costs namely carrying costs and ordering
costs. The model is explained in detail in this section.
The important assumptions for the EOQ Model are as follows;
i. Demand is deterministic and occurs at a constant rate.
ii. If an order of any size (say, q units) is placed, and ordering cost S is incurred.
iii. The lead time for each order is zero.
iv. No shortages are allowed.
v. The cost per unit-year of holding inventory is H
The cycle inventory levels in an EOQ system as shown in Figure 12.1

Figure 12.1: Cycle Inventory levels in EOQ System.

6
The behavior of the total annual cycle inventory costs is presented in figure 12.2

Figure 12.2: Inventory Costs behaviour with the Order Quantity

Let
Q= Number of pieces per order
Q*= Optimal number of pieces per order (EOQ)
D= Annual demand in units for the inventory item
S= Setup or ordering cost for each order
H= Holding or carrying cost per unit per year
We know that;
Annual ordering cost (S) = (Number of orders placed per year)  (order cost per order)
= (Annual demand/Number of units in each order)  (order cost per order)
𝐷
= [𝑄 ] (𝑆)

Annual holding cost = (Average inventory level)  (Holding cost per unit per year)
= (Order Quantity/2)  (Holding cost per unit per year)
𝑄
= [ ] (𝐻)
2

It is observed from fig.12.2, that the Optimal order quantity is achieved when annual
setup cost equals annual holding cost
𝐷 𝑄
Therefore, [𝑄 ] (𝑆)= [ 2 ] (𝐻)

Solving for Q* we get,

2DS= Q2H
Q2 = 2DS/H

7
 Q* = √2𝐷𝑆/𝐻 ----------------------------------- (1)
Equation (1) is called the EOQ formula as the Quantity arrived using this formula leads
to optimal cost (minimized inventory cost).
Now let us use the derived formula and understand its use with help of an illustrative
example.

ILLUSTRATIVE EXAMPLE 1
Determine the optimal number of packaging bags to order. D= 1000 units, S= Rs 10 per
order, H= Rs 0.50 per unit per year.

Q* = √2𝐷𝑆/𝐻

2100010
Q*= √ = √40000 =200 units
0.50

Expected number of orders = N= Demand/Order quantity = 𝐷⁄𝑄 ∗

N= 1000/200 = 5 orders per year
Expected time between orders = T= No. of Working days per year/N
T= 250/5 = 50 days between orders
Total annual cost = Ordering cost + Holding cost
𝐷 𝑄
TC = S (𝑄 ) + H ( 2 )

1000 200
TC = 10 ( ) + 0.50 ( )
200 2

TC = Rs. 100.

ILLUSTRATIVE EXAMPLE 2
Farm bucket sales are 18 pieces per week, and the supplier charges Rs. 60 per unit. The
cost of placing an order (S) with the supplier is Rs. 45. Annual holding cost (H) is 25%
of a bucket’s value, based on operations 52 weeks per year. Management chose a 390-
unit lot size (Q) so that new orders could be placed less frequently.
a) What is the annual cycle-inventory cost (C) of the current policy of using a 390-
unit lot size?
b) What would be the effect if the management decides to increase the lot size to
468 units?
c) What is the least annual inventory cost?
Solution:
(a) D = (18 /week)(52 weeks) = 936 units, H = 0.25 (Rs. 60/unit) = Rs. 15
𝑄 𝐷 390 936
C = ( 2 ) (H) + (𝑄 ) (S) = ( ) (15) + (390 ) (45)
2

8
 The annual cycle inventory cost C = Rs. 2925 + Rs. 108 = Rs. 3033
(b) Now if the management decides to increase the lot size of the order to 468 units
that what would be the annual cycle-inventory cost (C)
𝑄 𝐷 468 936
C = ( 2 ) (H) + (𝑄 ) (S) = ( ) (15) + (468) (45)
2

C = Rs. 3510 + Rs. 90 = Rs. 3600

It is seen that the annual cost increases to Rs. 3600 with the increase in the order
size.
Then, what is the best order size that will minimize the total annual inventory costs?
(c) The best lot size (EOQ) is the lowest point on the total annual cost curve which
is calculated by the equation.

Q* = √2𝐷𝑆/𝐻

293645
Q* = √ 15
= 74.94 or 75 units.

𝑄 𝐷 75 936
C = ( 2 ) (H) + (𝑄 ) (S) = ( 2 ) (15) + ( 75 ) (45)

C = Rs. 1124.10
We see that the EOQ of 75 units gives us the least annual inventory costs.

Check Your Progress 12.3

1. The EOQ determines:

a) The point at which a company should reorder

b) The point at which carrying costs equal ordering costs

c) The point at which the sum of carrying costs and ordering costs is
maximized

d) The relevant inventory flow for a particular time period

e) None of the above

2. Which of the following is not an assumption of the basic economic-order

quantity model?

a) Annual demand is constant and known

b) Lead time is constant

c) Ordering or setup costs are constant

9
 d) Quantity discounts are available

e) No backorders are allowed

3. The EOQ is defined by the intersection of the Ordering Cost Curve and the
Total Cost Curve (True/False)

4. In the basic EOQ model, the two main relevant costs are the ____ and the ____.

a) Order cost; Purchase cost

b) Order cost; Storage cost

c) Order cost; Holding cost

d) Safety stock cost; The lateness penalty cost

e) Purchase cost; The discount cost

5. __________ is the amount of inventory ordered that will minimize the total
inventory cost. (Options: EOQ, LOQ, BOQ, ERP, JIT)

6. In an economic order quantity problem, the total annual cost curve is at its
__________ where holding costs equal setup costs. (Options: Minimum,
Maximum, Moderate)

7. A company wishes to determine the EOQ for an item that has an annual demand
of 2,000 units, a cost per order of Rs. 75, and an annual carrying cost of Rs.
7.50 per unit. What is the EOQ?

a) 73

b) 40,000 units

c) 200 units

d) 100

8. If EOQ = 40 units, order costs are Rs. 2 per order, and carrying costs are Re .20
per unit, what is the usage in units?

a) 10 units.

b) 16 units.

c) 40 units.

d) 80 units.

10
12.5 FIXED ORDER QUANTITY SYSTEM (Q – SYSTEM)
Unlike the EOQ model discussed in the previous section wherein the orders are fixed at
definite intervals, in this system, a reorder is made when the inventory level falls to a
reorder level, which is equal to the lead time requirements plus buffer stocks. The
procurement and consumption cycle is shown in Fig. 12.3

Figure 12.3: Fixed Order Quantity System

It is observed from the above figure that a supply equal to EOQ is received at point A
and the quantity in stock reaches point E. The items are then issued and at time F when
the stock reaches ROL, an order is placed for quantity Q=EOQ and the issues continue.
At point B the supplies of order placed at point F are received and the stock reaches G.
At G, there is a delay in receiving the supplies due to lead time and so we issue from
the buffer stock. Thus for this system of ordering we, fix the size of the order that is,
every time the same quantity is ordered, but the time of ordering is allowed to vary
depending on the actual demand/usage.
In this system, we decide on the following three parameters

i. EOQ (Q*) = √2𝐷𝑆/𝐻

ii. Safety Stock = (Maximum lead time – Normal lead time) x consumption
iii. Reorder Level (ROL) = Safety Stock + Normal lead time consumption
Let us consider the following illustrative example to understand the above situation.
ILLUSTRATIVE EXAMPLE 1
A welding rod that is used in a fabrication shop has the following details:
Annual Demand= 1000 units/year
Holding costs = 0.30
Unit price = Re 0.50 per unit

11
Ordering Costs = Rs. 10 per order
Lead time = 2 years
Now let us calculate the EOQ and the Reorder level for this item.
We know that the formula for calculating the EOQ is as follows;

Q* = √2𝐷𝑆/𝐻

We get, EOQ = √2101000/(0.300.50) = 365 units

Further, time to order To = Q*/D = 365/1000 = 0.36 years
Since the lead time is 2 years and the time for placing the order is estimated as 0.36
years, reordering occurs when the level of inventory is sufficient to satisfy the demand
for (2-0.36) years = 1.64 years. Thus the optimum quantity Q* = 365 units is ordered
when the order of inventory reaches 1.64 x 1000 units. Therefore, the reorder point is
1640 units.

ILLUSTRATIVE EXAMPLE 2
For a fixed order quantity system find out
(i) EOQ
(ii) Safety stock
(iii) Reorder level with the following data
D= 10000 units, Unit cost = Re 1/unit, Carrying cost = Re 0.24 per unit,
Ordering Cost= Rs. 12/order
The normal lead time is observed to be 15 days. However, the lead time
may go as high as 30 days.
Historical Lead times: 15 days, 25 days, 13 days, 14 days, 30 days, 17 days.
Solution:

(i) Q* = √2𝐷𝑆/𝐻 = √2  12  10000/(0.24) = 1000 units

(ii) Safety Stock = (Maximum lead time – Normal lead time) x
consumption
Maximum lead time = 30 days
Normal lead time = 15 days
(30−15) 10000
Safety Stock = ( )
30 12
= 416.66 ≈ 417 units.
(iii) Normal lead time consumption = normal lead time  monthly
consumption

12
 15 10000
= ( ) ( ) = 417 units
30 12
Therefore the reorder level = Safety Stock + Normal lead time consumption
= 417 +417 = 834 units

(iv) Maximum inventory level = Safety stock + EOQ

= 417 +1000 = 1417 units
Since inventory would fluctuate from a maximum of 1417 units to a
minimum of 417 units the average inventory
1
= (2) (Safety Stock + Maximum inventory)
1
= (2) (417 + 1417) = 917 𝑢𝑛𝑖𝑡𝑠.

Check Your Progress 12.4

1. Which of the following statements is true about the Fixed-order quantity

system?

a) It does not have a reorder point but rather a target inventory.

b) The system is completely determined by the two parameters, Q and R.

c) It does not have an EOQ, since the quantity varies according to demand.

d) The order interval is fixed not the order quantity.

e) None of the above is true.

2. All of the following are possible reasons for using the fixed order interval
model except:

a) Supplier policy encourages use.

b) Grouping orders can save on shipping costs.

c) The required safety stock is lower than with an EOQ/ROP model.

d) It is suited to periodic checks of inventory levels rather than continuous

monitoring.

e) Continuous monitoring is not practical.

3. The fixed order interval model would be most likely to be used for this
situation:

13
 a) A company has switched from mass production to lean production.

b) Production is done in batches.

c) Spare parts are ordered when a new machine is purchased.

d) Grouping orders can save shipping costs.

e) none of these

4. The fixed-order interval model requires continuous monitoring of inventory

levels (True/False).

5. In a fixed order quantity system, the order size stays constant although the time
interval between orders may fluctuate (True/False)

6. An electrical appliances store has the following details available with him for
the product.

D= 10000 units, Unit cost = Re 1/unit, Carrying cost = Re 0.24 per unit,
Ordering Cost= Rs. 12/order. The normal lead time is observed to be 15 days.
However, the lead time may go as high as 30 days.

Past Lead times: 15 days, 25 days, 13 days, 14 days, 30 days, and 17 days.

Calculate:

a. EOQ

b. Safety stock

c. Reorder level

12.6 PERIODIC REVIEW (P) SYSTEM

As compared to a fixed order quantity system that we discussed in the previous section,
in a periodic review system an item’s inventory position is reviewed periodically- once,
weekly, monthly, or half-yearly rather than continuously. The other main features are:
It is also referred as
– P (period) is the time period after which the reviews are performed and a new
order is placed.
– Demand is not constant and certain
– The lot size, Q, may change from one order to the next. As the maximum order
to be placed should not go beyond the target level (T).
This system is also referred to as a fixed interval reorder system or a periodic reorder
system.

14
The following figure 12.4 gives the way in which the periodic review system operates.

Figure 12.4: Periodic Review System

In the above figure, the replenishment level is fixed at 100 units and the time period for
the review is taken as one month. We can see that the actual inventory level for the
month of January was 35 units and therefore a quantity of Q = 65 units was ordered.
Similarly, the reviews were carried out in the months of February, March, and April
and the quantities of 50 units, 35 units, and 55 units were ordered respectively.
First, we calculate the value of replenishment level (T) by considering the average rate
of daily consumption during the review period, normal lead time, and the safety stock.
The following formula may be used:
Replenishment level (T) = Average rate of consumption  (Review period + Normal
lead time) + Safety Stock
Further, the Ordering quantity (Q) = Replenishment Level (T) – Stock available (I) as
observed in the above example.
Let us consider the following illustration example:

ILLUSTRATIVE EXAMPLE 1
A Government Dispensary in a village gives Hepatises B vaccination. Every 10 days,
the vaccine supply authority comes by to check the inventory levels and order more of
the vaccine. It takes 3 days to get an order. The no of vaccines given per day is about
20 doses but can vary. The dispensary would like to keep a safety stock of about 30
doses to protect against stock-outs, just in case demand levels or lead times are greater
than expected.
For the above situation if we want to implement the periodic review system then the
first step would be to decide the replenishment level (T) as follows:
Replenishment level (T) = Average rate of consumption  (Review period + Normal
lead time) + Safety Stock

15
The average rate of consumption is 20 doses per day
The review period is every 10 days
The lead time is about 3 days
Safety stock to be maintained is 30 doses
Substituting the values in the above equation gives us:
T= 20  (10 + 3) +30 = 300 doses
The Replenishment level to be maintained is 300 doses.

Check Your Progress 12.5

1. Periodic review means:

a. Ordering inventory at a predetermined re-order level

b. Ordering inventory at a fixed and regular time interval

c. Ordering inventory to supplier call-offs on an as-required basis

d. Ordering inventory in consultation with suppliers relative to their available

capacity

e. Ordering inventory when it falls below the designated safety stock level

2. A periodic system of inventory

a. Reduces record keeping.

b. Increases record keeping.

c. Increases the cost of maintaining inventory.

d. Eliminates the need for a physical count.

3. A __________ system triggers inventory ordering on a uniform time-frequency.

4. A seeds vendor sells on average 50 packets (in 1kg pack) of marigold flower
seeds which varies depending upon the customer's needs. He checks the
inventory every 15 days and the order lead time is 5 days. Presently the safety
stock maintained by him is 15 kgs. Calculate the replenishment level.

12.7 LET US SUM UP

Inventory serves a useful purpose in the organization. That said, firms can help
minimize the need for inventory by carefully managing those factors that drive
inventory levels up. The major costs of inventory are the carrying costs and the

16
ordering costs. The inventory concepts may be applied to different types of inventory
that includes: raw materials, work in process, finished goods, etc. The economic order
quantity (EOQ) is the order quantity that minimizes total inventory costs. The fixed
order quantity helps us to minimize inventory costs by taking into consideration the
safety stocks and the variation in the demand. The ordering quantity is a constant value.
The periodic review system requires reviewing the inventory on a periodic basis (fixed
intervals) with varying order sizes.

12.8 KEYWORDS
Economic Order Quantity : EOQ is the order quantity that minimizes total
inventory holding costs and ordering costs.
Fixed Order Quantity : A method that only allows for a specific amount
of a given item to be ordered at one time.
Holding Costs : The associated price of storing inventory or assets
that remain unsold.
Inventory Cost : Inventory Costs are the costs related to storing
and maintaining its inventory over a certain
period of time.
Inventory : The raw materials, work-in-process goods, and
completely finished goods that are considered to
be the portion of a business's assets that are ready
or will be ready for sale.
Periodic Review System : A periodic inventory review system is one where
inventory is checked and reordered at a set time
interval.
Setup Cost : Expenses incurred in setting up a machine, work
center, or assembly line, to switch from one
production job to the next.

12.9 SUGGESTED FURTHER READINGS / REFERENCES

1. Chase, R. B., Shankar, R., Jacobs, F. R. and Aquilano, N. J. (2010), Operations
and Supply Management, Twelfth Edition, Tata McGraw-Hill Publishing
Company Ltd.
2. Gaither, N. and Frazier, G. (2004), Operations Management, Ninth Edition,
Cengage Learning India Private Ltd.
3. Heizer, J., Render, B. and Rajashekhar, J. (2008), Operations Management,
Ninth Edition, Prentice-Hall, India Private Ltd.

17
 4. Hira, D. S. and Gupta. P.K. (2007), Operation Research, S Chand and
Company, New Delhi.
5. Krajewski, L. J., Ritzman, L. P. and Malhotra, M. K. (2010), Operations
Management, Ninth Edition, Prentice-Hall, India Private Ltd.
6. Mahadevan, B. (2010), Operations Management – Theory and Practice, Second
Edition, Pearson Education Limited.
7. Markland, R. E., Vickery, S. K. and Davis, R. A. (1995), Operations
Management – Concept in Manufacturing and Services, West Publishing
Company.
8. Monks, J. G. (1987), Operations Management – Theory and Problems, Third
Edition, Tata McGraw-Hill Publishing Company Ltd.
9. Nigel Slack, Stuart Chambers and Robert Johnston (2010), Operations
Management, Sixth Edition, Pearson Education Limited.
10. Russell, R. S. and Taylor III, B. W. (2010), Operations Management – Along
the Supply Chain, Sixth Edition, John Wiley and Sons. Inc.
11. Schroeder, R. G., Goldstein, S. M. and Rungtusanatham, M. J. (2011),
Operations Management: Contemporary Concepts and Cases, Fifth Edition,
Tata McGraw-Hill Publishing Company Ltd.
12. Stevenson, W.J. (2005), Operations Management, Eighth Edition, Tata
McGraw-Hill Publishing Company Ltd.

12.10 ANSWERS TO CHECK YOUR PROGRESS

Check Your Progress 12.1
1. c
2. b
3. c
4. d

Check Your Progress 12.2

1. a
2. False
3. True

18
Check Your Progress 12.3
1. b
2. d
3. False
4. b
5. EOQ
6. Minimum
7. c
8. d

Check Your Progress 12.4

1. b
2. c
3. d
4. False
5. True
6. Solution:

a. Q* = √2𝐷𝑆/𝐻 = √21210000/(0.24) = 1000 units

b. Safety Stock = (Maximum lead time – Normal lead time)  consumption
Maximum lead time = 30 days
Normal lead time = 15 days
(30−15) 10000
Safety Stock = ( )
30 12
= 416.66 ≈ 417 units.
Normal lead time consumption = normal lead time  monthly consumption
15 10000
= ( ) ( ) = 417 units
30 12
c. Reorder level = Safety Stock + Normal lead time consumption
= 417 + 417 = 834 units
Maximum inventory level = Safety stock + EOQ
= 417 +1000 = 1417 units
Since inventory would fluctuate from a maximum of 1417 units to a
minimum of 417 units the average inventory

19
 1
= (2) (Safety Stock + Maximum inventory)
1
= (2) (417 + 1417) = 917 𝑢𝑛𝑖𝑡𝑠.

Check Your Progress 12.5

1. b
2. a
3. Periodic Review
4. Solution:
The review period is every 15 days
The lead time is about 5 days
Safety stock maintained is 15 doses
Substituting the values in the above equation gives us:
T= 50  (15 + 5) +15 = 1015 packets
The Replenishment level to be maintained is 1015 packets.

12.11 UNIT END QUESTIONS

1. Define Inventory management and discuss the various costs associated with it.
2. Define the terms
a. Pipeline
b. Cycle
c. Work in Process
d. Anticipatory
3. Discuss the major types of Inventory.
4. What is the purpose of the basic EOQ model?
5. What are the different assumptions of EOQ?
6. Discuss the Periodic Review System of Inventory Management.
7. Differentiate between Fixed Order Quantity and Periodic Review System of
Inventory Control.
8. Explain the fixed order quantity system. What parameters are needed to be
considered?

20
UNIT 13 GAME THEORY AND NETWORK ANALYSIS
Structure

13.0 Objectives
13.1 Introduction
13.2 Assumption and Basic Terminologies
14.3 Two Person Zero Sum Games
13.4 Solution of Games by Dominance
13.5 Programme Evaluation and Review Technique (PERT) & Critical Path Method
(CPM)
13.5.1 Critical Path and Project Management
13.6 Let Us Sum Up
13.7 Keywords
13.8 Suggested Further Readings / References
13.9 Answers to Check Your Progress
13.10 Practice Problems
13.11 Unit End Question

13.0 OBJECTIVES
After studying this unit, you should be able to:
 explain the concept and applicability of game theory;
 discuss the assumption and basic terminologies;
 explain the Two Person Zero Sum Games;
 describe the solution of games by Dominance; and
 apply Programme Evaluation and Review Technique (PERT) & Critical Path
Method (CPM) in agribusiness projects.

13.1 INTRODUCTION
Two competing packaged wheat flour brands namely Shaktiman atta and Ayush atta
are trying very hard to increase their market share and satisfy customer needs. In their
strategy to attract more customers, Shaktiman and Ayush atta have planned to launch
new products which have added nutritional value. They have different options to launch
their product and communicate to their customers like radio advertisement, TV

1
advertisement, print media, etc. However, Ayush atta is faced with budgeting
constraints and can afford to have limited advertisement expenditure. Keeping in mind
the actions of its competitor brand Shaktiman atta, what should be the strategy of
Ayush atta? Should it advertise heavily, moderately, or maintain a low profile? What if
Shaktiman atta advertises heavily? Will it affect the prospects of Ayush atta?
In the above case, the outcomes for Ayush atta are not only dependent on its own action
plan but also influenced by what the competitor does. Game theory or the theory of
games provides insight into such situations. It is a quantitative approach whose aim
would be to manage such conflicts and ascertain what each one can expect to gain in
view of the opposing interests.
In this Unit, we will discuss the basic terminology of Game theory, how to solve games
involving two people and understand the Rule of dominance while solving games.

13.2 ASSUMPTION AND BASIC TERMINOLOGIES

To begin with, let us first understand the assumption laid down for solving a game.
1. There are finite numbers of participants called players. Players have opposing
interests or the same objectives.
2. Each player has a finite number of strategies available to him and they act
rationally and intelligently.
3. Each game results in an outcome.
To understand the above assumptions clearly and play the game we will be required to
understand some basic terminologies. The commonly used terms are
1. Person: It refers to one of the opposing interests of players. It could be an
individual, a group of individuals, or an organization.
2. Game: The set of rules that defines what can and cannot be done; No change in
the rules is permitted during the game.
3. Strategy: A player’s course of action that is complete and ready before the
commencement of the game.
4. Value of the game: Expected payoff when each player is using his/her optimal
strategy.
5. Two person zero sum game: A game with only two persons is said to be a two-
person zero-sum game if the gain of one player is equal to the loss of the other.
Games are categorized based on the number of players involved in the game, that is if
we have two players it is referred to as a two-person game if three players then three-
game, and so on. The situation becomes complex to solve with the increase in the
number of persons therefore we shall restrict our scope of discussions in this unit to
only two-person sum games.

2
Check Your Progress 13.1
[

Note:

1. Game theory:

a. Analyses strategic interaction among rivals.

b. Examines pricing behavior of organizations in perfectly-competitive

industries.

c. Examines output decisions of rival firms in all industries.

d. Used for examining moves and counter-moves of players in sequential-

move games.

e. None of the above.

2. Game theory is about:

a. Strategies to develop market structure in the entertainment industry

b. Strategic interaction among players in the same market

c. Notions with no application in the real world.

d. An unnecessary exercise since all industries reach equilibrium solely on the

supply and demand analysis

3. Game theory is a complete and integrated theory of strategy based on

mathematical analysis.

a. True

b. False

4. Most games assume that there are only 2 players.

a. True

b. False

13.3 TWO PERSON ZERO SUM GAMES

Let us consider a simple two players game with players A and B. We further assume
that each of these players has two strategies available to them with the possible
outcomes in form of payoff given below in Table 13.1.

3
Table 13.1: Payoff Table
Value in Rs
Player A Player B
Strategy I Strategy II
Strategy I 5 lakhs 2 lakhs
Strategy II 7 lakhs 3 lakhs
Now let’s understand how to read this matrix:
1. Any payoff matrix is given from the viewpoints of the player listed on the left
side of the matrix (Rows). In this example, it is player A.
2. For player A, Negative payoffs represent loss and positive payoffs profits, and
gains. In the above matrix player A losses Rs. 5 lakhs if he plays strategy I and
player B also losses Rs. 2 lakhs if player B also plays strategy I.
3. Player A will gain Rs. 3 lakhs if he plays strategy II. Player A will gain Rs. 7
lakhs if he plays strategy II and player B decides to play strategy I.
With these observations, the next step is to understand what should be the strategy of
players A and B. It is clear from the given matrix that the above game is more
favourable for player A and player A will always choose strategy II as he will gain
profit. Whereas player B will choose strategy II so that player A will gain a minimum.
Hence the following generalized rules may be adopted.
1. Player A will use the maximum principle, i.e. will select the strategy which
gives him the maximum of the minimum gains.
2. Player B will use the minimax principle, i.e. will select the strategy which gives
him the minimum of the maximum losses.
Using the above two principles for players A and B we decide the following actions
Player A:
Minimum gains with Strategy I: Rs. - 5 lakhs
Minimum gains with Strategy II: Rs. 3 lakhs
The maximum of these two values is Rs. 3 lakhs, so player A will select strategy II.
Player B:
Maximum gains for Strategy I (For player A): Rs. 7 lakhs
Maximum gains for Strategy II (For player A): 3 lakhs
The minimum of these two strategies is strategy II with a minimum gain of Rs. 3 lakhs
for player A. Therefore player B will select strategy II.
To summarise the problem the following are the observations
o Player A plays strategy II
o Player B plays strategy II

4
The game value is Rs. 3 lakhs.
In the discussion, we had for the above problem we found that players A and B will
play a single strategy. In such cases, it is said to have a pure strategy. The game value
in such cases (we got Rs. 3 lakhs) is called a Saddle point. However, it is not necessary
that every problem should have a pure strategy or a saddle point. If these conditions are
not satisfied we say that the problem has a mixed strategy.
In conclusion, we say that if the maximin value for the row player and minimax value
for the column player is the same, then there exists a saddle point representing the
game value and the corresponding row and column the players' strategy.
For the payoff matrix given in Table 13.2 let us test whether the game has a saddle
point or not.
Table 13.2: Payoff Table in (Rs)
Player B Maximin
I II III
I 500 400 280 280
Player

II 700 500 450 450

A

III 750 550 500 500

Minimax 750 550 500
Examining the values of maximin for player A and minimax values for player B, we
infer that there exists a saddle point with game value Rs. 500, and the following pure
strategies are obtained.
o Player A – Strategy III
o Player B - Strategy III

Check Your Progress 13.2

Note:

1. In a zero-sum game,

a. One player wins, and the other loses.

b. The sum of each player’s winnings if the game is played many times must
be zero.

c. The game is fair—each person has an equal chance of winning.

d. Long-run profits must be zero.

2. In a zero-sum game

a. The sum of the payoffs for any given strategy pair is zero.

5
 b. The gain of one of the players is equal to the loss of the other player.

c. It is impossible for both players to earn positive profits for a given strategy
pair.

d. All of the above is true for zero-sum games.

3. The saddle point in a payoff matrix is always the __________.

a. largest number in the matrix

b. smallest number in its column and the smallest number in its row

c. smallest number in the matrix

d. largest number in its column and the smallest number in its row

e. none of the above

4. Consider the following two-person game, and determine the saddle point if it
exists.

Y1 Y2
X1 4 6
X2 5 -3
a. X2Y2

b. X1Y2

c. X1Y1

d. X1Y2

e. There is no saddle point.

13.4 SOLUTION OF GAMES BY DOMINANCE

Now let us consider one more example and solve it for obtaining the game value and
player strategies.
Two packaged fruit drink manufacturers A and B have the objective of increasing their
market share. The following strategies are available to both firms:
1. Launch guava flavour fruit drink (A1, B1)
2. Introduce a tetra pack of 250 ml (A2, B2)
3. Intensify the print advertisement (A3, B3)
4. Maintain present strategy (A4, B4)

6
The following payoff matrix in Table 13.3 shows the increase in the market share for
company A.
Table 13.3: Payoff table in (Rs)
Company B Maximin
B1 B2 B3 B4

Company A
A1 30 60 20 0 0
A2 25 15 10 0 0
A3 35 45 0 5 5
A4 50 55 8 22 8
Minimax 50 60 20 22

We find that the saddle point for the above matrix does not exist and hence we cannot
have a pure strategy for players A and B. In such cases we are required to reduce the
problem to a 2  2 matrix and to do so we shall use the rule of dominance.
The general rule for dominance
1. If all elements of a row, say Kth is less than or equal to the corresponding
elements of any other row, say rth, the Kth row is dominated by the rth row.
2. If all the elements of a column say Kth are greater than or equal to the
corresponding elements of any other column say r th, then Kth column is
dominated by rth column.
3. Delete/Remove dominated rows or columns.
Now let’s apply these rules of dominance to the game payoff matrix given in Table
13.4
Table 13.4: Payoff Table
Company B
B1 B2 B3 B4
Company A

A1 30 60 20 0
A2 25 15 10 0
A3 35 45 0 5
A4 50 55 8 22
Let us examine the rows for the rule of dominance. We find that the values of row A2
are less than or equal to values of A1 therefore A2 is getting dominated, so we reduce
row A2. Similarly, row A3 gets dominated by row A4 and is reduced. The modified
payoff matrix is:

7
Table 13.5: Payoff Table
Company B
B1 B2 B3 B4

Company
A1 30 60 20 0

A
A4 50 55 8 22
We check the rule of dominance for columns. We find that column B 2 is getting
dominated by B1 (i.e. B2 values are greater than B1 values). Further B3 and B4 dominate
column B1. Hence the reduced matrix is:
Table 13.6: Payoff Table
Company B
B3 B4
Company A

A1 20 0

A4 8 22

We can now see that there is no further dominance. The modified matrix has no saddle
point.
Table 13.7: Payoff Table
Company B
B3 B4
Company A

A1 20 0

A4 8 22

As there is no pure strategy to obtain the optimum mixed strategy we use the following
shortcut method.
Step 1: Subtract the smaller payoff in each row from the larger payoff and the smaller
payoff in each column from the larger payoff.
Table 13.8: Payoff Table
Company B
B3 B4
Company A

A1 20 0 20 – 0 = 20

A4 8 22 22 – 8 = 14

20 – 8 = 12 22 – 0 = 22

Step 2: Interchange each of these pairs of subtracted numbers

8
Table 13.9: Payoff Table
Company B
B3 B4

Company A
A1 20 0 20 14

A4 8 22 14 20

12 22
22 12
Step 3: Divide each pair of numbers by the sum of the pair
Table 13.10: Payoff Table
Company B
B3 B4
Company A

A1 20 0 14 / 34

A4 8 22 20 / 34

22 12
34 34
Based on the above calculations we get the following
Strategies for Company A: (14 / 34, 0, 0, 20 / 34)
Strategies for Company B: (0, 0, 22 / 34, 12 / 34)
Value of Game = [20  22 / 34 ] [0  12 / 34]
= 12.94
Company A will gain a market share of 12.94 %.

Check Your Progress 13.3

Note:

1. Dominance decreases the size of games by enabling some strategies to be

ignored.

a. True

b. False

2. When there is dominance in a game then

a. Least of the row  highest of another row

b. Least of the row  highest of another row

9
 c. Every element of the row  corresponding element of another row

d. Every element of the row  corresponding element of another row

−1 −1
3. The Column [−2] dominates the column [ 1 ]
−3 0
a. True

b. False

4. If a game has no saddle points, it may still be possible to reduce the game to a 1
 1 game using dominance.

a. True

b. False

5. Given the following two-person game, which strategy can be eliminated by the
use of dominance?

Y1 Y2
X1 9 13
X2 12 8
X3 6 4
a. X1

b. X2

c. X3

d. None of the above

Check Your Progress 13.4

1. A game is said to be _________ if the lower and upper values of the game are
the same as well as zero.
2. If a game involves more than two players, then it is called a _________ game.
3. The rules of _________ are used to reduce the size of the payoff matrix.
4. The __________ method is useful for the game with a larger payoff matrix
without saddle point.
5. The __________ method is useful for a game where the payoff matrix is of size
2 × n or m × 2.

10
13.5 PROGRAMME EVALUATION AND REVIEW TECHNIQUE
(PERT) & CRITICAL PATH METHOD (CPM)
PERT deals with the problem of uncertainty in activity time. As a manager’s tool, it
defines and coordinates what must be done to successfully accomplish the objectives of
a project on time. It aids the decision-maker but does not make decisions for him. In
PERT, time is the basic measure. It is usually expressed in calendar weeks. The time
associated with the activity is probabilistic and the expected times of each activity are
determined from three-time estimates namely optimistic, pessimistic, and most likely.
Whereas in CPM, a simple estimate is developed and the time estimate for an activity is
deterministic.
Advantages:
 PERT provides the management the ability to plan the best possible use of
resources to achieve a given goal within the overall time and cost limitations.
 It helps management to handle the uncertainties involved in the programme.
 It processes the right action at the right point and at right time in the organization.
 It provides information on the existence of a slack period between activities and
what activities are crucial to meet the completion date.
 PERT allows a large amount of data to be presented in a well-organized diagram
(network) from which both the executor and customer can make joint decisions.
Limitations:
 The basic difficulty comes in the way of time estimates for the completion of
activities because activities considered are of a non-repetitive type.
 The technique does not consider the resources required at various stages of the
project.
Network fundamentals
Networks are composed of activities and events as defined below:
Activity
Activity is defined as a process or operation of a job that consumes time and resources.
In a network diagram, it is represented by an arrow ( ).
Event
An event is defined as the starting or ending point of an activity or a group of activities.
It represents a milestone and does not consume time or resources. An event is described
by a circle (O) in the network.
To draw a network diagram interdependencies between events and activities must be
identified so that a master plan can be developed which illustrates the up-to-date picture
of operations during the project and which can also be easily understood by all.

11
Network analysis provides valuable information for planning, scheduling, and
controlling large complex programmes. It helps to eliminate the need for crisis
management by providing a graphic representation of the total programme from which
the decisions such as the impact of late starts, early starts, etc., can be taken.
Rules for network construction
1. Each activity must have a preceding and a succeeding event

1 2

2. Each event should have a distinct (unique) number

3. There should not be a loop in the project network as shown below:

1 3

4. Note more than one activity can have the same preceding and succeeding event

1 2

5. Any number of activities may merge with an event

Dummy activity
A dummy activity also known as “Zero-time activity” is necessary to reflect the
dependency relationship between activities in the network, and is needed to satisfy one
of the rules for constructing of network i.e., “no two events can be directly connected
by more than one activity”. It is generally denoted by X 1, X2 or Z1, Z2 ….. etc. in the
network with dotted lines (………) as shown below.

12
 3 D
5

A
Z1 G

1 2
B 7

Z2
F
C
4 6
E
In the above network, B is a common immediate activity predecessor of both D and E
activities, while activity A is an immediate predecessor of activity D alone, and activity
C is the immediate predecessor of activity E. Hence, two dummy activities Z1 and Z2
are used to indicate the precedence relationship as shown by broken lines in the above
network.
Steps in the development of network
1. Enumerate the activities in the project, which you feel need close monitoring in
terms of completion of the job intimate.
2. Define each activity
 Preceding activity (the activity which precedes it)
 Succeeding activity (the activity which followed)
 Concurrent activity (the activity that can be done concurrently)
3. Expected time for the completion of each activity
The estimation of the expected time for the completion of each activity is important in
the network analysis. This can be done considering three possible completion time
assumptions.
I) Optimistic time (a):
This time assumes that everything will go according to plan and with a minimum
amount of difficulties. This may be visualized approximately one percent of the time.
II) Most pessimistic time (b):
This time assumes that everything will not go according to plan and that the maximum
potential difficulties may develop. This type of situation may be expected
approximately one percent of the time.
III) Most likely or normal time (m):
This is the time that, in the mind of the manager, would most often occur should this
effort be repeated over time.

13
Considering the above, three types of lines, the expected time (t e) of the completion of
an activity is:
(𝑎 + 4𝑚 + 𝑏)
𝑡𝑒 =
6
where,
te = expected time
a = most optimistic time
m = normal time
b = most pessimistic time
and the associated degree of error also known as s.d. is calculated from the following
formula:
(𝑏 − 𝑎 )
𝑠. 𝑑. (𝑡𝑒 ) =
6
The s.d. of the expected time indicates the confidence of the manager in the completion
of the activity within the prescribed time. The larger the s.d. indicate less confidence in
the manager on the completion of the activity within the expected time.
Consider the following two examples:
Example 1:
a = 3, b = 7, m = 5 weeks
Then,
(3 + 4 × 5 + 7) 30
𝑡𝑒= = =5
6 6
(7 − 3)
𝑠. 𝑑. (𝑡𝑒 ) = = 0.67
6

Example 2:
a = 2, b = 12, m = 4 weeks
Then,
(2 + 4 × 4 + 12) 30
𝑡𝑒= = =5
6 6
(12−2)
𝑠. 𝑑. (𝑡𝑒 ) = = 1.67
6

In both the examples the expected time is 5 weeks but in example 2 the larger s.d.
indicates a higher degree of uncertainty because of the wider spread between the
optimistic and pessimistic time. Therefore care must be taken in the evaluation of risks
in the expected time.

14
The time estimates (weeks) for various activities in a dummy network are illustrated
below:

Name of the activity Symbol A M B Te Sd

1 A 9 12 21 13 2
2 B 6 12 18 12 2
3 C 2 4 6 4 0.67
4 D 4 8.5 10 8 1
5 E 6 8 10 8 0.67
6 F 1 2 3 2 0.33
7 G 2 4 6 4 0.67

Example:
Draw the flow chart and network for the following activities in a project.
Activity Table
Name of the activity Symbol Preceding activity te (weeks)
1 A -- 13
2 B -- 12
3 C A 2
4 D B, C 8
5 E A 15
6 F D 2

FLOW CHART
In a flow chart, the activity and its duration are shown in a box. The boxes are
connected with lines according to the preceding and succeeding activity relationship.
The critical path for the project can be identified by comparing the various path lengths
(sum of activity time on the path). The longest path in the chart is the critical path. The
flow charts do not give many details like the start and completion time of each activity.

15
 START

A 13 B 12

C 2

E 15 D 8

F 2

FINISH
In the above flow diagram, there are three alternate paths to reach from “Start” to
“Finish”. These are as under:
Sl. No. Path Path length Total
1 B-D-F 12+8+2 22
2 A-C-D-F 13+2+8+2 25
3 A-E 13+15 28 (Maximum)

Path A-E is the longest path i.e., having the maximum time is the critical path, and
activities A and E falling in this path are critical activities. Even though it is possible to
identify the critical path and the critical activities from a flow diagram, this does not
facilitate the computation of slacks or floats on the events. On the other hand, the
analysis of the network facilitates both identifications of critical paths and the
computation of slacks/floats.

16
NETWORK

2
13 15
A E
5
1
C 2 2

F
12
4
8
B 3
D

From the network, additional information on the time schedule of each event like Early
Starting Time (EST) and Late Starting Time (LST) can be worked out. This
information helps the Project Manager to reschedule his activity times and reallocation
of resources effectively.

Earliest Starting Time (EST)

EST is the time at which the activity cannot be started earlier than this time (Waiting
time).
In the above network, event 4 occurs when activities (A(1-2), B(1-3), C(2-4), and D(3-
4) are completed. In other words, event 4 occurs when paths (1-2-3-4) and (1-3-4) are
completed and the requirement of expected time for the two paths is 23 and 20 weeks,
respectively. In general, the EST of an event is the duration of the longest path among
the paths. In this case, 1-2-3-4 is the path with the longest time (23 weeks). Therefore,
the EST for the activity F is shown in the network as:

4
23

Latest Starting Time (LST)

LST is the time, at which the starting of an activity cannot be delayed beyond this time
(non-waiting time).
The LST for an event is obtained by making EST and LST equal for the end event and
calculated by working backward. The path with the shortest time among the paths is
considered as LST. In this example the LST of event 2 is 12 since between the two
paths 2-5 and 2-3-4-5, the path 2-3-4-5 is the shortest. The same can be shown in the
network for event 2 as:

17
 2
12

Similarly, the LST and EST of other events are calculated and shown below.

2
13 13
15
13 E
A 5
28 28
C 2
1 2
0 0 F
4
12 23 26

B 8
3
D
15 18

Identification of critical events

The critical events are the events with zero slack in the network. Event Slack is defined
as the difference between the Latest Starting Time (LST) and the Earliest Starting Time
(EST) of an event.
Event Slack Time = (LST-EST)
Slack time (week) for events 2, 3, and 4 are calculated and shown below:
Events LST EST Slack (LST-EST)
4 26 23 3
3 18 15 3
2 13 13 0
Besides the start and finish event, event “2” is the critical event in the above network.

Identification of critical activity

An activity can be called as a critical activity if the following conditions are satisfied.
i. LST and EST are equal at the head event.
ii. LST and EST are equal at the tail event.
iii. The difference between EST at the head and tail event of the activity equals to the
activity time.
iv. The difference between LST at the head and tail event of the activity equals to the
activity time.

18
In the above network, the activity A & E satisfying the above conditions are critical
activities.

Identification of critical path

The critical path is the chain of critical activity spanning the network from start to end
i.e. the path joining all the critical events. It is also the longest path from start to end of
the project network. Alternatively, therefore, comparing all the possible path lengths
can identify the critical path (see flow diagram). The critical path time is the shortest
duration of the project. The critical path is denoted preferably by denoting the critical
events on the path.
The critical path for the project is, A – E.
The critical path of the project can also be denoted in terms of the event numbers. In
the present project, it is 1- 2 – 5. To distinguish the critical path from other paths in the
project, it is preferable to use a thicker line to demarcate the critical path. It is quite
possible that a project can have multiple critical paths. In such cases, the length of all
the critical paths will be equal.

13.5.1 Critical Path and Project Management

The critical path time is the shortest project time any delay in completion of any of the
activities on the critical path would delay the entire project. Therefore it is a critical
activity that needs to be monitored for timely completion of the project. However, the
activities with positive event slack could be rescheduled within the available time
frame for efficient utilization i.e. smoothing of the demand on the available resources.
If the duration of the project requires to be reduced, activities on the critical path will
be the ones to be considered for completion at an early date with the allocation of
additional resources.
Determining the probability of completion of a project by a specific date (d)
Assuming the probability distributions of activities on the critical path are independent,
the variance of the time of the critical path is obtained by adding variances of activities
on the critical path (C.P.). In the previous example, A and E are the activities on the
critical path.
The variance of the activity times on C.P. (Te) = Var (ta) + Var (te)
s.d. (Te) = SQRT [Var (ta) + Var (te)]
With the information on the mean (m) and s.d (s) of critical path duration, and
assuming as normally distributed, we can compute the probability of completion of a
project by a specific date (d) as shown below:
Step 1 :
Find Z = (d – m) / sd
Step 2 :

19
Obtain cumulative probability up to z from the table “area under the normal curve”.
As an illustration, consider a dummy network with a mean (m) and s.d (s) of a critical
path as 28 weeks and s.d 3.0 weeks respectively. Then the probability of completion of
a project by certain specified dates is given by:
Specified period Z-value Probability of completion of a
(weeks) project by a specified period (d)
20 (20-28) / 3 = -2.67 0.005
25 (25-28) / 3 = -1.0 0.159
30 (30-28) / 3 = 0.67 0.73
It is seen from the above table that the project can be completed within 30 weeks with a
higher probability (0.73). On the other hand, it is highly impossible to complete it
within 20 weeks as indicated by the probability (0.005).
Example:
A farm manager has undertaken a project on to develop a farm in a regional station.
The activities identified in the project, their interrelationships, and the time distribution
of each activity are shown in the following activity table.
Activity Table
S. no. Name of the activity Symbol Preceding Time distribution
activity (weeks)
a M B
1. Land surveying A - 6 8 11
2. Plan preparation B A 4 7 10
3. Rock blasting C B 15 20 24
4. Land leveling D C 5 9 12
5. Drainage channel digging E D 4 6 10
6. Stone pitching of a F E 12 16 22
drainage channel
7. Digging of Wells G B 15 20 30
8. Laying of irrigation lines H G 10 12 18
9. Construction of pump I G 8 14 21
house
10. Procurement of pumps J G 4 6 8
11. Installation of pumps K I, J 5 7 12
12. Finishing works L F, H , K 4 6 10
Draw the flow chart, and network, and find the critical path. Also calculate the
expected time, slacks for each event, and the probability of completion of the project
for different specified periods.
Activity Number Symbol Te =(a+4m+b)/6 Sd(te)
1 A 8 0.83
2 B 7 1.00
3 C 20 1.50
4 D 9 1.20

20
 5 E 6 1.00
6 F 17 1.70
7 G 21 2.50
8 H 13 1.33
9 I 14 2.16
10 J 6 0.67
11 K 8 1.16
12 L 6 1.00

FLOW CHART

START

A 8

B 7

G 21 C 20

I 14 J 6 D 9

H 13 E 6

K 8
F 17

L 6

FINISH

21
 13
NETWORK
H

8 9 6 10
7
14
6 67 67 73 73
50 59 L
K
I
36 45

6 Z 0

8
17 F
21 G 42 59

0 8 1 7 2 20
3 9
4 6
5

B C D E 50 50
0 0 8 8 15 15 35 35 44 44
A

22
Critical path 0-1-2-3-4-5-9-10
Critical path duration = 73
Mean duration of C.P: 73 weeks
Sd (C.P) = 3.4 weeks
The probability for the completion of a project for specific dates (d) is calculated and
given below:
Specified date (weeks) (d) Z value Probability (from the
area under the normal
curve)
60 60 − 73 0.00005
= −3.82
3.4

65 65 − 73 0.015
= −2.35
3.4

70 70 − 73 0.47
= −0.09
3.4

75 75 − 73 0.61
= −0.60
3.4

80 80 − 73 0.98
= 2.06
3.4

From the table, it is seen, that the project may be completed within 60 days is highly
impossible. It may be completed within 75 weeks with a probability of 0.61. But the
project never goes beyond 80 weeks to complete as indicated by its associated
probability (0.98).

13.6 LET US SUM UP

The game theory presents a decision-maker to see the competitive situation from a
strategic perspective. Game theories are relatively used in abstract, theoretical, and
specific settings but the practical applicability of this technique cannot be neglected.
Game theory is used by business firms when making business decisions, including
which advertisement copy to be used, and how to price their products. With the use of
game theory, we can decide upon the best available strategies to the competitor and
also can estimate their financial gains or losses.

23
13.7 KEYWORDS
Dominance : States that if one strategy of a player dominates over the
other strategy in all conditions then the latter strategy can
be ignored. A strategy dominates over the other only if it
is preferable over the other in all conditions.
Game Theory : Set of concepts aimed at decision-making in situations of
competition and conflict (as well as cooperation and
interdependence) under specified rules.
Payoff matrix : A payoff is a number, also called utility that reflects the
desirability of an outcome to a player, for whatever
reason.
Saddle point : A point at which a function of two variables has partial
derivatives equal to zero but at which the function has
neither a maximum nor a minimum value.
Zero sum game : A game is said to be zero-sum if for any outcome, the
sum of the payoffs to all players is zero. In a two-player
zero-sum game, one player’s gain is the other player’s
loss, so their interests are diametrically opposed.

13.8 SUGGESTED FURTHER READINGS/REFERENCES

1. Dutta, P. K. (1999), Strategies and Games: Theory and Practice, Library congress
catalogues in publication.
2. Hira, D. S. and Gupta. P.K. (2007), Operation Research, S Chand and Company,
New Delhi.
3. Kalavathy, S. (2002), Operation Research, Fourth Edition, Vikas Publishing House
Private Ltd.
4. McKinsey, J. C. C. (2003), Introduction to the Theory of Games, Library congress
catalogues in publication.
5. Morris, P. (1994), Introduction to Game Theory, Springer - International Publisher

24
6. Panneerselvam, R. (2006), Operation Research, Second Edition, PHI Learning
Private Ltd, New Delhi, India.
7. Peters, H. (2008), Game Theory: A Multi-Leveled Approach, Springer -
International Publisher.
8. Sharma, A. (2009). Operations Research. Global Media, Himalaya Publishing
House.
9. Sharma, J.K. (2009), Operation Research – Theory and Application, Fourth
Edition, Macmillan Publishers India Ltd.
10. Tadelis, S. (2013), Game Theory: An Introduction, Library congress catalogues in
publication.
11. Taha, H. A. (2008), Operations Research – An Introduction, Eight Edition,
Prentice – Hall of India Private Ltd.
12. Thomas, L. C. (2003), Games, Theory and Applications, Library congress
catalogues in publication.
13. Webb, J. N. (2000), Game Theory: Decisions, Interaction and Evolution, Springer
- International Publisher.

13.9 ANSWERS TO CHECK YOUR PROGRESS

Check Your Progress 13.1
1. a
2. b
3. False
4. True

Check Your Progress 13.2

1. a
2. d
3. d
4. e

Check Your Progress 13.3

1. a
2. d

25
3. a
4. b
5. d

Check Your Progress 13.4

1. Fair Game
2. Unperson
3. Dominance
4. LP
5. Graphical

13.10 PRACTICE PROBLEMS

1. 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.
a.
Player B
B1 B2
A1 5 2
Player A
A2 7 4
b.
Player B
B1 B2
A1 1 1
Player A
A2 4 3
2. Solve the following game problem whose pay-off matrix is given below:
Player A
Player B 1 2 2
6 4 6
Determine the best strategy for players A and B and also the values of the game for
them. Is the game fair or strictly determinable?

26
3. Company management and the labor union are negotiating a new five-year policy
for employees. Each of these has four strategies.
i. Health and Safety
ii. Workforce conditions
iii. Employees Empowerment
iv. Job Enrichment
The costs to the company are given for every pair of stratified choices.

Company Strategies
Union Strategies I II III IV
I 20 15 12 35
II 25 14 8 10
III 40 2 10 5
IV 5 4 11 0
What strategy will the two adopt? Also, determine the value of the game.
4. Apply the rule of Dominance to the following market
B1 B2 B3
A1 9 8 7
A2 3 6 4
A3 6 7 7

SOLUTION TO PRACTICE PROBLEMS

1. a.
Player B
B1 B2 Row minima
A1 5 2 5
Player A A2 7 4 7

Column maxima 5 2

Maxi (minimum) = Max (5, 7) = 5

Mini (maximum) = Min (5, 2) = 5
Value of game = 5

27
Hence the game is strictly determinable.

1.b.
Player B
B1 B2 Row minima
A1 1 1 1
Player A
A2 4 3 3
Column maxima 4 1
Maxi (minimum) = Max (1, 3) = 1
Mini (maximum) = Min (4, 1) = 1
Value of game = 1
Hence the game is strictly determinable.
2. Solution:
Apply the steps of finding out the saddle; we obtain the saddle point which is enclosed
both in a rectangle and circle as shown in the following table
Player A Row minima
Player B 1 2 2 2
6 4 6 6
Column maxima 6 4 2
Value of the game = 2
Hence the game is not fair and strictly determinable.
3. Solution:
Applying the rule of finding out the saddle point, we obtain the saddle point is enclosed
both in a circle and a rectangle as shown in the following table
Company Strategies
Union Strategies I II III IV Row minimum
I 20 15 12 35 12 Maximin
II 25 14 8 10 8
III 40 2 10 5 2
IV 5 4 11 0 5
Column maximum 40 15 12 35
As shown in the above table;
Maximin = Minimax = Value of Game = 12

28
Therefore the company will adopt strategy III - Employees Empowerment and the
union will adopt strategy I - Health and safety.
4. Solution:
a. Rule of Row: It can be seen that pay-offs relating to A2, are less or equal to
those for row A3. Hence row A2 is dominated by row A3, and can be deleted.
(3 < 6; 6 < 7; 4 < 7 and A2 is dominated)
B1 B2 B3
A1 9 8 7
Reduced Matrix
A3 6 7 7
b. Rule of column: Now column B2 is dominated by B3 and hence B2 can be
deleted. Here 8 > 7, and 7 = 7, while comparing elements of columns B2 and
B3, column Be elements being more or equal B2, stands dominated
B1 B3
A1 9 7
Reduced Matrix
A3 6 7
There is no saddle point in this case.
13.11 UNIT END QUESTIONS
1. Game theory provides a systematic quantitative approach for analyzing completive
situations in which the competitor makes use of logical processes and techniques in
order to determine an optimal strategy for winning (refer Shaktiman atta and Ayush
atta). Comment.
2. What do you understand by the term Theory of Games? Explain its use in decision-
making.
3. What assumptions are made in the theory of games?
4. Describe a two-person zero-sum game.
5. Describe Saddle Point. Is it necessary that a game should always possess a saddle
point?
6. Explain the meaning of the Dominance principle in game theory.
7. Define the following terms
a. Pure Strategies
b. Mixed Strategies
c. Two persons zero game (or rectangular) game
d. Payoff matrix
8. How do you solve a game when;

29
 a. Saddle point exists
b. Saddle point does not exist
9. Discuss the fundamental difference between a pure strategy game and a mixed-
strategy game.

30