MAT506 Operations Research

YASHWANTRAO CHAVAN MAHARASHTRA OPEN UNIVERSITY
SCHOOL OF SCIENCES
(FORMERLY SCHOOL OF ARCHITECTURE, SCIENCE & TECHNOLOGY)
V151: M.Sc. Mathematics

2023 {As per NEP 2020}
Pattern
MAT506
OPERATIONS RESEARCH
(4 Credits)
Semester - I
Email: director.ast@ycmou.ac.in
Website: www.ycmou.ac.in
Phone: +91-253-2231473
Yashwantrao Chavan S25035
Maharashtra Open Operations Research

University
Brief Contents
Vice Chancellor’s Message .................................................. 3
Forward By The Director .................................................... 4
Credit 01 .................................................................................. 5
Unit 01-01: Basics of Operations Research ................. 6
Unit 01-02: Linear Programming Problems (L.P.P.) 23
Unit 01-03: The Simplex Method ................................... 44
Unit 01-04: Duality ............................................................ 79
Credit 02 .................................................................................. 103
Unit 02-01: Game Models and Related Theory .......... 104
Unit 02-02: Two Person Zero Sum Game .................... 123
Unit 02-03: Dominance in Games ................................. 139
Unit 02-04: Mixed Strategies (2×N and M×2 games) 166
Credit 03 .................................................................................. 196
Unit 03-01: Network Analysis ........................................ 197
Unit 03-02: Fulkerson’s Rule ......................................... 218
Unit 03-03: Critical Path Method (C.P.M.) ................. 241
Unit 03-04: Project Evaluation and Review Technique (Pert) 266
Credit 04.................................................................................. 288
Unit 04-01: Simulations and It’s Theory ..................... 289
Unit 04-02: Monte Carlo Simulation ............................ 301
Unit 04-03: Generation of Random Numbers ............ 323
Unit 04-04: Simulation Languages ............................... 335
Answers to Self-Tests and End of Unit Exercises ......... 340
S25035: Operations Research Page 1

S25035: Operations Research
Yashwantrao Chavan Maharashtra Open University

Vice-Chancellor: Prof. Dr. E. Vayunandan
School of Architecture, Science and Technology
Programme Advisory Committee (PAC)
Dr Sunanda More Dr Manoj Killedar Dr Chetana Kamlaskar
Director(I/c) & Associate Associate Professor, Assist. Professor,
Professor, School of School of Architecture, School of Architecture,
Architecture, Science & Science & Technology, Science & Technology,
Technology, YCMOU, YCMOU, Nashik YCMOU, Nashik
Nashik
Dr. T.M. Karade Prof. Dr Shivdas D Katore Prof. Dr J N Salunke
Retired Professor, Professor, Professor, Swami
R.T.M. Nagpur Sant Gadge Baba, Ramanand
University, Nagpur Amravati University
Teerth Marathwada
University, Nanded
Prof. Dr Meenakshi Prof. Dr S R Chaudhari, --
Wasadikar, Professor and HOD,
Professor and HOD, KBC North Maharashtra
Dr Babasaheb Ambedkar University, Jalgaon
Marathwada University
Development Team
Course Coordinator and Book Writer Book Editor
Instructional Technology
Editor
Dr. Subhash G. Pawar Dr. D. D. Pawar
Dr Chetana Kamlaskar M. Sc., M. Phil, PhD M.Sc., Ph.D.,
B.E., M. Tech., Ph.D., Head and Asso. Professor, Professor & Director,
Assistance Professor, K.T. H. M. College, School of Mathematical
Nashik. Sciences, SRTM University
School of Architecture,
Experience: 35 years Experience: 20+ Yrs
Science & Technology,
Credits written:01, 04
YCMOU, Nashik
Experience@YCMOU: Ms. Chetana V. Visave
20+ Yr M. Sc., NET, SET
Assistant Professor,
Bhavan’s College,
Andheri(W), Mumbai.
Experience: 05 years
Credits written:02, 03
This work by YCMOU is licensed under a Creative Commons
Attribution-NonCommercial-ShareAlike 4.0
International License .
 First Book Publication : 2 Oct 2021 Publication Number : 2428
 Publisher : Dr. Dinesh Bhonde, Registrar, YCMOU, Nashik- 422 222, MH, India
 ISBN Number : 978-93-92982-16-3
This SLM Book V130: M.Sc. (Mathematics) {2021 Pattern}, dtd. 02/10/2021
used in V151: M.Sc. (Mathematics) {2023 Pattern}, dtd. 31/08/2023
V ICE C HANCELLOR ’ S M ESSAGE
Dear Students, Greetings!!!
I offer cordial welcome to all of you for the Master’s degree programme of Yashwantrao
Chavan Maharashtra Open University. As a post graduate student, you must have autonomy to
learn, have information and knowledge regarding different dimensions in the fie ld of
Mathematics and at the same time intellectual development is necessary for application
of knowledge wisely. The process of learning includes appropriate thinking, understanding important
points, describing these points on the basis of experience and observation, explaining
them to others by speaking or writing about them. The science of education today accepts the
principle that it is possible to achieve excellence and knowledge in this regard.
The syllabus of this course has been structured in this book in such a way, to give you autonomy to
study easily without stirring from home. During the counseling sessions, scheduled at
your respective study centre, all your doubts will be clarified about the course and you will
get guidance from some qualified and experienced counsellors / professors. This guidance will not
only be based on lectures, but it will also include various techniques such as question-
answers, doubt clarification. We expect your active participation in the contact sessions at the study
centre. Our emphasis is on ‘self study’. If a student learns how to study, he will become independent in
learning throughout life. This course book has been written with the objective of helping in self-study and
giving you autonomy to learn at your convenience.
During this academic year, you have to give assignments, complete laboratory
activities, field visits and the Project work wherever required. You may have to opt for
specialization as per programme structure. You will get experience and joy in personally doing
above activities. This will enable you to assess your own progress and there by achieve a larger
educational objective.
We wish that you will enjoy the courses of Yashwantrao Chavan Maharashtra Open
University, emerge successful and very soon become a knowledgeable and honorable
Master’s degree holder of this university. I congratulate “Development Team” for the
development of this excellent high quality “Self- Learning Material (SLM)” for the
students. I hope and believe that this SLM will be immensely useful for all students of
this program.
Best Wishes!
- Prof. Dr. E. Vayunandan

Vice-Chancellor, YCMOU

F ORWARD B Y T HE D IRECTOR
This book aims at acquainting the students with advance Mathematics required at post
graduate degree level.
The book has been specially designed for Science students. It has a comprehensive
coverage of Mathematical concepts and its application in practical life. The book
contains numerous mathematical examples to build understanding and skills.
The book is written with self- instructional format. Each chapter is prepared with
articulated structure to make the contents not only easy to understand but also
interesting to learn.
Each chapter begins with learning objectives which are stated using Action Verbs as
per the Bloom’s Taxonomy. Each Unit is started with introduction to arouse or
stimulate curiosity of learner about the content/ topic. Thereafter the unit contains
explanation of concepts supported by tables, figures, exhibits and solved illustrations
wherever necessary for better effectiveness and understanding.
This book is written in simple language, using spoken style and short sentences. Topics
of each unit of the book presents from simple to complex in logical sequence. This book
is appropriate for low achiever students with lower intellectual capacity and covers the
syllabus of the course.
Exercises given in the chapter include MCQs, conceptual questions and practical
questions so as to create a ladder in the minds of students to grasp each and every
aspect of a particular concept.
I thank the students who have been a constant motivation for us. I am grateful to the
writers, editors and the School faculty associated in this SLM development of the
Programme.
- Dr. Sunanda More

Director (I/C) & Associate Professor,
School of Architecture, Science and Technology

C REDIT 01

UNIT 01-01: BASICS OF OPERATIONS RESEARCH
LEARNING OBJECTIVES
After successful completion of this unit, you will be able to
 Understand the different definition of operations research.
 Learn characteristics of operation research.
 Understand the scope of operation research in industry, public utilities , etc.
INTRODUCTION
Business managers face an endless list of complex issues every day. They must
make decisions about financing, where to build a plant, how much of a product to
manufacture, how many people to hire, and so on. Often, the factors that make up
business issues are complicated, and they may be difficult to comprehend. Operations
research is a way to deal with these thorny problems.
01-01: OPERATIONS RESEARCH AND IT’S CHARACTERISTICS

Definition Operation Research (O.R.):
Without a doubt we can say operation studies is the studies of operation. An

operation may be called a set of acts required for the achievement of a desired outcome.
Such complex, interrelated acts can be performed by four types of systems: Man,
Machine, Man-Machine unit and any organization of men, machines, and man -machine
units. OR is concerned with the operations of the last type of system.
Many definitions of O.R. had been suggested from time to time. It isn't always
smooth to define operations research as it is not a technological know -how representing
any properly-defined social, organic or physical phenomenon. Some of the one -of-a-
kind definitions advised are:.
Definitions of Operations Research (O.R.):

1) O.R. is a scientific method of providing executive departments with a quantitative
basis for decisions regarding the operations under their control.
–Morse and Kimball.
2) O.R. is a scientific approach to problem solving for executive management.
– H. M. Wagner
3) O.R. is the application of scientific methods, techniques and tools to
operation of a system with optimum solution to the problem.
- Churchman, Ackoff and Arnoff.

4) O.R. is the application of modern methods of mathematical science to complex
problem involving management of large system of men, machines, materials and
money in industry, business, government and defense. The distinctive approach is
developing a scientific model of the system incorporating measurements of factors
such as chance and risk to predicts and compare the outcomes of alternatives
decisions, strategies or controls. –J.O.R. Society, U.K.
5) The application of the scientific method to the study of operations of large complex
organizations or activities. It provides top level administrators with a quantitative
basis for decisions that will increase the effectiveness of such organizations in
carrying out their basic purposes.
– Committee on O.R .National Research Council, USA
In addition to the above there are hundreds of definitions availa ble to explain
what operations research is?. But many of them are not satisfactory because the
above-discussed definitions are given by various people at different times and
stages of development of O.R. as such they have considered the field in which they
are involved hence each definition is conce ntrating on one or two aspects. No
definition is having universal approach.
Characteristics of Operation Research:
The various definitions of O.R. bring out the essential characteristics of O.R.
They are as follows:
1) Interdisciplinary Team approach:
The highly optimized solution is found by a group of scientists who are selected
from diverse disciplines. For example, one may find mathematicians, statisticians,
physicist, psychologist, economist and an engineer working together on an O .R.
problem.
The use of interdisciplinary team can produce more unique solution with greater
probability success, than could be expected from the same number of persons from an
single discipline. Another reason for existence of O.R. team is that knowledge is
increasing at very fast rate.
2) Operations Research increases the creative ability of the decision maker:
O.R. provides manager mathematical tools, techniques and various models to

analyses the problem on hand and to evaluate the outcomes of various alterna tives and
make an optimal choice. This will definitely help him in making better and quick
decisions. A manager, without the knowledge of these techniques has to make decisions
by thumb rules or by guess work, which may click some times and many a time put him
in trouble. Hence, a manager who uses O.R. techniques will have a better creative-
ability than a manager who does not use the techniques.
3) Operations Research is a systems approach:
A business or a government organization or a defense organization may be

considered as a system having various sub-systems. The decision made by any sub -
system will have its effect on other sub-systems. Say for example, a decision taken by
marketing department will have its effect on production department. When dealing with
O.R. problems, one has to consider the entire system, and characteristics or sub -
systems, the inter-relationship between sub-systems and then analyze the problem,
search for a suitable model and get the solution for the problem. Hence, we say O.R. is
a systems Approach.
SOLVED PROBLEMS 01
Problem 01-01-01: Discuss the operations research approach to problem solving.
Solution: The maximum critical feature of operations studies is the use of the clinical
approach and the constructing of selection fashions. T he operations studies approach to
trouble fixing is primarily based on 3 levels, specifically (a) judgment phase; (b)
studies section, and (c) movement section.
a) Judgment phase: This segment includes: (i) identity of the actual-existence

problem, (ii) choice of the right objective and the values of various variables
related to this objective, (iii) application of the best scale of size, and (iv) formula
of the appropriate model of the problem and the abstraction of the vital
information, so that a technique to the decision-maker’s goals may be obtained.
b) Research phase: This phase is the most important and longest among all of the
levels. However, despite the fact that the last aren't as long, they are additionally
equally critical as they offer the premise for a scientific method. This section
makes use of: (i) observations and statistics collection for a higher information of
the problem, (ii) formula of hypothesis and model, (iii) remark and
experimentation to test the hypothesis on the basis of extra information, (iv)
analysis of the available records and verification of the hypothesis the usage of
pre-established measures of desirability, (v) prediction of various outcomes from
the hypothesis, and (iv) generalization of the result and consideration of
opportunity techniques.

c) Action phase: This phase consists of making pointers for implementing the
decision. This selection is implemented through a man or woman who is in a
position to put in force consequences. This character has to be aware about the
environment in which the trouble took place, be aware about the objective, of
assumptions behind the problem and the required omissions of the version.
Problem 01-01-02: Explain the role of operation research in decision making.
Solution: O.R. plays an important role in almost all areas of decision maki ng. Some
areas where operation research (O.R.) techniques can be used as listed below:
a) Financing and Investment Polices: This category includes the analysis of credit
and loan policy, fund flow and cash flow and also considers the dividend, share
and bonus policy. It reflects the portfolio of investment too.
b) Selling, Promotion, Marketing & Publicity: It emphasis on selection of the

product and timing, which means what type of commodity should be
manufactured at the proper passage of the time. It focuses on the media of
publicity like print media and electronic media may be chosen. It also used to
know about the number of sales persons are required to some speci fic area.
Moreover O.R. also use to select the product mixed which include product,
price, place and promotions.
c) Acquisition, Investigation, Manufacturing and Personnel Management : The

operational research techniques being used to purchase, substitute and
reorganize the most advantageous policies. These tec hniques also include
location and manufacturing size of retail outlets, factories and warehouses,
loading and unloading facilities for trucks, allocation and scheduling of
resources and optimum to merge the products, recruitment and selection of best
suitable employee for a company and job assignments , etc.
SELF-TEST 01
MCQ 01-01: Operation research approach is
a) Scientific.
b) Initiative.
c) Multi-disciplinary.
d) All of the above.
MCQ 01-02: Operation research is
a) Increase the creative capabilities of the decision maker.

b) Decrease the creative capabilities of the decision maker.
c) Stable the creative capabilities of the decision maker.
d) Not effect on the creative capabilities of the decision maker.
SHORT ANSWER QUESTIONS 01

SAQ 01-01-01: Define O.R. and explain are the characteristics of O.R.?
Solution: Characteristics of O.R. are as follows:
i) Interdisciplinary team approach.
ii) Scientific approach.
iii) There are also other characteristics like decision making, quantitative solution,
use of information technology, etc.
SAQ 01-01-02: Define O.R. and discuss O.R. as an interdisciplinary approach.
Solution: Write any one or two definitions of O.R.
Interdisciplinary Team approach: The highly optimized solution is found by a group

of scientists who are selected from diverse disciplines . For example, one may find
mathematicians, statisticians, physicist, psychologist, economist and an engineer
working together on an O.R. problem.
The use of interdisciplinary team can produce more unique solution with greater
probability success, than could be expected from the same number of persons from an
single discipline.
01-02: SCOPE OF OPERATIONS RESEARCH

Whenever there is problem of optimization, there is a scope of O.R. technique of
operation research is used in a wide range of situations:
1) In Industry
If the industry manager makes his policies simply on the basis of his past
experience and a day approaches when he gets retirement, then a serious loss is
encounter ahead of the industry. This heavy loss can be right away compensated
through appointing a young specialist of O.R. techniques in business management. O.R.
has been successfully applied in industry in the fields of production, blen ding, product
mix, inventory control, demand forecast, sale and purchase, transportation, repair and
maintenance, planning, etc. O.R. divides the company in three departments of
production, marketing and financial department. In production department , O.R.
reduces the minimize cost of production. In marketing, O.R. is useful to manager for

maximizing sales. In the financial department , O.R. minimize the capital required and
maintain any level of the business.
Industry
Business Defence
Scope
of OR
Hospitals Planning
L.I.C. Agriculture
2) In Defense:
O.R. has wide scope for application of O.R. in defense. There three are defense
agencies namely Air force, Navy and Army. The decision taken by one will have its
effect on the other. Hence, proper co-ordination of the three bases and smooth flow of
information is necessary. Here, O.R. techniques will help the departmental heads to take
appropriate decisions.
3) In Planning for Economic Growth:
In India, we have long-term anticipating consistent financial development. Each

state government needs to get ready designs for adjusted development of the state.
Different secretaries having a place with various offices needs to co -ordinate and plan
for consistent financial development. For this , all departments can use O.R. techniques
for planning purpose. The inquiry like the number of engineers, doctors, software
people are needed in future and what should be their quality to deal with the then issues
and so on, can be easily solved.
4) Agriculture:
With the sudden increase of population and resulting shortage of food, ever y
country is facing the problem of optimum allocation of land to variety of crops as per
the climate condition with available facilities. Also, every developing country is facing

a problem of optimum water distribution from various water resources. Therefo re, using
O.R. good quality work can be done in this direction.
5) Planning:
The basic in most of the developing countries in Asia and Africa is to remove the
poverty and hunger as quickly as possible. So, there is a great scope to economist,
statistician, administrator, politicians and technicians working in a team to solve this
problem by an O.R. approach.
6) Public Utilities:
i) O.R. approach may be additionally carried out in big hospital to reduce waiting
time of out-doors patients and to resolve the administrative problem.
ii) O.R. also used to solve the problem in area like transport to regulate train arrival
and their running times. Also similar to big hospital using queuing theory, we can
minimize congestion and passenger waiting time.
iii) O.R. is directly applicable to business and society. For example, it is increasingly
begin applied in L.I.C. offices to decide the premium rate of various policies.
iv) O.R. is also been used in petroleum, paper chemical, metal processing, aircraft,
transport and distribution, mining and textile industries.
v) O.R. also used in each and every small or big organization for managing employs,
production, purchasing row material, etc. Using O.R. we can minimize the cost
and maximize profit. Thus, we find that O.R. have a wide scope on the social,
economic and industrial problem today.
SOLVED PROBLEMS 02
Problem 01-01-03: Explain the scope of O.R. in industries.
Solution: O.R. plays very important role in industries. As company extended, it is not
possible for one man to manage them. Because of this the company is divided into the
different units like production department, marketing department and financial
department. In production department using O.R., we can minimize the cost of
production, increase the productivity and use the available resources carefully and for
healthy industrial growth. In marketing department, we can maximize profit and
minimize cost using O.R.
Problem 01-01-04: Explain the scope of O.R. in traffic control.
Solution: Because of population explosion, the increase within the wide variety and
verities of vehicles, road density is continuously increasing. Specifically in peak hours,

it will likely be a headache to manipulate the visitors. As a result, proper timing of site
visitors signaling is essential. Depending on the flow of commuters, proper signaling
time is to be worked out. This can be effortlessly finished by the application of
queuing theory form O.R.
SELF-TEST 02
MCQ 01-03: Which of the following statement is FALSE?
a) Operation research used for maximizes profit in the company.
b) Operation research used for minimize the profit in the company.
c) Operation research used for minimize the cost of production in the company.
d) Operation research used for increase the productivity in the company.
MCQ 01-04: Which of the following is not application of operation research?
a) A media planning.
b) Metal processing.
c) Marketing department.

SAQ 01-01-03: Define O.R. and discuss the scope of O.R. in defense.
Solution: O.R. has wide scope for application of O.R. in defense. There three are
defense agencies namely Air force, Navy and Army. The decision taken by one will
have its effect on the other. Hence, proper co-ordination of the three bases and smooth
flow of information is necessary. Here, operations research techniques will help the
departmental heads to take appropriate decisions.
SAQ 01-01-04: What is the role of O.R. in public utilities? Explain in briefly.
Solution: Scope of O.R. in public utilities is as follows:
i) O.R. approach may be additionally carried out in big hospital to reduce waiting
time of out-doors patients and to resolve the administrative problem .
ii) O.R. also used to solve the problem in area like transport to regulate train arrival
and their running times. Also, similar to big hospital using queuing theory, we can
minimize congestion and passenger waiting time.
iii) O.R. is directly applicable to business and society. For example, it is in creasingly
begin applied in L.I.C. offices to decide the premium rate of various policies.

iv) O.R. is also been used in petroleum, paper chemical, metal processing, aircraft,
transport and distribution, mining and textile industries, etc.
01-03: OPERATIONS RESEARCH MODELS

There is wide scope of O.R. hence there is no unique set of problems that can be
solved by O.R. technique. A model, is used in O.R., is defined as an idealized
representation of the real-life situation. Most O.R. studies involve the construction of a
mathematical model. The model is a collection of logical and mathematical relationship
that represents aspect of the situation under study.
Many O.R. models have been developed and applied to the problem tin business
and industry. Some of these models are described here.
Allocation
Models
Decision Queuing
Theory Models
Mathemati
cal Inventory
Teqhnique Models
s
OR Models
Statistical Simulation
Techniques Models
Dynamic Replaceme
Models nt Models
a) Allocation models: Allocation models are used to solve the problem in which there
are a number of jobs to be perform and there are alternative ways of doing them and
resources of facilities are limited. If the measure of effectiveness such as profit,
cost, etc., is represented as a linear function of several variables and limitations on
resources (constraints) are expressed as a system of linear equalities or inequalities,
the allocation problem is classified as a linear p rogramming problem.
The simple type of allocation model involve the association of the number of
jobs to the same number of resources (men). This is called assignment model. But if
the activities require more than one resource and conversely if the resou rces can be

used for more than one activity, the allocation problem is classified as a
transportation problem.
b) Waiting line (or Queuing) models: These models establish a trade-off between
costs of providing service and the waiting time of a customer in the queuing system.
A queuing model describes: Arrival process, queue structure and service process and
solution for the measure of performance – average length of waiting time, average
time spent by the customer in the line, traffic intensity, etc., of the w aiting system.
c) Inventory models: These model deals with idle resources such as raw materials
spare parts, semi-finished and finished products. These models are concern with two
decisions; how much to order and when to order, so that minimizes the total co st.
The main objective is to minimize the sum of three conflicting inventory costs: the
cost of holding or carrying extra inventory, the cost of shortage or delay in the
delivery of items when it is needed and the cost of ordering or set -up.
d) Simulation models: The meaning of simulation is imitation. These models have

mathematical structure but cannot be solved by using mathematical techniques. It
needs certain experimental analysis. To study the behavior of the system, we use
random numbers. More complex systems can be studied by simulation. Studying the
behavior of laboratory model, we can evaluate the required values in the system.
Only disadvantage of this method is that it does not have general solution method.
e) Dynamic models: These models are used where a problem requires optimization of
multistage (sequence of interrelated decisions) decision processes. The method
starts by dividing a given problem into stages or sub-problems and then solves those
sub-problems sequentially until the solution to the original problem is obtained.
f) Mathematical models: In these models, the variables of a problem is represented

by mathematical symbols, letters, etc. To show the relationships between variables
and constraints, we use mathematical symbols. Hence, these are known as symbolic
models or mathematical models. These models are used very much in O.R.
Examples of such models are resource allocation model, newspaper boy problem,
transportation model, etc.
g) Network models: These models are applied in a large and complex project
involving a number of interrelated activities, requiring a number of men, machine
and materials, it is not possible to make and execute an optimal schedule just by
intuition. PERT/CPM techniques help in identifying delay and project critical path.
These techniques help to achieve project co-ordination and efficient use of

resources. They are also used to determine time-cost trade off, resources allocation
and updating of networks.
h) Markov-chain models: This model is used for decision making is used in situations
where various states are defined. The probability of going from one state to another
is known and depends on the present state but independent of how that state was
reached. Using this probability percentage, we can take good decision. Markov
analysis the current behavior of some variable in order to predict the future
behavior of that variable. This was used by the Russian Mathematician A. Markov
to describe the behavior of gas particle in the cl osed container.
i) Replacement model: Any capital item, that is continuously used for supplying
service or for producing the product is subjected to put on and tear due to
utilization, and its performance goes on reducing. This discount in performance can
be expected by the growing range of breakdowns or reduced productiveness. The
worn-out parts or components are to be replaced to bring the gadget lower back to
work. This action is known as protection. A time is reached while the renovation fee
turns into very excessive and the supervisor feels to update the antique system by
means of new one. This sort of troubles known as replacement issues and may be
solved by using alternative fashions.
SOLVED PROBLEMS 03
Problem 01-01-05: State characteristics of good model.
Solution: The characteristics of good model are as follows:
a) The number of simplifying assumptions should be as few as possible.
b) The number of relevant variables should be as few as possible.
c) It should assimilate the system environmental changes with out change in its
framework.
d) It should be adaptable to parametric type of treatment.
e) It should be easy and economical to construct.
f) A model must take less time to construct.
Problem 01-01-06: What are the advantages and limitations of model?
Solution: Advantages of a model are as follows:
 It provides a logical and systematic approach to the problem.
 It indicates the scope as well as limitation of a problem.

 It helps in finding avenues for new research and improvement in a system.
 It permits experimentation and analysis of a complex system without directly

interfering in the working and environment of the system.
 It makes the overall structure of the problem more compressible and help in
dealing with problem in its entirety.
 Models saves resources like money, time, etc.
 Models help the analyst to find newer ways of solving the problem.
Limitations of a model:
i) Models are only idealized representation of reality and should not be regarded as
absolute in any case.
ii) The validity of a model for particular situation can be ascertained only by
conducting experiments on it.
SELF-TEST 03
MCQ 01-05: Every mathematical model
a) must be deterministic.
b) requires computer aid for its solution.
c) represents data in numerical form.
d) all of the above.
MCQ 01-06: A model is
a) an idealization.
b) an essence of reality.
c) an approximation.

SAQ 01-01-05: Define model and describe any one type of model.
Solution: A model is used in O.R., is defined as an idealized representation of the real-

life situation. Most O.R. studies involve the construction of a mathematical model. The
model is a collection of logical and mathematical relationship that represents aspect of
the situation under study. Some of the models are simulation model, network model,
dynamic model, inventory model, etc. Explain any one type of model.

SAQ 01-01-06: Discuss the inventory model briefly.
Solution: Inventory model deals with idle resources such as raw materials spare parts,
semi-finished and finished products. These models are concern with two decisions; how
much to order and when to order, so that minimizes the total cost. The main objective is
to minimize the sum of three conflicting inventory costs: the cost of holding or carrying
extra inventory, the cost of shortage or delay in the delivery of items when it is needed
and the cost of ordering or set-up.
01-04: METHODOLOGY OF OPERATIONS RESEARCH

Every O.R. specialist may have his/her own way of solving problems. However,
the effective use of O.R. techniques require to follow a sequence of steps .
Problem Formulation
Constructing a
Implimenting the
mathematical
solution
model
Implementation and Deriving a solution

control the solution from model
Testing the model
Step 1: Defining the problem
It is impossible to obtain ‘Right answers from wrong problem’. In formulating

the problem, objectives, alternative course of action, constraints and effect of the
system under study must be well established.
Step 2: Constructing a mathematical model

The next step is to construct a mathematical model of the problem with
following basic elements:
Decision or controllable variables and parameters or uncontrollable variables. A

controllable variable is one that can be directly controlled by the decision maker. The
value of these variables is to be determined. Uncontrollable variables are those which
are not under direct control of decision maker. Ex govt. policy, increasing prices.

Objective function: They are mathematical expression of profit or cost for a particular
operation.
Constraints: There are certain constraints (limitations) on the use of resources, and
such constraints arise due to organizational policy, legal restraints or limited resources
such as space, money, manpower, material, etc. The constraints on the use of resources
are expressed either in the form of equations or inequalities.
Functional relationships: In a decision problem, the decision variables in the objective

function and in the constraints are connected by a specific functional relationship .
Parameters: These are constants in the functional relationships amo ng decision

variables. Parameters can either be deterministic or probabilistic in nature. A
deterministic parameter is one whose value is assumed to occur with certainty.
Otherwise, it is probabilistic.
Step 3: Solving the mathematical model
Once a mathematical model of the problem has been formulated, the next step is
to obtain numerical values of decision variables. Obtaining these values depends on the
specific form or type of mathematical model .
Step 4: Testing the model and the solution
It involves testing of the model used. A model is valid if it can provide a

reliable prediction of system’s performance. A good operation research analyst
continuously tries to update his model.
Step 5: Implementing and control the solution
The next step is implementing the solution derived. Explanation of the solution
be made in terms of procedure used in actual system. After applying the solution,
observe response of the system. A control over the solution is established by proper
feedback.
SOLVED PROBLEMS 04
Problem 01-01-07: What are the limitations of O.R.? Explain briefly.
Solution: The demerits or limitations of O.R. are depicted below:
a) Mathematical models are applicable to only specific categories problem.
b) O.R. experts make very complex models for solving problems. These models
may not be realistic. Hence, they may not be useful for real -life situations.

c) O.R is very costly. This is because O.R. makes mathematical models for taking
decisions and solving problems. The company has to make various models for
solving different problems. All this increments the cost.
d) O.R. techniques try to find out an optimal solution taking into account all the
factors. In the modern society, these factors are enormous and expressing them
in quantity and establishing relationships among these require voluminous
calculations that can only be handled by computers.
e) O.R. is very complex concept. It is very difficult for an average manager to

understand it. Therefore, most managers do not use O .R. techniques.
Problem 01-01-08: Explain the role of computer in operation research.
Solution: The computer plays vital role in development of O.R. Many real-life O.R.
models require long and complex mathematical calculations. Thus, computer software
packages that are used to do these calculations rapidly and effectively have become a
part of O.R. approach to problem solving. Computer facilities such as spread sheets or
statistical and mathematical software packages that make such analysis readily
available to a decision-maker. Many computer manufacturing companies have
developed software packages for problem to solve by application of O .R. technique.
Companies such as IBM, ICL, UNIVAC, CDC have done so for solving scheduling,
inventory, simulations, queuing, networking (PERT/CPM) and many more O .R.
problem. Thus, the computer is an essential and integral part of OR.
SELF-TEST 04
MCQ 01-07: Operation research approach to decision maki ng based on
a) Judgments.
b) Research.
c) Action.
MCQ 01-08: The quantitative approach to decision analysis is a
a) logical approach.
b) rational approach.
c) scientific approach.

SAQ 01-01-07: Discuss the methodology of O.R.
Solution: The solving of a model of operational research need to follow the following
steps:
In the first step, we have to define a problem in the proper way for correct
answer. Then in the second step, we have to formulate the given problem in to
mathematical model by defining objective function, constraint, parameters , etc.
Once a mathematical model of the problem has been formulated, the next step is
to obtain numerical values of decision variables. Obtaining these values depends on the
specific form or type of mathematical model .
The next step is testing model, solution and implementing the solution derived.
Explanation of the solution made in terms of procedure used in actual system. After
applying the solution, observe response of the system. A control over the solution is
established by proper feedback.
SAQ 01-01-08: Define objective function and constraint.
Solution: Objective function: They are mathematical expression of profit or cost for a
particular operation. They are of maximization or minimization type.
Constraints: There are certain constraints (limitations) on the use of resources, and
such constraints arise due to organizational policy, legal restraints or limited resources
such as space, money, manpower, material, etc. The constraints on the use of resources
are expressed either in the form of equations or inequalities.
SUMMARY
Operations research is the clinical technique to decision making. This method
includes defining the problem, growing a version, fixing the model , checking out the
answer, studying the consequences and enforcing the consequences. In the use of
operations research technique, but there may be capacity troubles, which include
conflicting viewpoints, the effect of quantitative analysis on numerous sect ions of the
company, beginning assumptions, old answers, information the model, obtaining
applicable enter statistics, acquiring surest solution, checking out the solution, and
studying the consequences. In the usage of the operations study’s technique,
implementation isn't the final step because there can be a loss of commitment to the
method and resistance to exchange.

END OF UNIT EXERCISES
1. What is O.R.? Describe briefly its applications.
2. Define O.R. Discuss scope and limitation of O.R.
3. Discuss the scientific methods in O.R.
4. What are the situations where O.R. techniques are applicable?
5. Discuss the scope of O.R. in business.
6. Explain the role of computer in O.R. development.
KEY WORDS
Operation research (O.R.), Model, Objective function, Constraint.
REFERENCES
BOOKS
1) OPERATIONS RESEARCH, PREM KUMAR GUPTA AND D. S. HIRA, S. CHAND
& COMPANY LTD
2) OPERATIONS RESEARCH 9 TH EDITION, HAMDY A. TAHA, PEARSON.
3) OPERATIONS REEARCH 28 TH EDITION, B. S. GOEL AND S. K. MITTAL, A

PRAGATI EDITION
4) OPERATIONS REEARCH 6 TH EDITION 2017, J. K. SHARMA
MOOCS
YOUTUBE VIDEOS
 https://youtu.be/5KTtyFHi2ws
 https://youtu.be/C_EwiEzlHq0
 https://youtu.be/KA97JSw3kIE
WIKIPEDIA
https://en.wikipedia.org/wiki/Operations_research ,
https://theintactone.com/2020/01/04/nature-definition-characteristics-of-
operations-research/
OER
COURSE COMPANION WEBSITE
Visit here for course companion website for this course:

UNIT 01-02: LINEAR PROGRAMMING PROBLEMS (L.P.P.)
LEARNING OBJECTIVES
 Identify situations in which linear programming technique can be applied.
 Formulate linear programming problem.
 Solve L.P.P. using graphical methods.
 Recognize special cases such as, multiple optimal solutions, unbounded and
infeasible solutions.
INTRODUCTION
A model, which is used for optimum allocation of scarce or limited resources to
competing products or activities under such assumptions as certainty, linearity, fixed
technology, and constant profit per unit, is linear programming.
Linear programming is one of the maximum flexible, powerful and useful

strategies for making managerial selections. Linear programming technique can be used
for fixing huge variety of problem bobbing up in business, g overnment, industry,
hospitals, libraries, and so on. Whenever we want to allocate the available limited
resources for various competing activities for achieving our desired objective, the
technique that helps us is LINEAR PROGRAMMING. As a selection making tool, it
has established its price in diverse fields along with manufacturing, finance, advertising
and marketing, studies and improvement and personnel management. Determination of
optimal product mix (a mixture of products, which offers maximum profit) ,
transportation schedules, assignment problem and many more. In this section, we have
to discuss approximately various types of linear programming model.
The general linear programming problem (or model) with 𝑛 decision vaniables
and 𝑚 constraints can be stated in the following form:
Optimize (Max. or Min.) 𝑍 = c x + c x + ⋯ + c x
subject to the linear constraints,
a x + a x + ⋯ + a x (≤, =, ≥) b
𝑎 𝑥 + 𝑎 𝑥 + ⋯ + 𝑎 𝑥 (≤, =, ≥) 𝑏
⋮ ⋮ ⋮
𝑎 𝑥 + 𝑎 𝑥 + ⋯ + 𝑎 𝑥 (≤, =, ≥) 𝑏
and

x , x , … , x ≥ 0,
where the 𝑐 ′𝑠 are coefficients representing the per unit profit (or cost) of decision
variable 𝑥 to the value. The a are represent the input-output coefficients. These
represent the amount of resource, say 𝑖 consumed per unit of variable (activity) 𝑥 .
These coefficients can be positive, negative or zero. The 𝑏 represents the total
availability of the 𝑖 resource. The term resource is used in a very general sense to
include any numerical value associated with the right -hand side of a constraint. It is
assumed that b ≥ 0 for all 𝑖 . However, if any 𝑏 < 0, then both sides of constraint 𝑖 is
multiplied by −1 to make 𝑏 > 0 and reverse the inequality of the constraint.
In the general L.P. problem, the expression (≤, =, ≥) means that in any specific problem
each constraint may take only one of the three possible forms:
(i) less than or equal to (≤);
(ii) equal to (=);
(iii) greater than or equal to (≥ ).
02-01: FORMULATION OF L.P. PROBLEM (MAXIMIZATION)

Working rule for the formulation of linear programming problem or model is as
follows:
a) Identify decision variable that we seek to determine.
b) Identify objective (goal) that we need to optimize.
c) Identify constrains that the solution must satisfy.
SOLVED PROBLEMS 01
Problem 01-02-01: A company manufactures two products X and Y, which require the
following resources. The resources are the capacities machine 𝑀 , 𝑀 and 𝑀 . The
available capacities are 50, 25 and 15 hours , respectively in the planning period.
Product X requires 1 hour of machine 𝑀 and 1 hour of machine 𝑀 . Product Y requires
2 hours of machine 𝑀 , 2 hours of machine 𝑀 and 1 hour of machine 𝑀 . The profit
contribution of products X and Y are Rs.5/- and Rs.4/-, respectively. Formulate the L.P.
model that maximizes the profit.
Solution: Step I: First we need to decide the number of units of product X and Y. For
this we take help of variable x and y. Let the company manufactures x units of X and y
units of Y.
The content from the problem can be frame in table as follows:

Machines Products Availability in Hours
X Y
𝑀 0 2 50
𝑀 1 2 25
𝑀 1 1 15
Profit in Rs. Per unit 5 4
Step II: As the profit contributions of X and Y are Rs.5/ - and Rs. 4/- respectively. The
objective of the problem is to maximize the profit Z, Hence objective function is
𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍 = 5𝑥 + 4𝑦 Objective Function.
Step III: This should be done so that the utilization of machine hours by products x and
y should not exceed the available capacity. This can be shown as follows:
For Machine M 0x + 2y ≤ 50
For Machine M 1x + 2y ≤ 25 and Linear Structural Constraints.
For Machine M 1x + 1y ≤ 15
But the company can stop production of x and y or can manufacture any amount of x
and y. It cannot manufacture negative quantities of x and y. Hence, we get
both x and y are ≥ 0 Non -Negativity Constraint.
Hence, the complete L.P. model for the problem can be written as
𝑚𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍 = 5𝑥 + 4𝑦,
subject to constraints 2𝑦 ≤ 50,
𝑥 + 2𝑦 ≤ 25,
𝑥 + 𝑦 ≤ 15
where ≥ 0, 𝑦 ≥ 0.
This is the required L.P. model.
Problem 01-02-02(Production Allocation Problem): An advertising company wishes

to plan its advertising strategy in three different media-television, radio and magazines.
The purpose of advertising is to reach as large a number of potential customers as
possible. Following data have been obtained from mar ket survey:
Television Radio Magazine I Magazine II
Cost of an advertising 30,000 20,000 15,000 10,000

unit (In Rs.)

Number of potential 2,00,000 6,00,000 1,50,000 1,00,000
customers reach per unit
Number of female 1,50,000 4,00,000 70,000 50,000
customers per unit
The company want to spend no more than Rs.4,50,000 on advertising. Following are
he further requirements that must be met:
1. At least one million exposures take place among female customers.
2. Advertising on magazine be limited to Rs. 1,50,000.
3. At least three units bought on magazine I and 2 units on magazine II.
4. The number of advertising units on television and each radio should be between 5
and 10.
Formulate an L.P. model for the problem.
Solution: Step I: Let 𝑥 , 𝑥 , 𝑥 𝑎𝑛𝑑 𝑥 denote the number of advertising units to be

bought on television, radio and magazine I and magazine II , respectively.
Step II: The objective is to maximize the total number of potential customers reached .
i.e. Maximize Z = 2,00,000 𝑥 + 6,00,000 𝑥 + 1,50,000 𝑥 + 1,00,000𝑥 .
Step III: Constraints are on the advertising beget
30,000𝑥 + 20,000𝑥 + 15,000 𝑥 + 10, 000 𝑥 ≤ 4,50,000.
The number of female customers reached by an advertising campaign
1,50,000𝑥 + 4,00,000𝑥 + 70, ,000 𝑥 + 50, 000 𝑥 ≥ 10,00,000,
On expanse on magazine, I advertising:
15,000 𝑥 + 10, 000 𝑥 ≤ 1,50,000.
On no. of units on magazines: 𝑥 ≥ 55, 𝑥 ≥ 2.
On no. of units on television: 𝑥 ≥ 5 , 𝑥 ≤ 10.
On number of units on radio: 𝑥 ≥ 5, 𝑥 ≤ 10.
Where 𝑥 , 𝑥 , 𝑥 , 𝑥 ≥ 0 .
Hence, the complete LP model of given problem is
𝑚𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍 = 10 ( 2 𝑥 + 6 𝑥 + 1.5 𝑥 + 𝑥 ).
Subject to constraints 30𝑥 + 20𝑥 + 15𝑥 + 10 𝑥 ≤ 4,50,
15𝑥 + 40𝑥 + 7𝑥 + 5𝑥 ≥ 100, 15 𝑥 + 10𝑥 ≤ 150, 𝑥 ≥ 3, 𝑥 ≥ 2,

5 ≤ 𝑥 ≤ 10, 5 ≤ 𝑥 ≤ 10 , where 𝑥 , 𝑥 , 𝑥 , 𝑥 𝑒𝑎𝑐ℎ ≥ 0.
SELF-TEST 01
MCQ 02-01: Which of the following is a component of L.P. model?
a) Decision variable.
b) Objective function.
c) Constraint and non-negativity.
MCQ 02-02: The value of decision variables
a) Are under the decision maker’s control.
b) Is not under the decision maker’s control.
c) May be under the decision maker’s control.
d) May or may not be under the decision maker’s control .

SAQ 01-02-01: A company sells two different products A and B making a profit of Rs.
40 and Rs. 30 per unit, respectively. They are both produced with the help of a common
production process and are sold in two different markets. The production process has a
total capacity of 30,000 man-hours. It takes 3 hours to produce a unit A and 1 hour to
produce a unit of B. The market has been surveyed and company officials feel that the
maximum number of units of A that can be sold is 8,000 units and that of B is 12,000
units. Subject to these limitations, products can be sold in any combination. Formulate
this problem as an L.P. model to maximize profit.
Solution: Let 𝑥 , 𝑥 = number of units of products A and B to be produced ,

respectively. From the conditions given in problem, we get
𝑀𝑎𝑥 𝑍 = 40𝑥 + 30𝑥 ,
subject to 3𝑥 + 𝑥 ≤ 30,000 (Man-hours),
𝑥 ≤ 8,000 ; 𝑥 ≤ 12000 (Marketing) and 𝑥 , 𝑥 ≥ 0.
SAQ 01-02-02: A small manufacturer employs 5 skilled men and 10 semi -skilled men
and makes an article in two qualities, a deluxe model and an ordinary model. The
making of a deluxe model requires 2 hours work by a skilled man and 2 hours work by
a semi-skilled man. The ordinary model requires 1 hour work by a skilled man and 3
hours work by a semi-skilled man. By union rules no man can work more than 8 hours

per day. The manufacturer’s clear profit of the deluxe model is Rs. 10 and of the
ordinary model Rs. 8. Formulate the model of the problem.
Solution: Let 𝑥 𝑎𝑛𝑑 𝑥 are number of units of article, a deluxe model and an
ordinary model to be produced, respectively. From the conditions given in problem, we
get
Maximize 𝑍 = 10𝑥 + 8𝑥 ,
subject to 2𝑥 + 𝑥 ≤ 40,
2𝑥 + 3𝑥 ≤ 80 𝑎𝑛𝑑 𝑥 , 𝑥 ≥ 0.
02-02: FORMULATION OF L.P. MODEL (MINIMIZATION)

Working rule for the formulation of linear programming problem or model is as
follows: (Same as in maximization)
a) Identify decision variable that we seek to determine.
b) Identify objective (goal) that we need to optimize.
c) Identify the constrains that the solution must satisfy.
SOLVED PROBLEMS 02
Problem 01-02-03 (Diet Problem): A person wants to decide the constraints of a diet
which fulfill his daily requirements of proteins, fats and carbohydrates at the minimum
cost. The choice is to be made from four different types of food. The yield per unit of
these food are given in the table:
Food type Yield per unit Cost per unit (Rs.)

Proteins Fats Carbohydrates
1 3 2 6 45
2 4 2 4 40
3 8 7 7 85
4 6 5 4 65
Minimum Requirement 800 200 700
Formulate the linear programming model for problem.
Solution: Formulation of linear programming model
Step I: Let 𝑥 , 𝑥 , 𝑥 and 𝑥 denote the number of units of food type 1, 2, 3 and 4 ,
respectively.
Step II: Objective is to minimize the cost i.e.

𝑀𝑖𝑛𝑖𝑚𝑖𝑧𝑒 𝑍 = 𝑅𝑠. (45 𝑥 + 40 𝑥 + 85𝑥 + 65𝑥 ).
Step III: Constraints are on the fulfillment of the daily requirement of the various
constituents.
i.e. for proteins, 3𝑥 + 4 𝑥 + 8𝑥 + 6𝑥 ≥ 800,
for fats, 2𝑥 + 2 𝑥 + 7𝑥 + 5𝑥 ≥ 200,
for carbohydrate 6𝑥 + 4 𝑥 + 7 𝑥 + 4 𝑥 ≥ 700,
where 𝑥 , 𝑥 , 𝑥 , 𝑥 𝑒𝑎𝑐ℎ ≥ 0.
Problem 01-02-04: A company has two plants, each of which produces and supplies
two products: A and B. The plants can each work up to 16 hours a day. In plant 1, it
takes three hours to prepare and pack 1,000 gallons of A and one hour to prepare and
pack one quintal of B. In plant 2, it takes two hours to prepare and pack 1,000 gallons
of A and 1.5 hours to prepare and pack a quintal of B. In plant 1, it costs Rs . 15,000 to
prepare and pack 1,000 gallons of A and Rs. 28,000 to prepare and pack a quintal of B,
whereas in plant 2 these costs are Rs. 18,000 and Rs. 26,000, respectively. The
company is obliged to produce daily at least 10 thousand gallons of A and 8 quintals of
B.
Formulate this problem as an L.P. model to find out as to how the company
should organize its production so that the required amounts of the two products be
obtained at the minimum cost.
Solution: The data of the problem is summarized as follows:
Resources Product Total

Availability(hrs)
Plant A B
Preparation time 1 3 hrs/thousand 1 hr/quintal 16
(hrs) gallons
2 2 hrs/thousand 1.5 hr/quintal 16
gallons
Minimum daily 10 thousand 8 quintals
production gallons
Cost of production 1 15,000/thousand 28,000/quintals

(Rs.) gallons
2 18,000/thousand 26,000/quintals
gallons
Let 𝑥 , 𝑥 = quantity of product A (in thousand gallons) to be produced in plant 1 and
2, respectively.

𝑥 , 𝑥 = quantity of product B (in quintals) to be produced in plant 1 and 2,
respectively.
Then, the linear programming model is
𝑀𝑖𝑛𝑖𝑚𝑖𝑧𝑒 𝑍 = 15,000𝑥 + 18,000𝑥 + 28,000𝑥 + 26,000𝑥
subject to the constraints.
Preparation time 3𝑥 + 𝑥 ≤ 16,
2𝑥 + 1.5𝑥 ≤ 16.
Minimum daily production requirement 𝑥 + 𝑥 ≥ 10,
𝑥 + 𝑥 ≥ 8,
and 𝑥 , 𝑥 , 𝑥 , 𝑥 𝑒𝑎𝑐ℎ ≥ 0.
SELF-TEST 02
MCQ 02-03: A constraint in an L.P. model restricts
a) Value of objective function.
b) Value of a decision variable.
c) Use of the available resource.
MCQ 02-04: Constraints in an L.P. model represents
a) Limitations.
b) Requirements.
c) Balancing limitations and requirements.

SAQ 01-02-03: A firm produces an alloy having following significance:
1. Specific gravity ≤ 0.98,
2. Chromium ≥ 8%,
3. Melting point ≥ 450 ℃,
Raw materials A, B and C having the properties shown in the table can be used to make
the alloy.
Properties Properties of raw material

A B C
Specific gravity 0.92 0.97 1.04
Chromium 7% 13% 16%
Melting point 440 ℃ 490 ℃ 480 ℃
Cost of the various row material per ton are: Rs. 90 for A, Rs. 280 for B and Rs. 40 for
C. Formulate the L.P, model to find proportion in which A, B and C be used to obtain
an alloy of desired proportion while the cost of raw material is minimum.
Solution: Let 𝑥 , 𝑥 and 𝑥 be the proportion used of row materials A, B and C

Minimize 𝑍 = 90𝑥 + 280𝑥 + 40 x ,
subject to 0.92𝑥 + 0.97𝑥 + 1.04 𝑥 ≤ 0.98,
7𝑥 + 13𝑥 + 16 𝑥 ≥ 8, 440𝑥 + 490𝑥 + 480𝑥 ≥ 450,
where 𝑥 , 𝑥 , x each ≥ 0.
SAQ 01-02-04: A firm manufacture pain reliving pills in two sizes A and B. Size A
contain 4 grains of element X, 7 grains of element Y and 2 grains of element Z. Size B
contains 2 grains of element X, 10 grains of element Y and 8 grains of element Z. It is
found by user that it requires at least 12 grains of element X, 74 grains of element Y
and 24 grains of element Z to provide the immediate relief. It is required to determine
the least number of pills a patient should take to get immediate relief. Formulate the
problem as standard L.P.P.
Solution: Let 𝑥 𝑎𝑛𝑑 𝑦 are the number of pills a patient should take of size A and B,
Minimize 𝑍 = 𝑥 + 𝑦,
subject to constraint 4𝑥 + 2𝑦 ≥ 12, 7𝑥 + 10𝑦 ≥ 74,
2𝑥 + 8𝑦 ≥ 24 and 𝑥 ≥ 0 , 𝑦 ≥ 0.
02-03: GRAPHICAL METHOD OF L.P.P. SOLUTION

Once the problem is formulated as mathematical model, the next step is to solve
the problem to get an optimal solution. In graphical method, the inequalities (structural
constraints) are taken into consideration to be equations. This is because, one can't
draw a graph for inequality. Only two variable problems are taken into consideration,
because we will draw immediately lines in 2-dimensional plane (X- axis and Y-axis).
Moreover as we have non-negativity constraint with inside the problem, this is all of
the decision variables need to have positive values always the solution to the problem

lies in first quadrant of the graph. Sometimes, the value of variables can also
additionally fall in quadrants other than the first quadrant. In such cases, the line
becoming a member of the values of the variables must be extended in to the first
quadrant. The system of the method could be defined in detail while solving a
numerical problem. The characteristics of graphical technique are:
(i) Generally, the method is used to solve the problem, while it involves two
decision variables.
(ii) For three or more decision variables, the graph deals with planes and requires
high imagination to identify the answer area.
(iii) Always, the solution to the problem lies in first quadrant.
(iv) This method provides a basis for understanding the other methods of solution.
The graphical method consists of two methods
a) Corner (extreme) point method,
b) Iso-profit (Cost) method.
Here, we will discuss the only Corner point method.
Working rule for corner point method is as follows:
i. First formulate the L.P. Model. (If it is given in theoretical)
ii. Draw x and y (or 𝑥 and 𝑥 ) axis. The non-negativity 𝑥 ≥ 0 𝑎𝑛𝑑 𝑦 ≥ 0 implies
that the values of variable 𝑥 and 𝑦 lies in the first quadrant only.
iii. Plot each constraint on the graph for this considers constraint in the equality
form.
iv. Identity the feasible region (or solution space) that satisfies all the constraints
simultaneously. For ≥ type constraint, the area on or above the constraint line,
i.e. away from the origin will be considered. For ≤ type constraint, the area on
or below the constraint line, i.e. towards the origin will be considered. The area
common to all constraint is called the feasible region or convex polygon, which
we show by shaded region. This set is convex set.
v. Now, we have to start corner (extreme) point method. For this , find co-
ordinate of each corner points of the feasible region.
vi. Then, find the value of objective function at each corner point.

vii. The point at which objective function is maximum/minimum is the optimal
point and its co-ordinate gives the optimal solution.
SOLVED PROBLEMS 03
Problem 01-02-05: Find the maximum value 𝑍 = 2𝑥 + 3𝑦 subject to 𝑥 + 𝑦 ≤ 30, 𝑦 ≥ 3,
𝑦 ≤ 12, 𝑥 − 𝑦 ≥ 0 , 0 ≤ 𝑥 ≤ 20.
Solution: Now, we draw graph for this, we need to construct the table as follows:
Inequality Equations Points on X-axis Point on Y-axis Direction

𝑥 + 𝑦 ≤ 30 𝑥 + 𝑦 = 30 (30,0) (0,30) Towards origin
𝑦≥3 𝑦=3 - (0,3) Opposite to origin
𝑦 ≤ 12 𝑦 = 12 - (0,12) Towards origin
𝑥−𝑦 ≥0 𝑥−𝑦=0 (0,0) (5,5) Opposite to origin
𝑥 ≤ 20 𝑥 = 20 (20,0) - Towards origin
The solution space satisfying the given constraint and meeting the non -
negativity restriction 𝑥 ≥ 0 𝑎𝑛𝑑 𝑦 ≥ 0 is shown shaded in figure. Any point in this
region is a feasible solution to the given problem.
The co-ordinate of the corner vertices of the convex region ABCDE are A(3,3),
B(12,12), C(18,12), D(20,10) and E(20,3).
Now, we have to find values of the objective function 𝑍 = 2𝑥 + 3𝑦 at these vertices
𝑍(𝐴) = 𝑍(3, 3) = 2 × 3 + 3 × 3 = 15,
𝑍(𝐵) = 𝑍(12 , 12) = 2 × 12 + 3 × 12 = 60,
𝑍(𝐶) = 𝑍(18, 12) = 2 × 18 + 3 × 12 = 72,
𝑍(𝐷) = 𝑍(20, 10) = 2 × 20 + 3 × 10 = 70,
𝑍(𝐸) = 𝑍(20, 3) = 2 × 20 + 3 × 3 = 49.
We observe that the maximum value of Z is 72 at point C (18, 12), the solution to the
given problem is 𝑥 = 18, 𝑦 = 12 and 𝑍 = 72.

Problem 01-02-06: Use the graphical method to solve the following L.P. problem.
Minimize 𝑍 = 3𝑥 + 2𝑥 subject to the constraints 5𝑥 + 𝑥 ≥ 10, 𝑥 + 𝑥 ≥ 6, 𝑥 +

4𝑥 ≥ 12 𝑎𝑛𝑑 𝑥 , 𝑥 ≥ 0.
Solution: Now, we draw graph, for this we need to construct the table as follows:
Inequality Equations Points on Point on 𝑥 Direction

𝑥 -axis -axis
5𝑥 + 𝑥 ≥ 10 5𝑥 + 𝑥 = 10 (2,0) (0,10) Opposite to origin
𝑥 + 𝑥 ≥ 6, 𝑥 + 𝑥 = 6, (6,0) (0,6) Opposite to origin
𝑥 + 4𝑥 ≥ 12 𝑥 + 4𝑥 = 12 (12,0) (0,3) Opposite to origin
negativity restriction 𝑥 ≥ 0 𝑎𝑛𝑑 𝑥 ≥ 0 is shown shaded in figure. Any point in this
The co-ordinate of the corner vertices of the feasible region ABCD ( bounded
from below) are A(12,0), B(4,2), C(1,5) and D(0,10) .
Now, we have to find values of the objective function 𝑍 = 3𝑥 + 2𝑥 at these vertices
𝑍(𝐴) = 𝑍(12, 0) = 3 × 12 + 2 × 0 = 36,
𝑍(𝐵) = 𝑍(4 , 2) = 3 × 4 + 2 × 2 = 16,
𝑍(𝐶) = 𝑍(1, 5) = 3 × 1 + 2 × 5 = 13,
𝑍(𝐷) = 𝑍(0, 10) = 3 × 0 + 2 × 10 = 20.
We, observe that the minimum (optimal) value of Z is 13 at point C(1,5), the
solution to the given problem is 𝑥 = 1, 𝑥 = 5 𝑎𝑛𝑑 𝑍 = 13.
SELF-TEST 03
MCQ 02-05: The limitation of the graphical method of solving L .P. model is
a) It is useful for only two decision variable L.P. model.

b) It is useful for only three decision variable L .P. model.
c) It is useful for finitely many decision variable L.P. model
d) It is useful for only for maximization problem.
MCQ 02-06: Optimal solution of L.P. problem is
a) Feasible solution.
b) Feasible solution and satisfies non-negative condition.
c) Feasible solution and optimize objective function .
d) Feasible solution and satisfies non-positive condition.

SAQ 01-02-05: Use the graphical method to solve the following L.P. problem:
Maximize 𝑍 = 2𝑥 + 𝑥 subject to the constraints 𝑥 + 2𝑥 ≤ 10, 𝑥 + 𝑥 ≤ 6,
𝑥 – 𝑥 ≤ 2, 𝑥 – 2𝑥 ≤ 1 and 𝑥 , 𝑥 ≥ 0.
Solution: The solution space satisfying the given constraint and meeting the non-negativity
restriction 𝑥 ≥ 0 𝑎𝑛𝑑 𝑥 ≥ 0 is shown shaded in figure. Any point in this region is a feasible
solution to the given problem.
From figure, we see that the polygon ABCDEF shows the feasible region which made
shadow.
We get the co-ordinates of the points A, B, C, D, E and F. After finding the value of
𝑍 = 2𝑥 + 𝑥 at each point we get that Z is maximum at point D (4, 2) and hence, we get the
solution
𝑥 = 4, 𝑥 = 2 𝑎𝑛𝑑 𝑀𝑎𝑥 𝑍 = 10.

SAQ 01-02-06: Use the graphical method to solve the following L .P. problem:
Minimize 𝑍 = 3𝑥 + 2𝑦, subject to constraints 8𝑥 + 𝑦 ≥ 8, 2𝑥 + 𝑦 ≥ 6 𝑥 + 3𝑦 ≥ 6,
𝑥 + 6𝑦 ≥ 8, 𝑥 ≥ 0, 𝑦 ≥ 0.
Solution: The solution space satisfying the given constraint and meeting the non -
negativity restriction 𝑥 ≥ 0 𝑎𝑛𝑑 𝑦 ≥ 0 is shown shaded in figure. Any point in this
From figure, we see that the polygon ABCDE shows the feasible region which made
shadow.
We get the co-ordinates of the points A, B, C, D and E. After finding the value of
𝑍 = 3𝑥 + 2𝑦 at each point we get that Z is minimum at point C (2.4, 1.2) and hence, we
get the solution
𝑥 = 2.4, 𝑦 = 1.2 𝑎𝑛𝑑 𝑀𝑖𝑛 𝑍 = 9.6.
02-04: SOME EXCEPTIONAL CASES IN L.P.P.

In the previous section, we solved the L.P. problem which has unique optimal
solution this is because the objective function line passed only through the extreme
point located at the intersection of the two constraints lines. But it is not necessary for
every L.P. problem. In, general, a L.P. problem may have
a) A definite and unique optimal solution.
b) An infinite number of optimal solutions (multiple optimal solution).
c) An unbounded solution.
d) No feasible solution (infeasible solution).
First type is covered in the last section. Now, we have to discuss solution of the
multiple optimal solution, an unbounded solution and infeasible solution.

A) Multiple Optimal Solutions:
We have seen that the optimal solution of any linea r programming problem occurs at
corner point of the feasible region and that the solut ion is unique, i.e. no other solution
yields the same value of the objective function. However, in certain cases, a given L .P.
problem may have more than one solution yielding the same optimal objective function
value. Each of such optimal solutions is termed as alternate optimal solution. If there
are optimal solution at two points then all points on joining these two points are the
solutions of L.P. problem. There are two conditions that should be satisfied for
multiple optimal solutions to exist:
(i) The slope of the objective function should be the same as that of the constraint
forming the boundary of the feasible solutions region .
(ii) The constraint should form a boundary on the feasible region in the direction of
optimal movement of the objective function. In other words, the constraint
should be an active constraint. Geometrically active constraint is one that passes
through one of the extreme points of feasible solution space .
Remark: If slope of a constraint is parallel to the slope of the objectiv e function, but it
does not fall on the boundary of the feasible region, the multiple solutions will not
exist. This type of a constraint is called redundant constraint (a constraint whose
removal does not change the feasible region) .
B) Unbounded Solution:
It exists when an L.P. problem has no limit on the constraint i.e. the feasible region
is not bounded in any respect. In this case, if we consider decision variable in
unbounded region, it their value/ values can increase infinitely. Hence value of
objective function may be finite or infinite. If the value of objective function in this
case is infinite, the problem is said to have unbounded solution. In general, an
unbounded L.P. problem solution exists due to improper formulation of the real -life
problem.
C) Infeasibility Condition:
This condition is arising when there is no solution to L .P. problem which satisfies
all constraints and nonnegative restrictions.
SOLVED PROBLEMS 04
Problem 01-02-07: Maximize 𝑍 = 5𝑥 + 4𝑥 subject to constraint
𝑥 − 2𝑥 ≤ 1, 𝑥 + 2𝑥 ≥ 3 𝑎𝑛𝑑 𝑥 ≥ 0, 𝑥 ≥ 0.

Solution: Now, we draw graph, for this, we need to construct the table as follows:
Inequality Equations Points on Point on Direction

𝑥 -axis 𝑥 -axis
𝑥 − 2𝑥 ≤ 1 𝑥 − 2𝑥 = 1 (1,0) (0, -0.5) Towards the origin
𝑥 + 2𝑥 ≥ 3 𝑥 + 2𝑥 (3,0) (0, 1.5) Opposite to origin

=3
The solution space satisfying the given constraint and meeting the non -negativity
restriction 𝑥 ≥ 0 𝑎𝑛𝑑 𝑥 ≥ 0 is shown shaded in figure.
Any point in this region is a feasible solution to the given problem. The shaded
region in this problem is unbounded from above. The corners points of the shaded
region are A(0, 1.5) and B(2, 0.5). Then, values of objective function 𝑍 = 5𝑥 + 4𝑥 are
𝑍(𝐴) = 𝑍(0,1.5) = 5(0) + 4(1.5) = 6.
𝑍(𝐵) = 𝑍(2, 0.5) = 5(2) + 4(0.5) = 12 .
Since the given L .P. problem is of maximization, there exist a number of points in
the shaded region for which the value of the objective function is more than 12. For
example, the point (2, 2) lies in the region and the objective function value at this point
is Z (2, 2) = 18, which is more than 12. Thus, as value of variables 𝑥 and 𝑥 increases
arbitrarily large, the value of Z also starts increasing. Hence, the L.P. problem has an
unbounded solution.
Problem 01-02-08: Use the graphical method to solve the following LP problem.
𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍 = 10𝑥 + 6𝑥 subject to the constraints
5𝑥 + 3𝑥 ≤ 30, 𝑥 + 2𝑥 ≤ 18 and each 𝑥 , 𝑥 ≥ 0.
Solution: Now, we draw graph, for this, we need to construct the table as follows:

Constraint Equations Point on Point on Direction
𝑥 -axis 𝑥 -axis
5𝑥 + 3𝑥 ≤ 30 5𝑥 + 3𝑥 = 30 (6, 0) (0, 10) Towards origin
𝑥 + 2𝑥 ≤ 18 𝑥 + 2𝑥 = 18 (18, 0) (0, 9) Towards origin
negativity restriction 𝑥 ≥ 0 𝑎𝑛𝑑 𝑥 ≥ 0 is shown shaded in figure. Any point in this
The co-ordinate of the corner vertices of the feasible region OABC (bounded from
below) are A (0,9), B (6/7, 60/7), C (6,0) and O (0,0).
Now, we have to find values of the objective function 𝑍 = 10𝑥 + 6𝑥 at these

vertices
𝑍(𝐴) = 𝑍(0, 9) = 10 × 0 + 6 × 9 = 54.
𝑍(𝐵) = 𝑍 ( , ) = 10 × + 6 × = 60.
𝑍(𝐶) = 𝑍(6, 0) = 10 × 6 + 6 × 0 = 60 .
𝑍(𝑂) = 𝑍(0, 0) = 10 × 0 + 6 × 0 = 0.
We observe that, the maximum (optimal) value of Z is 60 at points two different

points B (6/7,60/7) and C(6,0) Therefore two alternate solution to the given problem
𝑥 = ,𝑥 = 𝑎𝑛𝑑 𝑥 = 6, 𝑥 = 0 .Thus, the give L.P. model has multiple optimal
solutions.
SELF-TEST 04
MCQ 02-07: A redundant constraint is a constraint
a) Whose removal change the feasible region.
b) Whose removal does not change the feasible region.

c) Fall on the boundary of the feasible region.
d) Its slope is not equal to the slope of objective function.
MCQ 02-08: L.P. model has an unbounded solution if
a) Feasible region is unbounded with any respect and objective value is infinite.
b) Feasible region is bounded.
c) Objective function value is finite.
d) Objective function value is 0.

SAQ 01-02-07: 𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍 = 3𝑥 + 4𝑥 subject to the constraints
𝑥 − 𝑥 ≥ 0 , 2.5 𝑥 − 𝑥 ≤ −3 and each 𝑥 , 𝑥 ≥ 0
Solution: Draw the graph from the given constraint.
Any point within this region satisfies constraints but not non -negativity restrictions.
Hence, infeasible solution to the given L.P.P.
SAQ 01-02-08: 𝑀𝑎𝑥 𝑍 = 4𝑥 + 4𝑥 subject to
𝑥 + 2𝑥 ≤ 10, 6𝑥 + 6𝑥 ≤ 36 , 𝑥 ≤ 6 𝑎𝑛𝑑 𝑒𝑎𝑐ℎ 𝑥 , 𝑥 ≥ 0.
Solution: We draw the graph first. Then, we get co-ordinate of the corner vertices of
the feasible region ABCD are A (0,0), B (6, 0), C (2, 4) and D(0, 5) .
We observe that, the maximum (optimal) value of Z is 24 at points two different

points B (6, 0) and C (2,4). Therefore two alternate solutions to the given problem
𝑥 = 6, 𝑥 = 0 𝑎𝑛𝑑 𝑥 = 2, 𝑥 = 4 . Thus, the give L.P. model has multiple optimal
solutions.

SUMMARY
Three important steps are important in construction of L .P. problem. These are
the Identify decision variable that we seek to determine, identify objective (goal) that
we need to optimize and identify the constrains that the solution must satisfy.
The graphical solution approaches (or methods) provide a conceptual basis for
solving large and complex L.P. problems. A graphical method is used to reach an
optimal solution to and L.P. problem that has a number of constraints binding the
objective function. Both corner point methods and the iso -profit (or cost) function line
method are used for graphically solving any L.P. problem that has only two decision
variables. The optimal solution of the L.P.P. is at one of the corner points. Also, we
discuss the L.P. problem has unbounded solution if the feasible region is not bounded
and objective value is infinite. If the L.P. problem have more than one solution then it
has infinitely many solutions.

1. Maximize 𝑍 = −𝑥 + 2𝑥 subject to constraint
𝑥 − 𝑥 ≤ −1, 0.5𝑥 + 𝑥 ≤ 2 𝑎𝑛𝑑 𝑥 ≥ 0, 𝑥 ≥ 0.
2. 𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍 = 20𝑥 + 10𝑥 subject to the constraints
𝑥 + 2𝑥 ≤ 40, 3𝑥 + 𝑥 ≥ 30 4𝑥 + 3𝑥 ≥ 60 and each 𝑥 , 𝑥 ≥ 0
3. Minimize 𝑍 = 60𝑥 + 40𝑥 subject to constraint
30𝑥 + 10𝑥 ≥ 240, 10𝑥 + 10𝑥 ≥ 160, 20𝑥 + 60𝑥 ≥ 480 𝑎𝑛𝑑 𝑥 ≥ 0, 𝑥 ≥ 0.
4. 𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍 = 𝑥 + 𝑦 subject to constraint
5𝑥 + 10𝑦 ≤ 50, 𝑥 + 𝑦 ≥ 1, 𝑦 ≤ 4 𝑎𝑛𝑑 𝑥 ≥ 0, 𝑦 ≥ 0

5. 𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍 = 4𝑥 + 5𝑦 subject to constraint
𝑥 + 𝑦 ≥ 1, −2𝑥 + 𝑦 ≤ 1, 4𝑥 − 𝑦 ≥ 1, 𝑥 ≥ 0, 𝑦 ≥ 0.
6. A manufacturer produces two different models X and Y of the same product.

Model X makes a contribution of Rs 50 per unit and model Y, Rs 30 per unit,
towards total profit. Raw materials 𝑟 and 𝑟 are required for production. At least
18 kg of 𝑟 and 12 kg of 𝑟 must be used daily. Also, at most 34 hours of labor are
to be utilized. A quantity of 2 kg of 𝑟 is needed for model X and 1 kg of 𝑟 for
model Y. For each of X and Y, 1 kg of 𝑟 is required. It takes 3 hours to
manufacture model X and 2 hours to manufacture model Y. How many units of
each model should be produced in order to maximize the profit?
7. A person requires 10, 12 and 12 units of chemical A, B and C , respectively for his
gardens. A liquid product 5, 2 and 1 units of A, B and C per gar. A dry product
contains 1, 2 and 4 units of A, B and C per carton. If the liquid product sells for
Rs. 3 per jar dry product sells Rs. 2 per carton. Formulate the above problem as a
L.P.P. to minimize the cost?
8. The cost of materials A and B is Re.1/- per unit, respectively. We have to

manufacture an alloy by mixing these to materials. The process of preparing the
alloy is carried out on three facilities X, Y and Z. Facilities X and Z are machines,
whose capacities are limited. Y is a furnace, where heat treatment takes place and
the material must use a minimum given time (even if it uses more than the
required, there is no harm). Material A requires 5 hours of machine X and it does
not require processing on machine Z. Material B requires 10 hours of machine X
and 1 hour of machine Z. Both A and B are to be heat treated at last one hour in
furnace Y. The available capacities of X, Y and Z are 50 hours, 1 hour and 4 hours,
respectively. Find how much of A and B are mixed so as to minimize the cost.
Formulate the above problem as a L.P.P. and solve it by using graphical method.
KEY WORDS
Linear Programming Problem (L.P.P.), optimal solution, maximization, minimization.
REFERENCES
BOOKS
& COMPANY LTD

PRAGATI EDITION
MOOCS
YOUTUBE VIDEOS
1) https://youtu.be/xrGVe6gMRyk
2) https://youtu.be/8IRrgDoV8Eo
3) https://youtu.be/34AHMgmjw10
4) https://youtu.be/_Y95eD3P2Mw
WIKIPEDIA
OER

UNIT 01-03: THE SIMPLEX METHOD
LEARNING OBJECTIVES
 Understand the meaning of the word ‘simplex’ and logic of using simplex
method.
 Convert an L.P. problem into its standard form by adding slack, surplus and/or
artificial variables.
 Set-up simplex tables and solve L.P. problems using the simplex algorithm.
 Interpret the optimal solution of L.P. problems.
 Recognize special cases such as degeneracy, multiple optimal solution,

unbounded and infeasible solutions.
INTRODUCTION
In previous section, we solved linear programming problems graphically. Since
we can only easily graph with two variables (x and y), this approach is not practical for
problems where there are more than two variables involved. To solve linear
programming problems in three or more variables, we will use somet hing called “The
Simplex Method”.
03-01: THE STANDARD FORM OF L.P. PROBLEM

After formulation L.P .model, we have to find the solution in the next step. But
before using any analytical method is used to obtain the solution, the problem must be
available in the particular forms like canonical form and Standard form. The canonical
form is useful in duality theory and standard form is used to develop the general
procedure for solving any L.P.P. We discuss canonical form in next section. Now, we
discuss standard form in this section which is useful in the simplex method.
The standard form of the L.P. problem is expressed as:
Maximum (or minimum) = ∑ 𝑐𝑥 ,
subject to ∑ 𝑎 𝑥 = 𝑏 , 𝑖 = 1, 2, 3, … , 𝑚.
𝑥 ≥ 0, 𝑗 = 1, 2, … , 𝑛.
and 𝑏 ≥ 0.
In the matrix-vector form, the standard form of the general L.P. problem can be
expressed as
Maximum (or minimum) 𝑍 = 𝒄𝒙
subject to Ax= 𝒃,
𝒙 ≥ 𝟎,
and 𝒃 ≥ 𝟎.
where, 𝒄 = (𝑐 , 𝑐 , . . . , 𝑐 ) is the row vector,𝒙 = (𝑥 , 𝑥 , . . . , 𝑥 ) ,
𝒃 = (𝑏 , 𝑏 , . . . , 𝑏 ) and 𝒔 = (𝑠 , 𝑠 , . . . , 𝑠 ) are column vectors, and A is the 𝑚 × 𝑛

matrix of coefficients, 𝑎 of variables 𝑥 , 𝑥 , . . . , 𝑥 in the constraints.
The standard form of the L.P. problem should have the following characteristics:
1) All variables are non-negative.
2) The right-hand side of each constraint should be made non-negative.
3) The objective function may be of the maximization or minimization type.
4) All constraint is expressed as equations.
Now, we see rule how to convert nonstandard LP problem to standard form.
i) Replace each unrestricted variable (if any) with the difference of the two non-
negative variables.
ii) Check whether all the 𝑏 (𝑖 = 1, 2, . . . , 𝑚) values are positive. If any one of them is
negative, then multiply the corresponding constraint by –1 in order to make 𝑏 > 0.
In doing so, remember to change a ≤ type constraint to a ≥ type constraint, and
vice versa.
iii) The inequality constraints are change to equality constraints by adding or

subtracting a non-negative variable from the left sides of such constraints. These
new variables called slack variable or simply slacks and they are always non -
negative. If the constraints are ≤ type, the slack variables are added in the left-hand
side of constraint. If the constraints are ≥ type, the slack variables are subtracted in
the left-hand side of constraint and it is commonly known as surplus variable. The
decision variables as well as slack and surplus variables are called the admissible
variables.
Such an L.P. problem formed after the introduction of slack variable or surplus variable
is called reformulated L.P. problem.
Remark: A slack variable represents an unused resource, either in the form of time on a
machine, labor hours, money, warehouse space, etc. Since these variables don’t yield

any profit, therefore such variables are added to the objective functi on with zero
coefficients.
A surplus variable represents the amount by which solution values exceed a

resource. These variables are also called negative slack variables. Surplus variables,
like slack variables carry a zero coefficient in the objective func tion.
SOLVED PROBLEMS 01
Problem 01-03-01: Express the following L.P. problem in standard form.
𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍 = 7𝑥 + 5𝑥 subject to 2𝑥 + 3𝑥 ≤ 20, 3𝑥 + 𝑥 ≥ 10, 𝑥 ≥ 0, 𝑥 ≥ 0.
Solution: Introduction slack 𝑠 and surplus 𝑠 variable, the problem can be expressed in
standard form as
𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍 = 7𝑥 + 5𝑥 + 0𝑠 + 0𝑠
subject to 2𝑥 + 3𝑥 + 𝑠 + 0𝑠 = 20,
3𝑥 + 𝑥 + 0𝑠 − 𝑠 = 10,
𝑥 , 𝑥 , 𝑠 , 𝑠 ≥ 0.
Problem 01-03-02: Reformulate the problem into standard form:
𝑀𝑖𝑛𝑖𝑚𝑖𝑧𝑒 𝑍 = 4𝑥 + 𝑥 − 3𝑥 subject to 𝑥 + 5𝑥 − 3𝑥 ≤ −20,
2𝑥 + 7𝑥 + 2𝑥 ≤ 10 , 𝑥 − 𝑥 − 3𝑥 ≥ 3, 𝑥 ≥ 0.
Solution: If we observed that the variable 𝑥 𝑎𝑛𝑑 𝑥 are unrestricted variable. Let us
consider 𝑥 = 𝑦 , 𝑥 = 𝑦 − 𝑦 𝑎𝑛𝑑 𝑥 = 𝑦 − 𝑦 , where 𝑦 , 𝑦 , 𝑦 , 𝑦 ≥ 0 . Hence, given
problem can be expressed as
𝑀𝑖𝑛𝑖𝑚𝑖𝑧𝑒 𝑍 = 4𝑦 + 𝑦 − 𝑦 − 3𝑦 + 3𝑦
subject to 𝑦 + 5𝑦 − 5𝑦 − 3𝑦 + 3𝑦 ≤ −20,
2𝑦 + 7𝑦 − 7𝑦 + 2𝑦 − 2𝑦 ≤ 10.
𝑦 − 𝑦 + 𝑦 − 3𝑦 + 3𝑦 ≥ 3.
𝑦 , 𝑦 , 𝑦 , 𝑦 , 𝑦 ≥ 0.
If observe that the first constraint right side is negative, hence multiply it by -1 for
making right side positive. We get
−𝑦 − 5𝑦 + 5𝑦 + 3𝑦 − 3𝑦 ≥ 20
Now, introduce slack variable 𝑠 and surplus variables 𝑠 , 𝑠 problem can be expressed
as in the standard form

𝑀𝑖𝑛𝑖𝑚𝑖𝑧𝑒 𝑍 = 4𝑦 + 𝑦 − 𝑦 − 3𝑦 + 3𝑦 + 0𝑠 + 0𝑠 + 0𝑠 + 0𝑠
subject to −𝑦 − 5𝑦 + 5𝑦 + 3𝑦 − 3𝑦 − 𝑠 = 20,
2𝑦 + 7𝑦 − 7𝑦 + 2𝑦 − 2𝑦 + 𝑠 = 10,
𝑦 − 𝑦 + 𝑦 − 3𝑦 + 3𝑦 − 𝑠 = 3,
𝑦 , 𝑦 , 𝑦 , 𝑦 , 𝑦 , 𝑠 , 𝑠 , 𝑠 ≥ 0.
SELF-TEST 01
MCQ 03-01: Which of the following statement is false about standard L.P. problem?
a) All decision variables are non-negative.
b) The right-hand side of each constraint is positive.
c) All constraints are expressed as equations.
d) Objective function may be of the maximizati on or minimization type.
MCQ 03-02: Surplus variable is
a) Added in a constraint if it is of the type ≥.
b) Added in a constraint if it is of the type ≤.
c) Subtracted in a constraint if it is of the type ≥.
d) Subtracted in a constraint if it is of the type ≤.

SAQ 01-03-01: Reformulate the following L.P. model on to the standard form.
𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍 = 5𝑥 + 3𝑥 + 4𝑥
subject to 2𝑥 − 5𝑥 ≤ 6, 2𝑥 + 3𝑥 + 𝑥 ≥ 5,
3𝑥 + 4𝑥 ≤ 3, 𝑥 , 𝑥 , 𝑥 ≥ 0.
Solution: Introduce the slack and surplus variable, we get
𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍 = 5𝑥 + 3𝑥 + 4𝑥 + 0𝑠 + 0𝑠 + 0𝑠
subject to 2𝑥 − 5𝑥 + 𝑠 = 6, 2𝑥 + 3𝑥 + 𝑥 − 𝑠 = 5,
3𝑥 + 4𝑥 + 𝑠 = 3, 𝑥 , 𝑥 , 𝑥 , 𝑠 , 𝑠 , 𝑠 ≥ 0.
SAQ 01-03-02: Reformulate the following L.P. model on to the standard form.
𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍 = 2𝑥 + 5𝑥 subject to 𝑥 + 2𝑥 − 𝑥 = −4, 3𝑥 + 4𝑥 − 𝑥 = −7,
𝑥 + 2𝑥 + 𝑥 = 9, 5𝑥 − 2𝑥 + 𝑥 = 9 , 𝑥 , 𝑥 , 𝑥 , 𝑥 ≥ 0 .

Solution: Introduce the slack and surplus variable for standard form.
Here, 𝑥 , 𝑥 , 𝑥 , 𝑥 are slack variables and 𝑥 , 𝑥 are unrestricted.
𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍 = 2𝑥 + 5𝑥 + 0𝑥 + 0𝑥 + 0𝑥 + 0𝑥 subject to
𝑥 − 𝑥 − 2𝑥 + 2𝑥 + 𝑥 = 4, −3𝑥 + 3𝑥 − 4𝑥 + 4𝑥 + 𝑥 = 7,
𝑥 − 𝑥 + 2𝑥 − 2𝑥 + 𝑥 = 9, 5𝑥 − 2𝑥 + 𝑥 = 9 ,
each 𝑥 , 𝑥 , 𝑥 , 𝑥 , 𝑥 , 𝑥 , 𝑥 , 𝑥 ≥ 0 .
03-02: THE SIMPLEX METHOD (TECHNIQUE OR ALGORITHM)

Here, we first define some important definitions. Consider the general linear
problem involving 𝑛 variables and 𝑚 constraints ( 𝑚 ≤ 𝑛) . Determine the values of
𝑥 , 𝑥 , . . , 𝑥 which
Maximum 𝑍 = ∑ 𝑐𝑥 ,
subject to ∑ 𝑎 𝑥 ≤ 𝑏 , 𝑖 = 1, 2, 3, … , 𝑚
𝑥 ≥ 0, 𝑗 = 1, 2, … , 𝑛 .
Introduce the slack variables 𝑥 ,𝑥 ,..,𝑥 in the constraint, it can be put in

the following standard form:
Maximum 𝑍 = ∑ 𝑐𝑥 , …..(1)
subject to
∑ 𝑎 𝑥 +𝑥 = 𝑏 , 𝑖 = 1, 2, 3, … , 𝑚. ….(2)
where 𝑥 ≥ 0, 𝑗 = 1, 2, … , 𝑛 + 𝑚. …..(3)
Definitions:
 Solution: A set of variables [𝑥 , 𝑥 , … , 𝑥 ] is called a solution to L.P. problem

if it satisfies the constraint (2).
 Feasible Solution: A set of variables [𝑥 , 𝑥 , … , 𝑥 ] is called a feasible

solution to L.P. problem if it satisfies the constraint (2) as well as non -negative
restrictions (3).
 Basic Solution: A solution obtained by setting any 𝑛 variables (among 𝑚 + 𝑛

variables) equal to zero and solving for remaining 𝑚 variables (provided the
determinant of the coefficient of the 𝑚 variable is non-zero.) is called the basic
solution. These 𝑚 variable (some of they may be zero) are called basic

variables and the remaining 𝑛 variables that have been put equal to zero each
are called non-basic variables.
( )
The number of basic solutions will at most be 𝐶 = , which the number of
combinations of (𝑚 + 𝑛) things taken 𝑚 at a time.
 Basic Feasible Solution (B.F.S.): It is the basic solution that also satisfies the non-
negativity restrictions (3). All variables in basic feasible solution are ≥ 0. Every
basic feasible solution of the problem is an extreme (corner) point of feasible
solutions and every extreme point is a basic feasible solution of the set of
constraints.
 Non-degenerate Basic Feasible Solution: It is the feasible solution in which all 𝑚

basic variables are positive (> 0) and the remaining 𝑛 variables are zero each.
 Degenerate Basic Feasible Solution: It is the basic feasible solutions in which one
or more of the 𝑚 basic variables are equal to zero.
 Optimal Basic Feasible Solution (O.B.F.S.): It is the basic solution that also
optimizes the objective function (1).
 Unbounded Solution: If the value of the objective function can be increased or

decreased indefinitely, the solution is called unbounded solution.
The graphical method discussed in the previous section cannot be applied when the
number of variables involved in L.P. problem is more than the tree or rather two,
since even with three variable the graphical solution become tedious as it involves
intersection of planes in three dimensions. The simplex method is developed by
Prof. George B. Dantzig, can be used to solve a ny L.P. problem (for which the
solution exists) involving any number of variable and constraint (hundreds or even
thousands).
The computational procedure in the simple method is based on the fundamental

property that, ‘the optimal solution to an L. P. problem, if it exists, occurs only at one
of the corner points of the feasible region’ . The Simplex method always starts with
initial basic feasible solution (I.B.F.S.) i.e. the origin (0.0), which is of the corner pint
of feasible region. Then, it is tested whether improvement in the value of objective
function is possible by moving to the next corner point of the feasible region. If so,
solution at this point obtained. This search for better corner point is repeated, till after
finite number of trials, the optimal solution if it exists, is obtained. This algorithm we
have to discuss with the help of some examples.

SOLVED PROBLEMS 02
Problem 01-03-03: Solve the following by using simplex method.
𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍 = 2𝑥 + 5𝑥 subject to 𝑥 + 4𝑥 ≤ 24, 3𝑥 + 𝑥 ≤ 21, 𝑥 + 𝑥 ≤ 9, where
𝑥 , 𝑥 ≥ 0.
Solution: Step 1: Express the L.P.P. in Standard form.
We have to introduce non-negative slack variable 𝑠 , 𝑠 𝑎𝑛𝑑 𝑠 for writing L.P.P. in

standard form. Then we get L.P.P. in standard form
𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍 = 2𝑥 + 5𝑥 + 0𝑠 + 0𝑠 + 0𝑠
subject to 𝑥 + 4𝑥 + 𝑠 + 0𝑠 + 0𝑠 = 24,
3𝑥 + 𝑥 + 0𝑠 + 𝑠 + 0𝑠 = 21,
𝑥 + 𝑥 + 0𝑠 + 0𝑠 + 𝑠 = 9,
where 𝑥 , 𝑥 , 𝑠 , 𝑠 , 𝑠 ≥ 0.
where the slack variable can be treated as imaginary produ cts, contributing zero profits.
Accordingly, they are assigned zero coefficients in the objective function.
Step 2: Find the Initial Basic Feasible Solution:
Assume the profit earned is zero. Since all 𝑏 (R.H.S. values) > 0, (𝑖 = 1, 2, 3),
Therefore choose the initial basic feasible solution as:
𝑥 = 𝑥 = 𝑥 = 0; 𝑠 = 24, 𝑠 = 21, 𝑠 = 9 𝑎𝑛𝑑 𝑍 = 0.
The problem in the standard for and the solution obtained above are now expressed in
the form of table, called the simplex table.
Contribution/ unit 𝑐 basis 2 5 0 0 0
Basic Basic Basic

Body Identity Matrix
Variable Variables Variables
matrix
Coefficient Value
𝑐 B 𝑏(= 𝑥 ) 𝑥 𝑥 𝑠 𝑠 𝑠
0 𝑠 24 1 4 1 0 0
0 𝑠 21 3 1 0 1 0
0 𝑠 9 1 1 0 0 1
Interpretation of the data in the table is given below.

1. The first row indicates the coefficient 𝑐 of the variable in the objective
function.
2. The second row indicates the variable in the problem for which 𝑐 coefficients
have already been written.
3. The first (𝑐 𝑐𝑜𝑙𝑢𝑚𝑛 ) represent the coefficient of the current basic variables in
the objective function.
4. The second column is the basis column (or product mix column) which
represent the basic variables of the basic variable of the current solutio n. The
basic variables are the slack variables 𝑠 , 𝑠 𝑎𝑛𝑑 𝑠 .
5. The body matrix (also called coefficient matrix) under non -basic variable
𝑥 𝑎𝑛𝑑 𝑥 represent their coefficient 𝑎 in the constraint. These coefficients may
be positive, negative or zero.
6. The identity matrix represents the coefficient of slack variable in the constraint.
It is always a square matrix and size is determined by the number of
constraints.
7. The b-column is the third column of the table. This is also called the quantity
column or the solution column ( 𝑥 ) column. This represents the R.H.S. value of
the constraints.
Step 3: Perform Optimality test.
Now, in this step add two more in the above table we get
𝑐 2 5 0 0 0 𝑏
F.R. 𝑐 Basis 𝑥 𝑥 𝑠 𝑠 𝑠
𝑥 𝜃
0 𝑠 1 (4) 1 0 0 24 6
1/4 0 𝑠 3 1 0 1 0 21 21
1/4 0 𝑠 1 1 0 0 1 9 9
𝑍 = ∑𝑐 𝑎 0 0 0 0 0 0
𝑐 −𝑍 2 5 0 0 0
I.B.F.S.
Now, we have to check whether I.B.F.S. is found in the step second can be
improved or not. This can be done by computing 𝑐 − 𝑍 , where 𝑍 = ∑ 𝑐 𝑎 . Here,

𝑎 matrix element in the 𝑖 row and 𝑗 column. If 𝑐 − 𝑍 is positive under any

Column, the current feasible solution is not a optimal solution and at least one better
solution is possible.
𝑧 = 0 (1) + 0 (3) + 0 (1) = 0 𝑓𝑜𝑟 𝑥 − 𝑐𝑜𝑙𝑢𝑚𝑛,
𝑧 = 0 (4) + 0 (1) + 0 (1) = 0 𝑓𝑜𝑟 𝑥 − 𝑐𝑜𝑙𝑢𝑚𝑛,
𝑧 = 0 (1) + 0 (0) + 0 (0) = 0 𝑓𝑜𝑟 𝑠 − 𝑐𝑜𝑙𝑢𝑚𝑛 and so on, we get
𝑧 = 0 𝑎𝑛𝑑 𝑧 = 0, 𝑧 = 0 for column 𝑠 , 𝑠 𝑎𝑛𝑑 𝑥 , respectively.
Now, we have to find 𝑐 − 𝑧 .
𝑐 − 𝑍 = 2 − 0 = 2, 𝑐 − 𝑍 = 5 − 0 = 5, 𝑐 − 𝑍 = 0 − 0 = 0 and so on.
Since 𝑐 − 𝑍 is positive under the 𝑥 − 𝑐𝑜𝑙𝑢𝑚𝑛 and 𝑥 -𝑐𝑜𝑙𝑢𝑚𝑛 , I.B.F.S. is not

an optimal solution.
Step-4: - Iterate towards an optimal solution:
Now select key row, key column and key element. Select key row with minimum
𝑏
non-negative = 𝑎 𝑖 . If all ratios are negative or infinity, the current solution is
𝑖𝑗
unbounded and stops the computation. From the table minimum non negative = =
4. Thus 𝑠 row is the key-row. Variable 𝑥 is chosen to enter into the basis because
𝑐 – 𝑧 = 5 is the largest positive number in the 𝑥 -column, where all elements are
positive. This means that for every unit of variable 𝑥 , the objective function will
increase in value by 5. The 𝑥 -column is the key column.
The intersection of the key-row and key-column is (4), called key element. Mark,
they key row, key column and key element as shown in table. Hence, 𝑥 is the incoming
variable which replace the outgoing variable 𝑠 in the next table.
Since the key element (4) enclosed in the (.) in above table is not 1, divide all
elements of the key row by 4 in order to obtain new values of the elements in this row.
The new values of the elements in the remaining rows for the new table can be obtained
by performing the following elementary row operations on all rows so that all elements
except the key element 1 in the key column are zero.
𝑅 (𝑛𝑒𝑤) → 𝑅 (𝑜𝑙𝑑) ÷ 4 (𝑘𝑒𝑦 𝑒𝑙𝑒𝑚𝑒𝑛𝑡)
1 4 1 0 0 24 1 1
→ ( , , , , , ) = ( , 1 , , 0, 0, 6 )
4 4 4 4 4 4 4 4
𝑹 (𝒏𝒆𝒘) = 𝑹 (𝒐𝒍𝒅) − 𝑹 (𝒏𝒆𝒘)

3− = , 1 − 1 = 0, 0 − = − , 1 − 0 = 1, 0 − 0 = 0 , 21 − 6 = 15
1− = , 1 − 1 = 0, 0 − = − , 1 − 0 = 0, 0−0=0 , 9−6=3
Then, we get the new following table:
𝑐 2 5 0 0 0
𝑏
𝜃
𝑥
F.R. 𝒄 Basis 𝑥 𝑥 𝑠 𝑠 𝑠
1
5 𝑥 ¼ 1 ¼ 0 0 6 24
3
11 11 1 60
0 𝑠 0 − 1 0 15
3 4 4 11
3 1 4
0 𝑠 ( ) 0 − 0 1 3
4 4
5 5
𝑍 = ∑𝑐 𝑎 5 0 0 30
4 4
−5 0
𝑐 −𝑍 0 0
4 Second FS
The improved basic feasible solution can be read from above table as:
𝑥 = 6, 𝑠 = 15, 𝑠 = 3 and 𝑥 = 𝑠 = 0.
Once again, calculate values of 𝑐 – 𝑧 to check whether the solution shown in above
table is optimal or not. Since 𝑐 – 𝑍 > 0, the current solution is not optimal.
Step 5: Iterate towards an optimal solution: (repeat stem 4 again)
Now, mark the column 𝑥 as a key column and 𝑥 is the incoming variable. Then
the replacement ratios 𝜃 are 24, 60/11 and 4. Since 4 is the minimum non-negative
ratio, 𝑠 row marked as a key row, 𝑠 is outgoing variable. Hence, 𝑠 is replace by 𝑥 in
the next table.
Here (3/4) is the key-element. Now we have to make the key element 1 and
remaining element in 𝑥 column zero. For this perform the following operations.
4
𝑅 (𝑛𝑒𝑤) → 𝑅 (𝑜𝑙𝑑) ÷ (𝑘𝑒𝑦 𝑒𝑙𝑒𝑚𝑒𝑛𝑡)
3
1 4
→ (1, 0 , − , 0 , , 4 )
3 3

𝟒
− ( ) = 0, 0 − (0) = 0, − − (− ) = , 1 − (0) = 1, 0 − (1) = − ,

15 − (3) = 4.

𝟒
− ( ) = 0, 1 − (0) = 1, − (− ) = , 0 − (0) = 0, 0 − (1) = − 6 − (3) = 5
Then, we get the new table as follows:
𝑐 2 5 0 0 0 𝑏
𝑐 Basis 𝑥 𝑥 𝑠 𝑠 𝑠
𝑥
5 𝑥 0 1 1/3 0 -1/3 5
0 𝑠 0 0 2/3 1 -11/3 4
2 𝑥 1 0 -1/3 0 4/3 4
𝑍 = ∑𝑐 𝑎 2 5 1 0 1 33
𝑐 −𝑍 0 0 -1 0 -1 Third FS
𝑍 −row and 𝑐 − 𝑍 − row values calculated. Since all 𝑐 − 𝑍 values are negative or
zero, third feasible solution is optimal solution. The optimal solution is given by
𝑥 = 4, 𝑥 = 5, 𝑠 = 4 (basis), and 𝑠 = 0, 𝑠 = 0 (non-basic) and 𝑍 = 33.
Problem 01-03-04: Solve by simplex method the following L.P. problem. 𝑀𝑖𝑛𝑖𝑚𝑖𝑧𝑒 𝑍 =
𝑥 − 3𝑥 + 3𝑥 subject to 3𝑥 − 𝑥 + 2𝑥 ≤ 7, 2𝑥 + 4𝑥 ≥ −12, −4𝑥 + 3𝑥 + 8𝑥 ≤ 10,
where 𝑥 , 𝑥 , 𝑥 ≥ 0.
Solution: Step 1: Set-up the problem in standard form. If we observe the given
problem, we see that the second constrain R.H.S. value is negative, we have to make it
positive by multiplying second constraint by −1. Then the inequality is also reverse. We
get −2𝑥 − 4𝑥 ≤ 12.
Now, introduce the slack variables 𝑠 , 𝑠 𝑎𝑛𝑑 𝑠 , the problem can be expressed
in the standard form as follows.

𝑀𝑖𝑛𝑖𝑚𝑖𝑧𝑒 𝑍 = 𝑥 − 3𝑥 + 3𝑥 + 0𝑠 + 0𝑠 + 0𝑠
subject to 3𝑥 − 𝑥 + 2𝑥 + 𝑠 = 7,
−2𝑥 − 4𝑥 + 𝑠 = 12 ,
−4𝑥 + 3𝑥 + 8𝑥 + 𝑠 = 10,
where 𝑥 , 𝑥 , 𝑥 , 𝑠 , 𝑠 , 𝑠 ≥ 0.
Step 2: Find initial basic feasible solution (I.B.F.S.).
Putting 𝑥 = 𝑥 = 𝑥 = 0 in the constraint
we get, 𝑠 = 7, 𝑠 = 12 , 𝑠 = 10: 𝑍 = 0
𝑐 1 -3 3 0 0 0
𝑏 𝜃=
F.R. 𝑐 Basis 𝑥 𝑥 𝑥 𝑠 𝑠 𝑠
𝑥
1/3 0 𝑠 3 -1 2 1 0 0 7 -7
4/3 0 𝑠 -2 -4 0 0 1 0 12 -3
0 𝑠 -4 (3) 8 0 0 1 10 10/3
𝑍 = ∑𝑐 𝑎 0 0 0 0 0 0 0
𝑐 −𝑍 1 -3 3 0 0 0 I.B.F.S.
Step 3: Perform the optimality test:
𝑍 = ∑ 𝑐 𝑎 and 𝑐 − 𝑍 row added in the above table.
Note that the objective is to minimize Z. Therefore, if any 𝑐 − 𝑍 coefficient is

negative, the solution is not optimal. The column having most negative 𝑐 − 𝑍 value
will be the key column. Accordingly, 𝑥 − 𝑐𝑜𝑙𝑢𝑚𝑛 is marked the key column.
Replacement ratios are -7, -3 and 10/3. As 10/3 is the only non-negative ratio, 𝑠 − 𝑟𝑜𝑤
is marked the key row and (3) is the key element.
Step 4: Iterate towards an optimal solution.
Variable 𝑠 is replaced by 𝑥 . By performing the row operation in the usual

ways. Solution will become optimal when all 𝑐 − 𝑍 coefficients become zero or
positive. Then, we get following table:
𝑐 1 -3 3 0 0 0 𝑏

F.R. Basis 𝑥 𝜃=
𝑐 𝑥 𝑥 𝑥 𝑠 𝑠 𝑠
0 𝑠 (5/3) 0 14/3 1 0 1/3 31/3 31/5
22/5 0 𝑠 -22/3 0 32/3 0 1 4/3 76/3 -38/11
4/5 -3 𝑥 -4/3 1 8/3 0 0 1/3 10/3 -5/2
𝑍 = ∑𝑐 𝑎 4 -3 -8 0 0 -1 -10
𝑐 −𝑍 -3 0 11 0 0 1 Second FS
If we see the above table again we can observe that not all
𝑐 − 𝑍 are positive or zero. So, second feasible solution is not an optimal solution.
Step 5: Iterate towards an optimal solution.
Variable 𝑠 is replaced by 𝑥 . By performing the row operation in the usual

ways. We get new table as follow:
𝑐 1 -3 3 0 0 0 𝑏
𝑐 Basis 𝑥 𝑥 𝑥 𝑠 𝑠 𝑠 𝑥
1 𝑥 1 0 14/5 3/5 0 1/5 31/5
0 𝑠 0 0 156/5 22/5 1 14/5 354/5

-3 𝑥 0 1 32/5 4/5 0 3/5 58/5
𝑍 = ∑𝑐 𝑎 1 -3 -82/5 -9/5 0 -8/5 -143/5
𝑐 −𝑍 0 0 97/5 9/5 0 8/5

Third F.S.
Since all 𝑐 − 𝑍 are non-negative. The third feasible solution is optimal solution. The
optimal solution to the problem is
𝑥 = ,𝑥 = ,𝑠 = (basis), and 𝑠 = 0, 𝑠 = 0 (non-basic) and 𝑍 =− .

SELF-TEST 02:
MCQ 03-03: the simplex method is useful for the finding solution of L.P. problem
involving
a) Only two variables.
b) Only Three variables.
c) Almost three variables.
d) Any finite number of variables.
MCQ 03-04: Solving minimization L.P. problem using simplex method, the solution is
optimal if
a) All 𝑐 − 𝑍 are positive.
b) All 𝑐 − 𝑍 are negative.
c) All 𝑐 − 𝑍 are non-positive.
d) All 𝑐 − 𝑍 are non-negative.

SAQ 01-03-03: Explain the simplex method of solving linear programming problem.
Solution: The following steps are to be performing for simplex method:
Step 1: Formulation of the mathematical model : Express the L.P. problem in

standard form by introducing slack and surplus and/ or artificial variable. Assign
coefficient 𝑐 to the in the objective function.
Step 2: Set-up the initial solution: Write the initial simplex table and obtain the initial
basic feasible solution. Compute 𝑐 and 𝑐 − 𝑍 values.
a) If problem is maximization select the key column with largest positive 𝑐 − 𝑍 .
b) If problem is minimization select the key column with most negative 𝑐 − 𝑍 .
c) Select key raw with minimum non-negative . If all ratios are negative or
infinity, the current solution is unbounded solution and stop the computation.
d) Identify the key element at the intersection of key row and key column.
e) Divide all element of the key row by key element to get modified key row for
the next simplex table.

f) Write the new simplex table by performing elementary row operation for the
remaining rows.
g) Compute 𝑐 and 𝑐 − 𝑍 values.
h) If he problem is maximization, check are all 𝑐 − 𝑍 ≥ 0, if yes the solution

optimal solution if not then repeat from step 2.
i) If he problem is minimization, check are all 𝑐 − 𝑍 ≤ 0, if yes, the solution

optimal solution and if not repeat step 2 .
SAQ 01-03-04: Solve the following by using simplex method

𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍 = 3𝑥 + 5𝑥 + 4𝑥 subject to 2𝑥 + 3𝑥 ≤ 8 , 2𝑥 + 5𝑥 ≤ 10, 3𝑥 +
2𝑥 + 4𝑥 ≤ 15, where 𝑥 , 𝑥 , 𝑥 ≥ 0.
Solution: Step 1: Introducing non-negative slack variables 𝑠 , 𝑠 and 𝑠 to convert the

given L. P. problem into its standard form:
Maximize 𝑍 = 3𝑥 + 5𝑥 + 4𝑥 + 0𝑠 + 0𝑠 + 0𝑠
subject to the constraints 2𝑥 + 3𝑥 + 𝑠 = 8, 2𝑥 + 5𝑥 + 𝑠 = 10,
3𝑥 + 2𝑥 + 4𝑥 + 𝑠 = 15 𝑎𝑛𝑑 𝑥 , 𝑥 , 𝑥 , 𝑠 , 𝑠 , 𝑠 ≥ 0
Step 2: Find Initial Basic Feasible Solution (I.B.F.S.).
Putting 𝑥 = 𝑥 = 𝑥 = 0 in the constraint, we get 𝑠 = 8, 𝑠 = 10, 𝑠 = 15: 𝑍 = 0
𝑐 3 5 4 0 0 0
𝑏 𝜃=
𝑥
- 0 𝑠 2 (3) 0 1 0 0 8 8/3
- 0 𝑠 2 0 5 0 1 0 10 0
2 0 𝑠 3 2 4 0 0 1 15 5
𝑍 = ∑𝑐 𝑎 0 0 0 0 0 0 0
𝑐 −𝑍 3 5 4 0 0 0 I.B.F.S.
Step 3: Perform the optimality test:
𝑍 = ∑ 𝑐 𝑎 and 𝑐 − 𝑍 row added in the above table.
Note that, the objective is to maximization Z. Therefore, if any 𝑐 − 𝑍

coefficient is positive, the solution is not optimal. The column having most positive

𝑐 − 𝑍 value will be the key column. Accordingly, 𝑥 − 𝑐𝑜𝑙𝑢𝑚𝑛 is marked the key
column. Replacement ratios are 8/3, 0 and 5. As 8/3 is the smallest positive ratio,
𝑠 − 𝑟𝑜𝑤 is marked the key row and (3) is the key element . Now make key element 1 by
dividing this row by 3 and make 0 in this key column by usin g fix ratio.
𝑐 3 5 4 0 0 0 Min
𝑏 𝜃=
𝑥
- 5 𝑠 2/3 1 0 -1/3 0 0 8/3
- 0 𝑠 -4/3 0 (5) -2/3 1 0 14/3 -14/15
2 0 𝑠 5/3 0 4 -2/3 0 1 29/3 29/12
𝑍 = ∑𝑐 𝑎 10/3 5 0 5/3 0 0 0
𝑐 −𝑍 -1/3 0 4 -5/3 0 0 I.B.F.S.
The improved basic feasible solution can be read from table as:
𝑥_2 = 8/3, 𝑠 = 14/3, 𝑠 = 29/3 𝑎𝑛𝑑 𝑥 = 𝑥 = 𝑠 = 0. The improved value of the

objective function becomes: 𝑍 = 5 (8/3) + 0 (14/3) + 0 (29/3) = 40/3
Once again, calculate values of 𝑐 – 𝑧 to check whether the solution shown in
table is optimal or not. Since 𝑐 – 𝑧 > 0, the current solution is not optimal.
Repeat the step 3 up to all 𝑐 – 𝑧 < 0 for non-basic variables. Therefore, the
optimal solution is reached with 𝑥 = ,𝑥 = ,𝑥 = 𝑎𝑛𝑑 𝑀𝑖𝑛 𝑍 = .
03-03: THE BIG M-METHOD

In the previous section, we discuss the problem, the constraints were of ≤ type
with nonnegative R.H.S. and solved these L.P.P. using simplex method. But if the
problem with constraint with ≤ type with negative R.H.S. or with ≥ 𝑜𝑟 = type with
non-negative R.H.S. we are not able to find the I.B.F.S. as because it does not satisfy
non-negativity conditions of slack variables or it does not satisfy non-negativity
conditions of surplus variables. In such a case, artificial variables are added to get an
initial basic feasible solution. They are fictitious and have no physical meaning. They
assume the role of slack variables in the first iteration, only to be replaced at a later
iteration. Thus, they are just a device to get the I.B.F.S. so that the simplex algorithm
can be applied as usual to get optimal solution. There are two techniques are available

to solve such problem Big M-Method or Method of Penalties and the Two-Phase
Method.
In this section, we discuss The Big M-Method. The method consists of the
following steps.
(i) Express the L.P. problem in the standard form by adding slack variables, surplus
variables and/or artificial variables. Assign a zero coefficient to both slack and
surplus variables. Then, assign a very large coefficient + M (minimization case) and
– M (maximization case) to artificial variable in the objective function.
(ii) The initial basic feasible solution is obtained by assigning zero value to decision
variables, 𝑥 𝑥 , … , etc.
(iii) Solve the modified L.P.P. by simplex method.
Remarks:
At any iteration of the simplex algorithm one of the following cases may arise:
i) If at least one artificial variable is a basic variable (i.e. variable that is present in
the basis) with zero value and the coefficient M in each 𝑐 – 𝑍 ( 𝑗 = 1, 2, . . . ) values
is non-negative, then the given L.P. problem has no solution. That is, the current
basic feasible solution is degenerate.
ii) If at least one artificial variable is present in the basis with a positive value and the
coefficients M in each 𝑐 – 𝑍 ( 𝑗 = 1, 2, . . . ) values is non-negative, then the given
L.P. problem has no optimum basic feasible solution. In this case, the given L.P.
problem has a pseudo-optimum basic feasible solution.
SOLVED PROBLEMS 03
Problem 01-03-05: Use penalty (Big-M) method to solve the following L.P. problem
𝑀𝑖𝑛𝑖𝑚𝑖𝑧𝑒 𝑍 = 5𝑥 + 3𝑥 subject to 2𝑥 + 4𝑥 ≤ 12, 2𝑥 + 2𝑥 = 10, 5𝑥 + 2𝑥 ≥ 10,
where 𝑥 , 𝑥 ≥ 0.
Solution: Step 1: Set-up the problem in standard form.
Adding slack variable 𝑠 , surplus variable 𝑠 , and artificial variables 𝐴 , 𝐴 in the

constraints of the given L.P. problem, the standard form of the L.P. problem becomes.
𝑀𝑖𝑛𝑖𝑚𝑖𝑧𝑒 𝑍 = 5𝑥 + 3𝑥 + 0𝑠 + 0𝑠 + 𝑀𝐴 + 𝑀𝐴
subject to the constraints 2𝑥 + 4𝑥 + 𝑠 = 12,
2𝑥 + 2𝑥 + 𝐴 = 10,

5𝑥 + 2𝑥 – 𝑠 + 𝐴 = 10,
and 𝑥 , 𝑥 , 𝑠 , 𝑠 , 𝐴 , 𝐴 ≥ 0,
Step 2: Find initial basic feasible solution:
Put 𝑥 = 𝑥 = 𝑠 = 0 , we get the I.B.F.S. 𝑠 = 12, 𝐴 = 10, 𝐴 = 10 ;
𝑍 = 10𝑀 + 10𝑀 = 20𝑀.
Step-3: Perform optimality test:
𝑐 5 3 0 0 M M 𝑏
𝜃=
𝑐 Basis 𝑥 𝑥 𝑠 𝑠 𝐴 𝐴 𝑥
6
0 𝑠 2 4 1 0 0 0 12
M 𝐴 2 2 0 0 1 0 10 5
2
M 𝐴 (5) 2 0 -1 0 1 10
𝑍 = ∑𝑐 𝑎 7M 4M 0 -M M M Z=20M
5-7M
𝑐 −𝑍 3-4M 0 M 0 0 I.B.F.S.
Since the value 𝑐 – 𝑍 = 5 – 7𝑀 is the smallest value, therefore variable 𝑥 is

chosen to enter into the basis (solution mix). To decide a current basic variable to leave
the basis, calculated minimum ratio as shown in above table.
Step 3: Iteration towards an optimal solution.
Introduce variable 𝑥 into the basis and remove 𝐴 from the basis by applying the
following row operations
𝑅 (𝑛𝑒𝑤) → 𝑅 (𝑜𝑙𝑑) ÷ 5 (𝑘𝑒𝑦 𝑒𝑙𝑒𝑚𝑒𝑛𝑡); 𝑅 (𝑛𝑒𝑤) → 𝑅 (𝑜𝑙𝑑)– 2𝑅 (𝑛𝑒𝑤);
5
𝑅 (𝑛𝑒𝑤) → 𝑅 (𝑜𝑙𝑑)– 𝑅 (𝑛𝑒𝑤).
16
The improved basic feasible solution is shown in below table.
𝜃
𝑐 5 3 0 0 M 𝑏
=
𝑐 Basis 𝑥 𝑥 𝑠 𝑠 𝐴 𝑥
5/2
0 𝑠 0 (16/5) 1 2/5 0 8

5
M 𝐴 0 6/5 0 2/5 1 6
5
5 𝑥 1 2/5 0 -1/5 0 2
𝑍 = ∑𝑐 𝑎 5 6M/5+2 0 2M/5-1 M Z=10+6M
-6M/5+1
𝑐 −𝑍 0 0 -2M/5+1 0
Second F.S.
Since the value 𝑐 – 𝑍 = − + 1 is the smallest value, therefore variable 𝑥 is
chosen to enter into the basis (solution mix). To decide a current basic variable to leave
the basis, calculate minimum ratio as shown in above table.
Introduce variable 𝑥 into the basis and replace it by 𝑠 from the basis by
applying the following row operations. Thus, to get an improved basic feasible solution,
apply, the following row operations:
5 6
𝑅 (𝑛𝑒𝑤) → 𝑅 (𝑜𝑙𝑑) ÷ (𝑘𝑒𝑦 𝑒𝑙𝑒𝑚𝑒𝑛𝑡); 𝑅 (𝑛𝑒𝑤) → 𝑅 (𝑜𝑙𝑑)– 𝑅 (𝑛𝑒𝑤);
16 5
2
𝑅 (𝑛𝑒𝑤) → 𝑅 (𝑜𝑙𝑑)– 𝑅 (𝑛𝑒𝑤).
5
The improved basic feasible solution is shown in below table.
𝜃=
𝑐 5 3 0 0 M 𝑏
𝑐 Basis 𝑥 𝑥 𝑠 𝑠 𝐴 𝑥
40
3 𝑥 0 1 5/16 1/8 0 5/2
M 𝐴 0 0 -3/8 1/4 1 3 12
-4
5 𝑥 1 0 -1/8 -1/4 0 1
- M/4-
𝑍 = ∑𝑐 𝑎 5 3 M Z=10+6M
3M/8+5/16 7/8
-M/4+7/8
𝑐 −𝑍 0 0 3M/8-5/16 0
Third FS
Since 𝑐 – 𝑧 < 0 (negative) in 𝑠 -column, the current solution is not optimal. Thus,
non-basic variable s2 is chosen to replace artificial variable 𝐴 in the basis. To get an
improved basic feasible solution, apply the following row operations:

1 1
𝑅 (𝑛𝑒𝑤) → 𝑅 (𝑜𝑙𝑑) ÷ (𝑘𝑒𝑦 𝑒𝑙𝑒𝑚𝑒𝑛𝑡); 𝑅 (𝑛𝑒𝑤) → 𝑅 (𝑜𝑙𝑑)– ( ) 𝑅 (𝑛𝑒𝑤);
4 8
1
𝑅 (𝑛𝑒𝑤) → 𝑅 (𝑜𝑙𝑑) + ( ) 𝑅 (𝑛𝑒𝑤).
4
The improved basic feasible solution is shown in table.
𝑐 5 3 0 0
𝑏(𝑥 )
𝑐 Basis 𝑥 𝑥 𝑠 𝑠
1
3 𝑥 0 1 1/2 0
0 𝑠 0 0 -3/2 1 12
4
5 𝑥 1 0 -1/2 0
23
𝑍 = ∑𝑐 𝑎 5 3 -1 0
Optimal
𝑐 −𝑍 0 0 1 0 Solution
In table, all 𝑐 – 𝑍 ≥ 0 . Thus, an optimal solution is arrived at with the value of

variables as: 𝑥 = 4, 𝑥 = 1, 𝑠 = 0, 𝑠 = 12 𝑎𝑛𝑑 𝑀𝑖𝑛 𝑍 = 23.
Problem 01-03-06: 𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍 = 3𝑥 − 𝑥 subject to 2𝑥 + 𝑥 ≤ 2 , 𝑥 + 3𝑥 ≥ 3 ,

𝑥 ≤ 4, where 𝑥 , 𝑥 ≥ 0.
Solution: Step1: First set up the problem in slandered form:
We have to write the given L.P.P. in standard form, for this introduce the slack, surplus
and artificial variable.
𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍 = 3𝑥 − 𝑥 + 0𝑠 + 0𝑠 + 0𝑠 − 𝑀𝐴
subject to 2𝑥 + 𝑥 + 𝑠 = 2,
𝑥 + 3𝑥 − 𝑠 + 𝐴 = 3
𝑥 +𝑠 = 4
where 𝑥 , 𝑥 , 𝑠 , 𝑠 , 𝑠 , 𝐴 ≥ 0.
Step-2: Find initial basic feasible solution:
Put 𝑥 = 0, 𝑥 = 0, 𝑠 = 0 we get 𝑠 = 2, 𝐴 = 3, 𝑠 = 4 and 𝑍 = 3𝑀.
Step-3: Perform optimal test:

-
𝑐 3 -1 0 0 0 𝑏 𝜃=
M
Basi 𝑥
𝑐 𝑥 𝑥 𝑠 𝑠 𝑠 𝐴
s
0 𝑠 2 1 1 0 0 0 2 2
- 1
𝐴 1 (3) 0 -1 0 1 3
M
0 𝑠 0 1 0 0 1 0 4 4
𝑍 = ∑𝑐 𝑎 -M -3M 0 M 0 -M -3M
-1+3M
𝑐 −𝑍 3+M 0 -M 0 0 I.B.F.S.
Since 𝑐 − 𝑍 is positive under some variable, above table is not optimal and can
be improved.
Step 4: Iterate towards an optimal solution:
Successive iteration gives the following tables :
𝑐 3 -1 0 0 0 𝑏 𝜃
𝑐 Basis 𝑥 𝑥 𝑠 𝑠 𝑠 𝑥 =
3/5
0 𝑠 (5/3) 0 1 1/3 0 1
-1 𝑥 1/3 1 0 -1/3 0 1 3
0 𝑠 -1/3 0 0 1/3 1 3 -9
𝑍 = ∑𝑐 𝑎 -1/3 -1 0 1/3 0 -1
𝑐 −𝑍 10/3 0 0 -1/3 0
Second F.S.
Since 𝑐 − 𝑍 is positive under some variable, above table is not optimal and can be
improved.
𝑐 3 -1 0 0 0
𝑏
𝑥
𝑐 Basis 𝑥 𝑥 𝑠 𝑠 𝑠
3 𝑥 1 0 3/5 1/5 0 3/5
-1 𝑥 0 1 -1/5 -2/5 0 4/5

0 𝑠 0 0 1/5 2/5 1 16/5
𝑍 = ∑𝑐 𝑎 3 -1 2 1 0 1
0
𝑐 −𝑍 0 0 -2 -1
𝟑𝒓𝒅 F.S.
The optimal solution is 𝑥 = ,𝑥 = ;𝑍 = 1.
SELF-TEST 03
MCQ 03-05: For the maximization problem, the objective function coefficient for an
artificial variable is
a) M.
b) –M.
c) Zero.
d) None of them.
MCQ 03-06: The role of artificial variable in the simplex method is
a) To aid in finding initial basic feasible solution.
b) To start phase of simplex method.
c) To find optimal solution.

SAQ 01-03-05: Explain the Big-M method and discuss the cases arises in the solution
of L.P.P. by simplex method.
Solution: The Big M-Method consists of the following steps
(i) Express the L.P. problem in the standard form by adding slack variables, surplus
variables and/or artificial variables. Assign a zero coefficient to both slack and
surplus variables. Then assign a very large coefficient + M (minimization case) and
– M (maximization case) to artificial variable in the objective function.
(ii) The initial basic feasible solution is obtained by assigning zero value to decision
variables, 𝑥 𝑥 , … , etc.
(iii) Solve the modified L.P.P. by simplex method.
At any iteration of the simplex algorithm one of the following cases may arise:

i) If at least one artificial variable is a basic var iable (i.e., variable that is present in
the basis) with zero value and the coefficient M in each 𝑐 – 𝑍 ( 𝑗 = 1, 2, . . . )
values is non-negative, then the given LP problem has no solution. That is, the
current basic feasible solution is degenerate.
ii) If at least one artificial variable is present in the basis with a positive value and
the coefficients M in each 𝑐 – 𝑍 ( 𝑗 = 1, 2, . . . ) value is non-negative, then the
given L.P. problem has no optimum basic feasible solution. In this case, the given
L.P. problem has a pseudo-optimum basic feasible solution.
SAQ 01-03-06: Use Big-M method to solve the following L.P. problem.
𝑀𝑖𝑛𝑖𝑚𝑖𝑧𝑒 𝑍 = 600𝑥 + 500𝑥 subject to 2𝑥 + 𝑥 ≥ 80, 𝑥 + 2𝑥 ≥ 60, where

𝑥 , 𝑥 ≥ 0.
Solution: Adding surplus variables 𝑠 , 𝑠 and artificial variables 𝐴 , 𝐴 in the

constraints, the standard form of the LP problem becomes
Minimize 𝑍 = 600𝑥 + 500𝑥 + 0𝑠 + 0𝑠 + 𝑀𝐴 + 𝑀𝐴 subject to the constraints
2𝑥 + 𝑥 – 𝑠 + 𝐴 = 80, 𝑥 + 2𝑥 – 𝑠 + 𝐴 = 60 and 𝑥 , 𝑥 , 𝑠 , 𝑠 , 𝐴 , 𝐴 ≥ 0.
Now, we have to solve the problem using simplex method.
The initial basic feasible solution: 𝐴 = 80, 𝐴 = 60 and Min 𝑍 = 80𝑀 + 60𝑀 =
140𝑀 obtained by putting 𝑥 = 𝑥 = 𝑠 = 𝑠 = 0 is shown in table.
𝑐 600 500 0 0 M M 𝑏
𝑥 𝜃=
𝑐 Basis 𝑥 𝑥 𝑠 𝑠 𝐴 𝐴
M 𝐴 2 1 -1 0 1 0 80 80/1
M 𝐴 1 (2) 0 -1 0 1 60 60/2
𝑍 = ∑𝑐 𝑎 3M 3M -M -M M M 140M
𝑐 −𝑍 600- 500- M M 0 0
3M 3M I.B.F.S.
Since 𝑐 – 𝑧 value in 𝑥 -column of Table is the largest negative, therefore non-

basic variable 𝑥 is chosen to replace basic variable 𝐴 in the basis. For this, apply
following row operations
𝑅 (𝑛𝑒𝑤) = 𝑅 (𝑜𝑙𝑑) ÷ 2 (𝑘𝑒𝑦 𝑒𝑙𝑒𝑚𝑒𝑛𝑡); 𝑅 (𝑛𝑒𝑤) → 𝑅 (𝑜𝑙𝑑) – 𝑅 (𝑛𝑒𝑤)

to get an improved basic feasible solution as shown in table.

𝑐 600 500 0 0 M 𝑏
𝑥 𝜃=
𝑐 Basis 𝑥 𝑥 𝑠 𝑠 𝐴
M 𝐴 (3/2) 0 -1 1/2 1 50 100/3
500 𝑥 1/2 1 0 -1/2 0 30 60
𝑍 = ∑𝑐 𝑎 3M/2+250 500 -M M/2-250 M 15000+50M
𝑐 −𝑍 350-3M/2 0 M 250-M/2 0 S.B.F.S.
Again, since value 𝑐 – 𝑧 of in 𝑥 -column of above table is the largest negative,

non-basic variable 𝑥 is chosen to replace basic variable 𝐴 in the basis. For this, apply
following row operations:
1
𝑅 (𝑛𝑒𝑤) → 𝑅 (𝑜𝑙𝑑) × (2/3) ; 𝑅 (𝑛𝑒𝑤) → 𝑅 (𝑜𝑙𝑑) – ( ) 𝑅 (𝑛𝑒𝑤)
2
to get an improved basic feasible solution as shown in below table
𝑐 600 500 0 0 𝑏(𝑥 )

600 𝑥 1 0 -2/3 1/3 100/3
500 𝑥 0 1 1/3 -2/3 40/3

𝑍 = ∑𝑐 𝑎 600 500 -700/3 -400/3 80000/3
𝑐 −𝑍 0 0 700/3 400/3 Optimal B.F.S.
In above Table, all 𝑐 – 𝑧 ≥ 0 and no artificial variable is present in the basis

(solution mix). Hence, an optimum solution is arrived at with 𝑥 = 100/3 , 𝑥 =
40/3, 𝑚𝑖𝑛𝑖𝑚𝑢𝑚 𝑍 = 𝑅𝑠. 80,000/3.
03-04: TWO- PHASE METHOD

In the previous section, we discuss the Big M-method which is used if the L.P.P.
standard form contains artificial variables. There is another method to handle this
problem called Two-Phase method. In the first phase of this method, the sum of the
artificial variables is minimized subject to the given constraints in order to get a basic
feasible solution of the L.P. problem. The second phase minimizes, the original
objective function starting with the basic feasible solution obtained at the end of the
first phase. Since the solution of the L.P. problem is completed in two phases, this
method is called the Two-Phase method.

Steps of Algorithm:
Phase I:
a) If all the constraints in the given L.P. problem are ‘less than or equal to’ (≤) type,
then Phase II can be directly used to solve the problem. Otherwise, the necessary
number of surplus and artificial variables are added to convert constraints into
equality constraints.
b) Now, we have to find an I.B.F.S. to the original problem. For this we have to put
all artificial variable zero. To do this we have to form an artificial objective
function ( 𝑤) which is sum of all artificial variables. Now, we have to minimize
artificial objective function with original constraint using simplex method. Note
that, that the new objective function is always minimization type regardless the
original function maximization or minimization. There may be arising following
cases
i. If 𝑀𝑖𝑛 𝑤 > 0 , and at least one of the artificial variables appear in the basis at a
positive level, then given problem has no feasible solution and the procedure
terminate.
ii. If 𝑀𝑖𝑛 𝑤 = 0 and no artificial variable appear in the basis, then the basic
feasible solution is obtained to the given problem.
iii. If 𝑀𝑖𝑛 𝑤 = 0 and one or more artificial variable appear in the basis at the zero,
then the basic feasible solution is obtained to the given problem.
Phase II: Now use the basic feasible solution which is obtained in the Phase I as a
starting basic feasible solution for the original problem. Then , use simplex method to
obtain the optimal solution to the problem.
SOLVED PROBLEMS 04
Problem 01-03-07: Use the two-phase simplex method to 𝑚𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍 = 5𝑥 − 4𝑥 +
3𝑥 subject to 2𝑥 + 𝑥 − 6𝑥 = 20, 6𝑥 + 5𝑥 + 10𝑥 ≤ 76, 8𝑥 − 3𝑥 + 6 𝑥 ≤
50, where 𝑥 , 𝑥 , 𝑥 ≥ 0.
Solution:
Phase I: It consist of the following steps.
Introduce slack and artificial for standard L.P.P. , we get
𝑚𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍 = 5𝑥 − 4𝑥 + 3𝑥 + 0𝑠 + 0𝑠 + 𝐴
subject to 2𝑥 + 𝑥 − 6𝑥 + 𝐴 = 20,

6𝑥 + 5𝑥 + 10𝑥 + 𝑠 = 76,
8𝑥 − 3𝑥 + 6 𝑥 + 𝑠 = 50,
where 𝑥 , 𝑥 , 𝑥 , 𝑠 , 𝑠 , 𝐴 ≥ 0.
The new objective function is
𝑀𝑖𝑛𝑖𝑚𝑖𝑧𝑒 𝑤 = 𝐴
subject to 2𝑥 + 𝑥 − 6𝑥 + 𝐴 = 20,
6𝑥 + 5𝑥 + 10𝑥 + 𝑠 = 76
8𝑥 − 3𝑥 + 6 𝑥 + 𝑠 = 50
where 𝑥 , 𝑥 , 𝑥 , 𝑠 , 𝑠 , 𝐴 ≥ 0.
Now we have to use the Simplex method to minimize the new objective function.
Step I: Find an I.B.F.S..
Put 𝑥 = 0, 𝑥 = 0, 𝑥 = 0 in new L.P.P., we get 𝐴 = 20, 𝑠 = 76, 𝑠 = 50.
Now, format the simplex table,
𝑐 0 0 0 0 0 1 𝑏 𝜃
𝑐 Basis 𝑥 𝑥 𝑥 𝑠 𝑠 𝐴 𝑥 =
1 𝐴 2 1 -6 0 0 1 20 10
0 𝑠 6 5 10 1 0 0 76 38/3
0 𝑠 (8) -3 6 0 1 0 50 25/4
𝑍 = ∑𝑐 𝑎 2 1 -6 0 0 1 76
-2 0
𝑐 −𝑍 -1 6 0 0 I.B.F.S. phase 1
Step II: Perform Optimality test:
Since 𝑐 − 𝑍 is negative under some variable, above table does not gives optimal
solution and can be improved.
Step III: Iterate towards an optimal solution:

From the above table, 𝑥 is the incoming variable, 𝑠 is the outgoing variable and 8 is
an key element. This key element we have to make 1 and then remaining entries in this
column is zero. For this, perform the row operations, we get new table as follows:
𝑐 0 0 0 0 0 1 𝜃
𝑏
=
𝑐 Basis 𝑥 𝑥 𝑥 𝑠 𝑠 𝐴 𝑥
1 𝐴 0 (7/4) -15/2 0 -1/4 1 15/2 30/7
0 𝑠 0 29/4 11/2 1 -3/4 0 77/2 154/29
0 𝑥 1 -3/8 3/4 0 1/8 0 25/4 -50/3
𝑍 = ∑𝑐 𝑎 0 7/4 -15/2 0 -1/4 1 15/2
𝑐 −𝑍 0 -7/4 15/2 0 1/4 0

Second F.S.
Since 𝑐 − 𝑍 is negative under some variable, above table is not optimal and can
be improved. From the above table, 𝑥 is the incoming variable, 𝐴 is the outgoing
variable and (7/4) is a key element. This key element we have to make 1 and then
remaining entries in this column is zero. For this , perform the row operations, we get
new table as follows:
𝑐 0 0 0 0 0 1 𝑏
𝑐 Basis 𝑥 𝑥 𝑥 𝑠 𝑠 𝐴 𝑥
0 𝑥 0 1 -30/7 0 -1/7 4/7 30/7
0 𝑠 0 0 256/7 1 2/7 -29/7 52/7
0 𝑥 1 0 -6/7 0 1/14 3/14 55/7
𝑍 = ∑𝑐 𝑎 0 0 0 0 0 0 0
𝑐 −𝑍 0 0 0 0 0 0 Optimal Solution
Since 𝑐 − 𝑍 is non-negative under all column, above table gives optimal. Also,
𝑤 = 0 and no artificial variable is appearing in the basis, this table gives B .F.S. to
the original problem.

Phase II: Since the above table does not contain the artificial variable, we can delete
this column and just replace the new objective function in abov e table by original
objective function. Then, do corresponding change in last two column.
𝑐 5 -4 3 0 0 𝑏
𝑐 Basis 𝑥 𝑥 𝑥 𝑠 𝑠 𝑥
0 𝑥 0 1 -30/7 0 -1/7 30/7
0 𝑠 0 0 256/7 1 2/7 52/7
0 𝑥 1 0 -6/7 0 1/14 55/7
𝑍 = ∑𝑐 𝑎 5 -4 90/7 0 13/14 155/7
𝑐 −𝑍 0 0 -69/7 0 -13/14 Optimal Solution
Since 𝑐 − 𝑍 is negative or zero under all column, above table gives optimal solution to
the original problem.
𝑥 = ,𝑥 = ,𝑥 = 0 𝑎𝑛𝑑 𝑚𝑎𝑥𝑖𝑚𝑢𝑚 𝑍 = .
Problem 01-03-08: Discuss the comparison between the Big M-method and the two-
phase method.
Solution: After observation of Big M-Method and the two-phase method we see the
following points:
i) The basic approach is same. Both add artificial variable to convert the pr oblem in
slandered form and then eliminate these from the solution.
ii) The Big M-method solves the L.P. problem in one pass, whereas two-phase
method solves it in two stages as two linear problems.
iii) The Big M-method is computationally inconvenient due to the existence of the
large number, M. The two-phase method does not involve M during computation.
iv) The Big M-method presents a difficulty when problem is solved on digital
computer because this method contains M. Therefore, for computer calculation the
two-phase method has been commonly used.

SELF-TEST 04
MCQ 03-07: If all 𝑎 values in the incoming variable column of the simplex table are
negative, then
a) Solution is unbounded.
b) There are multiple solutions.
c) There exists no solution.
d) The solution is degenerate.
MCQ 03-08: The per unit improvement in the solution of the minimization L .P.
problem is indicated by the negative values of
a) 𝑐 − 𝑍 in the decision variable column.
b) 𝑐 − 𝑍 in the slack variable column.
c) 𝑐 − 𝑍 in the surplus variable column .
d) None of them.

SAQ 01-03-07: Explain the algorithm of two-phase simplex method of solving the L.P.
problem.
Solution: Steps of Two-Phase Algorithm: Phase I:
a) If all the constraints in the given L.P. problem are ‘less than or equal to’ (≤) type,
then Phase II can be directly used to solve the problem. Otherwise, the necessary
number of surplus and artificial variables are added to convert constraints into
equality constraints.
b) Now, we have to find an I.B.F.S. to the original problem. For this, we have to put
all artificial variable zero. To do this, we have to form an artificial objective
function (𝑤), which is sum of all artificial variables. Now, we have to minimize
artificial objective function with original constraint using simplex method. Note
that, that the new objective function is always minimization type regardless the
original function maximization or minimization. There may be arising following
cases:
i. If 𝑀𝑖𝑛 𝑤 > 0 , and at least one of the artificial variables appear in the basis at a
positive level, then given problem has no feasible solution and the procedure
terminate.

ii. If 𝑀𝑖𝑛 𝑤 = 0 and no artificial variable appear in the basis, then the basic
feasible solution is obtained to the given problem.
iii. If 𝑀𝑖𝑛 𝑤 = 0 and one or more artificial variable appear in the basis at the zero,
then the basic feasible solution is obtained to the given problem.
Phase II: Now, use the basic feasible solution which is obtained in the Phase I as a
starting basic feasible solution for the original problem. Then , use simplex method to
obtain the optimal solution to the problem
SAQ 01-03-08: Use two-phase simplex method to solve the following L.P. problem:
𝑚𝑖𝑛𝑖𝑚𝑖𝑧𝑒 𝑍 = 𝑥 + 𝑥 subject to 2𝑥 + 𝑥 ≥ 4, 𝑥 + 7𝑥 ≥ 7 , where 𝑥 , 𝑥 ≥ 0.
Solution: Step-1: Convert the problem in to standard L.P. form
𝑚𝑖𝑛𝑖𝑚𝑖𝑧𝑒 𝑍 = 𝑥 + 𝑥
subject to 2𝑥 + 𝑥 − 𝑠 + 𝐴 = 4,
𝑥 + 7𝑥 − 𝑠 + 𝐴 = 7,
where 𝑥 , 𝑥 , 𝑠 , 𝑠 , 𝐴 , 𝐴 ≥ 0.
Then, the new objective function with original constraint is
𝑚𝑖𝑛𝑖𝑚𝑖𝑧𝑒 𝑤 = 𝐴 + 𝐴
subject to 2𝑥 + 𝑥 − 𝑠 + 𝐴 = 4,
𝑥 + 7𝑥 − 𝑠 + 𝐴 = 7,
where 𝑥 , 𝑥 , 𝑠 , 𝑠 , 𝐴 , 𝐴 ≥ 0.
Now we have to minimize the above objective function using simplex method.
Step II: Find an I.B.F.S.
Put 𝑥 = 0, 𝑥 = 0, 𝑠 = 0, 𝑠 = 0, we get 𝐴 = 4, 𝐴 = 7.
𝑐 0 0 0 0 1 1 𝑏 𝜃
𝑥 =
𝑐 Basis 𝑥 𝑥 𝑠 𝑠 𝐴 𝐴
1 𝐴 2 1 -1 0 1 0 4 4
1 𝐴 1 (7) 0 -1 0 1 7 1
𝑍 = ∑𝑐 𝑎 3 8 -1 1 1 1 11

𝑐 −𝑍 -3 -8 1 -1 -1 -1 I.B.F.S. phase
Since 𝑐 − 𝑍 is negative under some variable, above table does not give optimal
Step III: Perform the optimality test and iterat e towards the optimal solution.
Form the above table, we see that 𝐴 is outgoing variable and 𝑥 is the incoming
variable with 7 key elements. Now, we have to make key element 1 and remaining
entries in this column is zero. Perform corresponding operations , we get new tale as
below. Delete the column 𝐴
𝑐 0 0 0 0 1 𝜃
𝑏
𝑐 Basis 𝑥 𝑥 𝑠 𝑠 𝐴 𝑥 =
1 𝐴 (13/7) 0 -1 1/7 1 3 21/13
0 𝑥 1/7 1 0 -1/7 0 1 7
𝑍 = ∑𝑐 𝑎 13/7 0 -1 1/7 1 3
𝑐 −𝑍 -13/7 0 1 -1/7 -1 Second F.S.
Since 𝑐 − 𝑍 is negative under some variable, above table does not give optimal
Step IV: Perform the optimality test and iterate towards the optimal solution.
Form the above table we see that 𝐴 is outgoing variable and 𝑥 is the incoming
variable with 13/7 key element. Now, we have to make key element 1 and remaining
entries in this column is zero. Perform corresponding operations , we get new tale as
below.
𝑐 0 0 0 0 𝑏
𝑐 Basis 𝑥 𝑥 𝑠 𝑠 𝑥
0 𝑥 1 0 -7/13 1/13 21/13
0 𝑥 0 1 1/13 -2/13 10/13
𝑍 = ∑𝑐 𝑎 0 0 0 0 0

𝑐 −𝑍 0 0 0 0
Second F. S.
Since 𝑐 − 𝑍 is non-negative under all variables, above table give optimal solution.
Thus, the optimal solution for new objective function is 𝑥 = ,𝑥 = − , 𝑍 = 0.
Phase II: Since the above table does not contain the artificial variable and just replace
the new objective function in above table by original objective function. Then , do
corresponding change in last two columns.
𝑐 1 1 0 0 𝑏
𝑥
1 𝑥 1 0 -7/13 1/13 21/13
1 𝑥 0 1 1/13 -2/13 10/13

-1/13
𝑍 = ∑𝑐 𝑎 1 1 -6/13 31/13
𝑐 −𝑍 0 0 6/13 1/13 F.S. phase 2
Since 𝑐 − 𝑍 is non-negative under all variables, above table give optimal solution.
Thus, the optimal solution for new objective function is 𝑥 = ,𝑥 = − ,𝑍 = .
SUMMARY
The simplex method is an interactive procedure for reaching the optimal
solution to any L.P. problem. It consists of a series of rules that, in effect, algebraically
examine corner (extreme) points of the solution space in a systematic way. Each step
moves towards the optimal solution by increasing profit or decreasing cost, whil e
maintaining feasibility. The simplex method consists of five steps: (i) identifying the
key column, (ii) identifying the key row and key element, (iii) replacing the key row,
(iv) computing new values for each remaining row; and (v) computing the 𝑧 and 𝑐 – 𝑧
values and examining for optimality. Each step of this iterative procedure is displayed
and explained in a simplex table both for maximization and minimization L .P.
problems. Also, we can see that the two-phase simplex method and Big M-method are
useful to solve the L.P. model involving artificial variable and ≥ inequality in
constraint with positive R.H.S.

1. Solve the following by using simplex method 𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍 = 7𝑥 + 5𝑥
subject to 2𝑥 + 3𝑥 ≤ 20, 3𝑥 + 𝑥 ≥ 10, where 𝑥 , 𝑥 ≥ 0.

2. Using simplex method 𝑀𝑖𝑛𝑖𝑚𝑖𝑧𝑒 𝐶 = 8𝑥 + 5𝑦 subject to 20𝑥 + 12𝑦 ≥ 200,
8𝑥 ≥ 40, 6𝑦 ≥ 30, where 𝑥, 𝑦 ≥ 0.
3. A company makes two kinds of leather belts, belt A and belt B. Belt A is a high -
quality belt and belt B is of lower quality. The respective profits are Rs. 4 and
Rs. 3 per belt. The production of each of type A requires twice as much time as a
belt of type B, and if all belts were of type B, the company could make 1,000
belts per day. The supply of leather is sufficient for only 800 belts per day (both
A and B combined). Belt A requires a fancy buckle and only 400 of these are
available per day. There are only 700 buckles a day available for belt B. What
should be the daily production of each type of belt? Formulate this problem as
an L.P. model and solve it using the simplex method.
4. Use the Charnes penalty method to 𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍 = 3𝑥 − 𝑥 subject to
2𝑥 + 𝑥 ≥ 2, 𝑥 + 3𝑥 ≤ 3, 𝑥 ≤ 4 𝑎𝑛𝑑 𝑥 , 𝑥 ≥ 0 .
5. Using Big M-method 𝑀𝑖𝑛𝑖𝑚𝑖𝑧𝑒 𝑍 = 4𝑥 + 5𝑥 − 3𝑥 subject to
𝑥 − 𝑥 ≥ 1,, 𝑥 + 𝑥 + 𝑥 = 10, 2𝑥 + 3𝑥 + 𝑥 ≤ 40, where 𝑥 , 𝑥 , 𝑥 ≥ 0.
6. Using two-phase simplex method 𝑀𝑖𝑛𝑖𝑚𝑖𝑧𝑒 𝑍 = 𝑥 + 𝑥 subject to 2𝑥 + 𝑥 ≥ 4,

𝑥 + 7𝑥 ≥ 7, where 𝑥 , 𝑥 ≥ 0.
7. Using two phase simplex method 𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍 = 5𝑥 − 2𝑦 − 3𝑧 subject to 2𝑥 +

2𝑦 − 𝑧 ≥ 2, 3𝑥 − 4𝑦 ≤ 3, 3𝑥 − 4𝑦 ≤ 3, 𝑦 + 3𝑧 ≤ 5, where 𝑥, 𝑦, 𝑧 ≥ 0.
KEY WORDS
Simplex Method, optimal, two-phase method, Big M-Method, Initial Basic Feasible
Solution, Iteration.
REFERENCES
BOOKS
& COMPANY LTD

PRAGATI EDITION

MOOCS
YOUTUBE VIDEOS
1) https://youtu.be/M8POtpPtQZc
2) https://youtu.be/0PAOSZhwsnc
3) https://youtu.be/W_6ACHrjuWM
4) https://youtu.be/wBMwK1eF4M8
WIKIPEDIA
https://people.richland.edu/james/ictcm/2006/simplex.html
http://www.math.wsu.edu/students/odykhovychnyi/M201-04/Ch06_1-
2_Simplex_Method.pdf
OER
Visit here for course companion website for this course :

UNIT 01-04: DUALITY
LEARNING OBJECTIVES
 Formulate the dual L.P. problem
 Appreciate the significance of the duality concept.
 Understand the relationship between primal and dual L .P. problems.
 Solve L.P. model using dual simplex method.
INTRODUCTION
The meaning of dual in general is two or double. For every L.P. problem there is
a related unique L.P. problem involving the same data which also describe the original
problem. That is any L.P. problem (either maximization or minimization) can be stated
in another equivalent form based on the same data. The given original problem is called
primal programme (P). The new L.P. problem is called dual linear programming
problem or in short dual. In general, it is immaterial which of the two problems is
called primal or dual, since the dual of the dual is primal . The variable of the dual
problem (or programme) is called dual variable or shadow prices of the various
resources. The shadow price is also defined as the rate of change in the optimal
objective function value with respect to the unit change in the availability of a resource.
Duality is an extremely important and interesting feature of linear programming. Some

advantages of the duality are:
1. It is good way to solve the dual of a primal that has a smaller number of constraints
because the number of constraints usually equals the number of iterations required
to solve the problem.
2. It gives the additional information as to how the optimal solution changes as result
of the changes in the coefficient and the formulation of the problem. This technique
is called the sensitivity analysis.
3. Using calculation of the dual we can the accuracy of the primal solution.
4. Duality is used to solve an L.P. problem by the simplex method in which the initial
solution is infeasible.
5. Economic interpretation of the dual helps the management in making future

decision.

6. It avoids the necessity of adding surplus or artificial and solve the problem quickly.
The technique is called primal-dual method.
04-01: DUAL PROBLEM WHEN PRIMAL IS IN CANONICAL FORM

Suppose the primal L.P. problem is given in the canonical form as follow
Minimize 𝑍 = 𝑐 𝑥 + 𝑐 𝑥 + ⋯+ 𝑐 𝑥 ,
subject to 𝑎 𝑥 + 𝑎 𝑥 + ⋯+ 𝑎 𝑥 ≤ 𝑏 ,
𝑎 𝑥 + 𝑎 𝑥 + ⋯+ 𝑎 𝑥 ≤ 𝑏 ,
……………………………….. …….(1)
…………………………………
𝑎 𝑥 +𝑎 𝑥 +⋯+ 𝑎 𝑥 ≤𝑏 ,
where 𝑥 ,𝑥 ,..,𝑥 ≥ 0.
If the above problem is considered primal, then the dual is
Minimize 𝑍 = 𝑏 𝑦 + 𝑏 𝑦 + ⋯+ 𝑏 𝑦 ,
subject to 𝑎 𝑦 +𝑎 𝑦 + ⋯+ 𝑎 𝑦 ≤𝑐 ,
𝑎 𝑦 +𝑎 𝑦 + ⋯+ 𝑎 𝑦 ≤𝑐 ,
……………………………….. ……(2)
…………………………………
𝑎 𝑦 +𝑎 𝑦 + ⋯+ 𝑎 𝑦 ≤𝑐 ,
where 𝑦 ,𝑦 ,..,𝑦 ≥ 0.
The above equation (1) and (2) are called symmetric -primal dual pairs.
In general, the primal-dual relationship between a pair of L.P. problems can be

expressed as follows:
Primal Dual
Maximize 𝑍 = ∑ 𝑐𝑥 Minimize 𝑊 = ∑ 𝑏 𝑦
Subject to constraint Subject to constraint
∑ 𝑎 𝑥 ≤𝑏, 𝑖 = 1,2, … , 𝑚 , ∑ 𝑎 𝑦 ≤𝑐, 𝑖 = 1,2, … , 𝑛,
where 𝑥 ≥ 0, 𝑗 = 1,2, … , 𝑛. where 𝑦 ≥ 0, 𝑖 = 1,2, … , 𝑚.

A summary of the general relationships between primal and du al L.P. problems is given
in table
Primal Then Dual
i) Maximization problem. i) Minimization problem.
ii) n variable and m constraint. ii) m variable and n constraint.
iii) ≤ type. iii) ≥ type.
iv) Objective function coefficients. iv) R.H.S. constants of the

constraints.
v) RHS constants of the constraints.
v) Objective function coefficients.
vi) = type constraints.
vi) Unrestricted 𝑖 variable.
vii) Variable 𝑥 unrestricted.
vii) 𝑗 constraint = type.
SOLVED PROBLEMS 01
Problem 01-04-01: Write the dual of the following L.P. problem
Minimize 𝑍 = 3𝑥 – 2𝑥 + 4𝑥
Subject to constraint 3𝑥 + 5𝑥 + 4𝑥 ≥ 7,
6𝑥 + 𝑥 + 3𝑥 ≥ 4,
7𝑥 – 2𝑥 – 𝑥 ≤ 10,
𝑥 – 2𝑥 + 5𝑥 ≥ 3,
4𝑥 + 7𝑥 – 2𝑥 ≥ 2 and 𝑥 , 𝑥 , 𝑥 ≥ 0.
Solution: Since the objective function of the given L.P. problem is of minimization, the
direction of each inequality has to be changed to ≥ type by multiplying both sides by
– 1. The standard primal L.P. problem so obtained is:
Minimize 𝑍 = 3𝑥 – 2𝑥 + 4𝑥
Subject to constraint 3𝑥 + 5𝑥 + 4𝑥 ≥ 7,
6𝑥 + 𝑥 + 3𝑥 ≥ 4,
−7𝑥 + 2𝑥 + 𝑥 ≥ − 10,
𝑥 – 2𝑥 + 5𝑥 ≥ 3,

4𝑥 + 7𝑥 – 2𝑥 ≥ 2 and 𝑥 , 𝑥 , 𝑥 ≥ 0.
If 𝑦 , 𝑦 , 𝑦 , 𝑦 𝑎𝑛𝑑 𝑦 are dual variables corresponding to the five primal constraints in

the given order, the dual of this primal L.P. problem is stated as:
Maximize 𝑊 = 7𝑦 + 4𝑦 – 10𝑦 + 3𝑦 + 2𝑦
subject to constraint 3𝑦 + 6𝑦 − 7𝑦 + 𝑦 + 4𝑦 ≤ 3,
5𝑦 + 𝑦 + 2𝑦 − 2𝑦 + 7𝑦 ≤ −2,
4𝑦 + 3𝑦 + 𝑦 + 5𝑦 − 2𝑦 ≤ 4.
Problem 01-04-02: Construct the dual of the problem
Maximize 𝑍 = 3𝑥 + 17𝑥 + 9𝑥 subject to the constraint 𝑥 − 𝑥 ≥ 3 ,
−3𝑥 + 2𝑥 + 𝑥 ≤ 1 and 𝑥 , 𝑥 , 𝑥 ≥ 0.
Solution: The problems maximization L.P. problem, hence all constraints are of ≤ type.
So, multiply first constraint by −1, we get primal in form
Maximize 𝑍 = 3𝑥 + 17𝑥 + 9𝑥 subject to the constraint −𝑥 + 𝑥 ≤ −3 ,
−3𝑥 + 2𝑥 + 𝑥 ≤ 1 and 𝑥 , 𝑥 , 𝑥 ≥ 0.
Let 𝑦 , 𝑦 are the dual variable, then the dual of the above primal is
Minimize 𝑤 = −3𝑦 + 𝑦 subject to constraints – 𝑦 − 3𝑦 ≥ 3, 𝑦 + 2𝑦 ≥ 17 ,
𝑦 ≥ 9 𝑎𝑛𝑑 𝑦 , 𝑦 ≥ 0.
SELF-TEST 01
MCQ 04-01: The dual of the primal minimization L.P. problem having 𝑚 constraints
and 𝑛 non-negative variables should
a) have 𝑛 constraints and 𝑚 non-negative variables.
b) be a maximization problem.
c) Both a) and b).
d) Neither a) nor b).
MCQ 04-02: The right-hand side constant of a constraint in a primal problem appears
in the corresponding dual as
a) a right-hand side constant of a constraint.
b) an input-out coefficient.
c) a coefficient in the objective function.

d) None of the above.

SAQ 01-04-01: Construct the dual of to the primal problem: Maximize 𝑍 = 3𝑥 + 5𝑥
subject to the constraints 2𝑥 + 6𝑥 ≤ 50 , 3𝑥 + 2𝑥 ≤ 35, 5𝑥 − 3𝑥 ≤ 10, 𝑥 ≤ 20 and
𝑥 , 𝑥 ≥ 0.
Solution: The dual of the given problem is
Minimize 𝑤 = 50𝑦 + 35𝑦 + 10𝑦 + 20𝑦 subject to constraints
2𝑦 + 3𝑦 + 5𝑦 ≥ 3, 6𝑦 + 2𝑦 − 3𝑦 + 𝑦 ≥ 5 𝑎𝑛𝑑 𝑦 , 𝑦 , 𝑦 , 𝑦 ≥ 0.
SAQ 01-04-02: Construct the dual of the problem: Maximize 𝑍 = 3𝑥 + 17𝑥 + 9𝑥

subject to the constraints 𝑥 − 𝑥 + 3 ≥ 3 , −3𝑥 + 2𝑥 ≤ 1 and 𝑥 , 𝑥 , 𝑥 ≥ 0.
Solution: The dual of the given problem is
Minimize 𝑤 = −3𝑦 + 𝑦 subject to constraints
– 𝑦 − 3𝑦 ≥ 3, 𝑦 + 2𝑦 ≥ 17 , −𝑦 + 2𝑦 ≥ 9 𝑎𝑛𝑑 𝑦 , 𝑦 ≥ 0.
04-02: DUAL PROBLEM WHEN PRIMAL IS IN THE STANDARD FORM

We discuss in section 03-01 the standard form of the L.P. problem in which all
constraints are equations (=) form. In this section , we have to discuss that an equality in
primal corresponds to an unconstrained variable in the dual and vice versa. We prove by
taking one example.
Consider the primal with = sign
Maximize 𝑍 = 𝑐 𝑥 + 𝑐 𝑥 , subject to constraint
𝑎 𝑥 + 𝑎 𝑥 = 𝑏 , 𝑎 𝑥 + 𝑎 𝑥 ≤ 𝑏 𝑎𝑛𝑑 𝑥 ≥ 0, 𝑥 ≥ 0.
The first constraint is of = type, it is equivalent to the canonical form
𝑎 𝑥 + 𝑎 𝑥 ≤ 𝑏 𝑎𝑛𝑑 𝑎 𝑥 + 𝑎 𝑥 ≥ 𝑏 .
The problem is of maximization form hence all inequalities are of ≤ type. Then we can
write 𝑎 𝑥 + 𝑎 𝑥 ≤ 𝑏 𝑎𝑛𝑑 − 𝑎 𝑥 − 𝑎 𝑥 ≤ −𝑏 .
Now, we have L.P. problem with three constraint and two variables. Now , find the dual
of this, for this suppose 𝑦 , 𝑦 , 𝑦 are the dual variables. Then, we get dual
Minimize 𝑤 = 𝑏 (𝑦 − 𝑦 ) + 𝑏 𝑦 subject to constraint 𝑎 (𝑦 − 𝑦 ) + 𝑎 𝑦 ≥ 𝑐 ,

𝑎 (𝑦 − 𝑦 ) + 𝑎 𝑦 ≥ 𝑐 , 𝑦 , 𝑦 , 𝑦 ≥ 0 .

We observe that, 𝑦 − 𝑦 is in the objective function as well as in the constraint. If we
put 𝑦 − 𝑦 = 𝑦 , which is the difference of the two non-negative numbers. So, 𝑦 may
be positive, negative or zero i.e. it is unrestricted in sign. The dual problem becomes
Minimize 𝑤 = 𝑏 𝑦 + 𝑏 𝑦 subject to constraint
𝑎 𝑦 + 𝑎 𝑦 ≥ 𝑐 , 𝑎 𝑦 + 𝑎 𝑦 ≥ 𝑐 , 𝑎𝑛𝑑 𝑦 ≥ 0 , 𝑢𝑛𝑟𝑒𝑠𝑡𝑟𝑖𝑐𝑡𝑒𝑑 𝑖𝑛 𝑠𝑖𝑔𝑛 𝑦 .
Hence, the dual variable which corresponds to an equality constraint must be

unrestricted in sign. Similarly, we can prove this for 𝑛 variables and 𝑚 constraint L.P.
problem. Conversely, when a primal variable is unrestricted in sign, its dual constraint
must be in equality form.
SOLVED PROBLEMS 02
Problem 01-04-03: Obtain the dual L.P. problem of the following primal L.P. problem
𝑀𝑖𝑛𝑖𝑚𝑖𝑧𝑒 𝑍 = 𝑥 + 2𝑥 subject to constraints 2𝑥 + 4𝑥 ≤ 160, 𝑥 – 𝑥 = 30,

𝑥 ≥ 10 𝑎𝑛𝑑 𝑥 , 𝑥 ≥ 0 .
Solution: Since the objective function of the primal L.P. problem is of minimization,
change all ≤ type constraints to ≥ type constraints by multiplying the constraint on
both sides by – 1. Also, write = type constraint equivalent to two constraints of the type
≥ and ≤. i.e. 𝑥 – 𝑥 = 30 is equivalent to the pair of inequalities
𝑥 – 𝑥 ≥ 30 𝑎𝑛𝑑 𝑥 − 𝑥 ≤ 30.
Then the given L.P. problem can be written as
𝑀𝑖𝑛𝑖𝑚𝑖𝑧𝑒 𝑍 = 𝑥 + 2𝑥 subject to constraint
− 2𝑥 − 4𝑥 ≥ − 160, 𝑥 – 𝑥 ≥ 30, − 𝑥 + 𝑥 ≥ − 30 , 𝑥 ≥ 10 𝑎𝑛𝑑 𝑥 , 𝑥 ≥ 0 .
Let 𝑦 , 𝑦 , 𝑦 𝑎𝑛𝑑 𝑦 be the dual variables corresponding to the four constraints in the
given order. The dual of the given primal L.P. problem is
𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍𝑦 = – 160𝑦 + 30(𝑦 – 𝑦 + 10𝑦 subject to constraint
−2𝑦 + (𝑦 − 𝑦 ) + 𝑦 ≤ 1, − 4𝑦 − (𝑦 − 𝑦 ) ≤ 2 𝑎𝑛𝑑 𝑦 , 𝑦 , 𝑦 ≥ 0.
Put (𝑦 − 𝑦 ) = 𝑦 we get 𝑦 is unrestricted in sign. Then, we get
𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍𝑦 = – 160𝑦 + 30𝑦 + 10𝑦 subject to constraint −2𝑦 + 𝑦 + 𝑦 ≤ 1,
−4𝑦 − 𝑦 ≤ 2 𝑎𝑛𝑑 𝑦 ≥ 0 , 𝑦 𝑖𝑠 𝑢𝑛𝑟𝑒𝑠𝑡𝑟𝑖𝑐𝑡𝑒𝑑 𝑖𝑛 𝑠𝑖𝑔𝑛.
Problem 01-04-04: Construct the dual of the primal problem
Maximize 𝑍 = 2𝑥 + 𝑥 + 𝑥 , subject to constraint 𝑥 + 𝑥 + 𝑥 ≥ 6, 3𝑥

−2𝑥 + 3𝑥 = 3, −4𝑥 + 3𝑥 − 6𝑥 = 1, 𝑥 , 𝑥 , 𝑥 ≥ 0.
Solution: As the problem is of the maximization type all constraint must be of the ≤
type. So, multiply first constraint by −1, we get −𝑥 − 𝑥 − 𝑥 ≤ −6 .
Also, the constraint 3𝑥 − 2𝑥 + 3𝑥 = 3 is equivalent to the pair of inequalities

3𝑥 − 2𝑥 + 3𝑥 ≥ 3 𝑎𝑛𝑑 3𝑥 − 2𝑥 + 3𝑥 ≤ 3 . Or it is written in the from
−3𝑥 + 2𝑥 − 3𝑥 ≤ −3 𝑎𝑛𝑑 3𝑥 − 2𝑥 + 3𝑥 ≤ 3.
Also, the constraint −4𝑥 + 3𝑥 − 𝑥 = 1 is equivalent to the pair of inequality −4𝑥 +

3𝑥 − 6𝑥 ≥ 1 𝑎𝑛𝑑 − 4𝑥 + 3𝑥 − 6𝑥 ≤ 1 i.e.
4𝑥 − 3𝑥 + 6𝑥 ≤ −1 𝑎𝑛𝑑 − 4𝑥 + 3𝑥 − 6𝑥 ≤ 1.
Then the given primal problem is in the form
Maximize 𝑍 = 2𝑥 + 𝑥 + 𝑥 ,
subject to constraint −𝑥 − 𝑥 − 𝑥 ≤ −6,
3𝑥 − 2𝑥 + 3𝑥 ≤ 3,
−3𝑥 + 2𝑥 − 3𝑥 ≤ −3 ,
−4𝑥 + 3𝑥 − 6𝑥 ≤ 1,
4𝑥 − 3𝑥 + 6𝑥 ≤ −1,
𝑥 , 𝑥 , 𝑥 ≥ 0.
The dual of the primal is
Minimize 𝑤 = −6𝑦 + 3(𝑦 − 𝑦 ) + (𝑦 − 𝑦 )
Subject to the constraint– 𝑦 + 3(𝑦 − 𝑦 ) − (4𝑦 − 𝑦 ) ≥ 2,
−𝑦 − 2(𝑦 − 𝑦 ) + 3(𝑦 − 𝑦 ) ≥ 1 , −𝑦 + 3(𝑦 − 𝑦 ) − 6(𝑦 − 𝑦 ) ≥ 1
𝑎𝑛𝑑 𝑦 , 𝑦 , 𝑦 , 𝑦 , 𝑦 ≥ 0.
Put 𝑦 − 𝑦 = 𝑦 𝑎𝑛𝑑 𝑦 − 𝑦 = 𝑦 , where 𝑦 𝑎𝑛𝑑 𝑦 are unrestricted in sign.
The dual is in the form
Minimize 𝑤 = −6𝑦 + 3𝑦 + 𝑦
Subject to the constraint
– 𝑦 + 3𝑦 − 4𝑦 ≥ 2, −𝑦 − 2𝑦 + 3𝑦 ≥ 1 , − 𝑦 + 3𝑦 − 6𝑦 ≥1
𝑦 ≥ 0 and 𝑦 , 𝑦 are unrestricted in sign.

SELF-TEST 02
MCQ 04-03: An equality constraint in primal corresponds to an
a) Unrestricted variable in the dual.
b) Restricted variable in the dual.
c) Any variable in the dual.
d) Variable with zero coefficient in the dual.
MCQ 04-04: The difference of the two non-negative number is
a) Negative number.
b) Positive number.
c) 0.
d) May be negative or positive number or 0.

SAQ 01-04-03: Obtain the dual of the primal: Minimize 𝑍 = 3𝑥 − 2𝑦 + 𝑧 subject to
2𝑥 − 3𝑦 + 𝑧 ≤ 5, 4𝑥 − 2𝑦 ≥ 9, − 8𝑥 + 4𝑦 + 3𝑧 = 8 and 𝑥, 𝑦 ≥ 0, 𝑧 unrestricted.
Solution: As z is unrestricted put 𝑧 = 𝑧 − 𝑧 , 𝑤ℎ𝑒𝑟𝑒 𝑧 ≥, 𝑧 ≥ 0 in given problem

become Minimize 𝑍 = 3𝑥 − 2𝑦 + 𝑧 − 𝑧 subject to 2𝑥 − 3𝑦 + 𝑧 − 𝑧 ≤ 5,
4𝑥 − 2𝑦 ≥ 9, − 8𝑥 + 4𝑦 + 3𝑧 − 3𝑧 = 8 and 𝑥, 𝑦 , 𝑧 , 𝑧 ≥ 0.
The problem is minimization hence all constraint must be in the form ≥. Hence, do
multiply inequality 1 by -1 we get −2𝑥 + 3𝑦 − 𝑧 + 𝑧 ≥ −5,
Also, write equation −8𝑥 + 4𝑦 + 3𝑧 = 8 as −8𝑥 + 4𝑦 + 3𝑧 ≥ 8, and −8𝑥 + 4𝑦 + 3𝑧 ≤ 8,
Hence, we get the given primal as
Minimize 𝑍 = 3𝑥 − 2𝑦 + 𝑧 − 𝑧 subject to −2𝑥 + 3𝑦 − 𝑧 + 𝑧 ≥ −5, 4𝑥 − 2𝑦 ≥ 9,
−8𝑥 + 4𝑦 + 3𝑧 − 3𝑧 ≥ 8, 8𝑥 − 4𝑦 − 3𝑧 + 3𝑧 ≥ −8 and 𝑥, 𝑦 , 𝑧 , 𝑧 ≥ 0.
Then, dual in simplified form is
𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑤 = −5𝑦 + 9𝑦 + 8𝑦 Subject to
−2𝑦 + 4𝑦 − 8𝑦 ≤ 3, 3𝑦 − 2𝑦 + 4𝑦 ≤ −2,
−𝑦 + 3𝑦 = 1 𝑎𝑛𝑑 𝑦 , 𝑦 ≥ 0, 𝑦 unrestricted
SAQ 01-04-04: Show that dual variable which corresponds to an equality constraint
must be unrestricted in sign with the help of example.

Solution: Consider the primal i with equality sign constraint.
Maximize 𝑍 = 𝑐 𝑥 + 𝑐 𝑥 , subject to constraint
𝑎 𝑥 + 𝑎 𝑥 = 𝑏 , 𝑎 𝑥 + 𝑎 𝑥 ≤ 𝑏 𝑎𝑛𝑑 𝑥 ≥ 0, 𝑥 ≥ 0.
The first constraint is of = type, it is equivalent to the canonical form
𝑎 𝑥 + 𝑎 𝑥 ≤ 𝑏 𝑎𝑛𝑑 𝑎 𝑥 + 𝑎 𝑥 ≥ 𝑏 .
The problem is of maximization form hence all inequalities are of ≤ type. Then, we
can write 𝑎 𝑥 + 𝑎 𝑥 ≤ 𝑏 𝑎𝑛𝑑 − 𝑎 𝑥 − 𝑎 𝑥 ≤ −𝑏 .
Now, we have L.P. problem with three constraint and two variables. Now , find the dual
of this, for this suppose 𝑦 , 𝑦 , 𝑦 are the dual variables. Then, we get dual
Minimize 𝑤 = 𝑏 (𝑦 − 𝑦 ) + 𝑏 𝑦 subject to constraint 𝑎 (𝑦 − 𝑦 ) + 𝑎 𝑦 ≥ 𝑐 ,

𝑎 (𝑦 − 𝑦 ) + 𝑎 𝑦 ≥ 𝑐 , 𝑎𝑛𝑑 𝑦 , 𝑦 , 𝑦 ≥0 .
We observe that, 𝑦 − 𝑦 is in the objective function as well as in the constraint. If we

put 𝑦 − 𝑦 = 𝑦 , which is the difference of the two non-negative numbers. So, 𝑦 may
be positive, negative or zero i.e. it is unrestricted in sign. The dual problem becomes
Minimize 𝑤 = 𝑏 𝑦 + 𝑏 𝑦 subject to constraint
𝑎 𝑦 + 𝑎 𝑦 ≥ 𝑐 , 𝑎 𝑦 + 𝑎 𝑦 ≥ 𝑐 𝑎𝑛𝑑 𝑦 ≥ 0 , 𝑢𝑛𝑟𝑒𝑠𝑡𝑟𝑖𝑐𝑡𝑒𝑑 𝑖𝑛 𝑠𝑖𝑔𝑛 𝑦 .
Hence, the dual variable which corresponds to an equality constraint must be

unrestricted in sign. Similarly, we can prove this for 𝑛 variables and 𝑚 constraint L.P.
problem.
04-03: SOLUTION OF DUAL PROBLEM

In this section, we have to study some important result on duality, principle of
complementary slackness and the correspondence between primal and dual optimal
solution.
A) The standard result on duality:
1. The dual of the dual is the primal.
2. The value of the objective function Z for any feasible solution of the primal is
≤ the value of the objective function W for any feasible solution of the dual.
3. If either the primal or the dual problem has an unbounded solution, then the
solution to the other problem is infeasible.

4. The main duality theorem or Fundamental theorem of duality: If both the
primal and the dual problems have feasible solutions, then both have optimal
solution and max. Z=min.W.
5. Complementary slackness property of primal-dual relationship states that for a

positive basic variable in the primal, the corresponding dual variable will be
equal to zero. Alternatively, for a non -basic variable in the primal (which is
zero), the corresponding dual variable will be basic and positive.
B) Correspondence between Primal and Dual Optimal Solution:
(i) Value for the non-basic variable of the primal are given by the base row of the
dual solution, under the slack variable (if there are any), neglecting the – 𝑣𝑒
sign if any and under the artificial variable (if there is no slack variable in a
constraint) neglecting the – 𝑣𝑒 sign if any, after deleting the constant M.
(ii) Value of the slack variable of the primal are given by the base row under the
non-basic variable of the dual solution neglecting – 𝑣𝑒 sign if any.
(iii) The value of the objective function is same for primal and dual solution.
We take some examples on so it will be clear more perfectly.
SOLVED PROBLEMS 03
Problem 01-04-05: Use duality to solve the problem Minimize 𝑍 = 𝑥 − 𝑥 subject to
constraint 2𝑥 – 𝑥 ≥ 2, −𝑥 + 𝑥 ≥ 1 𝑎𝑛𝑑 𝑥 , 𝑥 ≥ 0 .
Solution: The dual of the give primal problem is
Maximize 𝑤 = 2𝑦 + 𝑦
subject to constraint 2𝑦 − 𝑦 ≤ 1, − 𝑦 + 𝑦 ≤ −1 , 𝑦 , 𝑦 ≥ 0.
Step I: Set up the problem in the standard form:
Multiply the second dual constraint both side by−1, it becomes
𝑦 − 𝑦 ≥ 1.
Now, write the above problem in standard L.P. form for this, we need to introduce the
slack and surplus variable.
Maximize 𝑤 = 2𝑦 + 𝑦 + 0𝑠 + 𝑠 − 𝑀𝐴
subject to constraint 2𝑦 − 𝑦 + 𝑠 = 1,
𝑦 −𝑦 −𝑠 +𝐴 =1,

𝑦 , 𝑦 , 𝑠 , 𝑠 , 𝐴 ≥ 0.
Step II: Find an initial basic feasible solution:
Put 𝑦 = 0, 𝑦 = 0, 𝑠 = 0, we get 𝑠 = 1, 𝐴 = 1, 𝑊 = −𝑀.
Since the basic variable 𝑠 , 𝐴 are positive, the above basic solution for the artificial
system is also feasible and non-degenerate. Then, the simplex table is
𝑐 2 1 0 0 -M 𝑏 𝜃
𝑐 Basis 𝑦 𝑦 𝑠 𝑠 𝐴 𝑥 =
1/2
0 𝑠 (2) -1 1 0 0 1
-M 𝐴 1 -1 0 -1 1 1 1
𝑍 = ∑𝑐 𝑎 -M M 0 M -M -M
2+M
𝑐 −𝑍 1-M 0 -M 0
I.B.F.S.
Since 𝑐 − 𝑍 is positive under 𝑦 column, I.B.F.S. is not optimal.
Step III: iterate towards an optimal solution:
From table we see that 𝑠 is outgoing variable and 𝑦 is incoming variable. Replace 𝑠
by 𝑦 . Now, make key element 2 is one and reaming element in this column zero by
performing suitable row operations. Then, the next simplex table is as follows.
𝑐 2 1 0 0 -M
𝑏
𝑥
𝑐 Basis 𝑦 𝑦 𝑠 𝑠 𝐴
2 𝑦 1 -1/2 1/2 0 0 1/2
-M 𝐴 0 -1/2 -1/2 -1 1 1/2
𝑍 = ∑𝑐 𝑎 2 M/2 1+M/2 M -M 1-M
0
𝑐 −𝑍 0 -1+M/2 -1-M/2 M
Second B.F.S.

Since all 𝑐 − 𝑍 is negative or zero all columns but the artificial variable 𝐴 appears in
the basis at non-zero level ½, the dual problem has no feasible solution.
It follows, that the given primal problem also has infeasible solution.
Problem 01-04-06: Solve the L.P.P. by using its dual: Maximize 𝑍 = 5𝑥 − 2𝑥 + 3𝑥

subject to constraints 2𝑥 + 2𝑥 − 𝑥 ≥ 2, 3𝑥 − 4𝑥 ≤ 3,
𝑥 + 3𝑥 ≤ 5 𝑎𝑛𝑑 𝑥 , 𝑥 , 𝑥 ≥ 0 .
Solution: In given problem first constraint is of ≥ type and problem is of maximization

so multiply first constraint by −1, Then we get the primal in the form
Maximize 𝑍 = 5𝑥 − 2𝑥 + 3𝑥 subject to constraint −2𝑥 − 2𝑥 + 𝑥 ≤ −2, 3𝑥 −

4𝑥 ≤ 3, 𝑥 + 3𝑥 ≤ 5 𝑎𝑛𝑑 𝑥 , 𝑥 , 𝑥 ≥ 0.
The dual of the given primal is given by
Minimize 𝑤 = −2𝑦 + 3𝑦 + 5𝑦 subject to the constraint
−2𝑦 + 3𝑦 ≥ 5, − 2𝑦 − 4𝑦 + 𝑦 ≥ −2, 𝑦 + 3𝑦 ≥ 3 𝑎𝑛𝑑 𝑦 , 𝑦 , 𝑦 ≥ 0.
Step I: Express the problem in the standard form:
In the above, primal second constraint RHS is negative, make it positive by

multiplying −1. Then we get 2𝑦 + 4𝑦 − 𝑦 ≤ −2 .
Now, introduce slack, surplus and artificial variables for standard L .P.P.
Minimize 𝑤 = −2𝑦 + 3𝑦 + 5𝑦 + 0𝑠 + 0𝑠 + 0𝑠 + 𝑀𝐴 + 𝑀𝐴
subject to the constraint −2𝑦 + 3𝑦 − 𝑠 + 𝐴 = 5,
2𝑦 + 4𝑦 − 𝑦 + 𝑠 = 2,
𝑦 + 3𝑦 − 𝑠 + 𝐴 = 3
𝑎𝑛𝑑 𝑦 , 𝑦 , 𝑦 , 𝑠 , 𝑠 , 𝑠 , 𝐴 , 𝐴 ≥ 0.
Step 2: Find the initial basic feasible solution:
Putting 𝑦 = 𝑦 = 𝑦 = 𝑠 = 𝑠 = 0, we get 𝐴 = 5, 𝑠 = 2, 𝐴 = 3 are the basic feasible

solution to the system.
Then, the simplex table is
𝑐 -2 3 5 0 0 0 M M 𝑏 𝜃
𝑐 Basis 𝑦 𝑦 𝑦 𝑠 𝑠 𝑠 𝐴 𝐴 𝑥 =

M 𝐴 -2 3 0 -1 0 0 1 0 5 5/3
0 𝑠 2 (4) -1 0 1 0 0 0 2 1/2
∞
M 𝐴 1 0 3 0 0 -1 0 1 3
𝑍 - -
-M 3M 3M 0 M M
= ∑𝑐 𝑎 M M
- 3- 5-
𝑐 −𝑍 3M M 0 M 0 0
2+M 3M I.B.F.S.
Since the 𝑐 − 𝑧 is negative under some column, above table not gives optimal
solution.
Step 3: Iterate toward an optimal solution.
Iterating thrice, we get, the final simplex table
𝑐 -2 3 5 0 0 0 M M 𝑏
𝑐 Basis 𝑦 𝑦 𝑦 𝑠 𝑠 𝑠 𝐴 𝐴 𝑥
0 𝑠 -15 0 0 -4 -3 1 4 -1 11
3 𝑦 -2/3 1 0 -1/3 0 0 1/3 0 5/3
5 𝑦 -14/3 0 1 -4/3 -1 0 4/3 0 14/3
𝑍 = ∑𝑐 𝑎 -76/3 3 5 -23/3 5 0 23/3 0
M
𝑐 −𝑍 70/3 0 0 23/3 5 0 -23/3+M
O.B.F.S.
The optimal solution is
𝑦 = 0, 𝑦 = , 𝑦 = , 𝑀𝑖𝑛 𝑊 = .
𝑇ℎ𝑒 𝑜𝑝𝑡𝑖𝑚𝑎𝑙 𝑏𝑎𝑠𝑖𝑐 𝑓𝑒𝑎𝑠𝑖𝑏𝑙𝑒 𝑠𝑜𝑙𝑢𝑡𝑜𝑝𝑛 𝑓𝑜𝑟 𝑡ℎ𝑒 𝑔𝑖𝑣𝑒𝑛 𝑝𝑟𝑖𝑚𝑎𝑙 𝑝𝑟𝑜𝑏𝑙𝑒𝑚 𝑖𝑠
𝑥 = , 𝑥 = 5, 𝑥 = 0 𝑎𝑛𝑑 max 𝑍 = .
SELF-TEST 03
MCQ 04-05: If the primal ha an unbounded solution, then the solution to the dual is
a) Optimal.

b) Feasible.
c) Unbounded.
d) Infeasible.
MCQ 04-06: The both the primal and dual problem has feasible solutions, then
a) Both have optimal solution.
b) Both have not optimal solution.
c) Both have optimal solution and max. Z = min. Z.
d) Both have optimal solution and max. Z ≠ min. Z.

SAQ 01-04-05: Write the dual of Maximize 𝑍 = 5𝑥 + 8𝑥 + 10𝑥 subject to
constrain 𝑥 + 𝑥 + 2𝑥 ≤ 120, 3𝑥 − 2𝑥 − 𝑥 ≥ 90,
2𝑥 + 4𝑥 + 2𝑥 = 100 𝑎𝑛𝑑 𝑥 , 𝑥 , 𝑥 ≥ 0.
Use simplex method and obtain the zeroth and first iteration of the dual.
Solution: After simplification and assigning 𝑦 , 𝑦 , 𝑦 , 𝑦 be the associated dual

variable. Then the dual problem is
𝑚𝑖𝑛𝑖𝑚𝑖𝑧𝑒 𝑊 = 120𝑦 − 90 + 100(𝑦 − 𝑦 )
Subject to 𝑦 − 3𝑦 + 2(𝑦 − 𝑦 ) ≥ 5, 𝑦 + 2𝑦 + 4(𝑦 − 𝑦 ) ≥ 8,
2𝑦 + 𝑦 + 2(𝑦 − 𝑦 ) ≥ 10 𝑎𝑛𝑑 𝑦 , 𝑦 , 𝑦 , 𝑦 ≥ 0.
The solution of this dual problem by simplex method consists of the following steps
Step I: Set up problem in the standard form:
Introducing surplus and artificial variable, we get an artificial system given by
𝑚𝑖𝑛𝑖𝑚𝑖𝑧𝑒 𝑊 = 120𝑦 − 90 + 100𝑦 − 100𝑦 + 0𝑠 + 0𝑠 + 0𝑠 + 𝑀𝐴 + 𝑀𝐴 + 𝑀𝐴
Subject to 𝑦 − 3𝑦 + 2𝑦 − 2𝑦 − 𝑠 + 𝐴 = 5,
𝑦 + 2𝑦 + 4𝑦 − 4𝑦 − 𝑠 + 𝐴 = 8,
2𝑦 + 𝑦 + 2𝑦 − 2𝑦 − 𝑠 + 𝐴 = 10
𝑎𝑛𝑑 𝑦 , 𝑦 , 𝑦 , 𝑦 , 𝑠 , 𝑠 , 𝑠 , 𝐴 , 𝐴 , 𝐴 ≥ 0.
Step II: Find an initial basic solution:
Setting 𝑦 , 𝑦 , 𝑦 , 𝑦 , 𝑠 , 𝑠 , 𝑠 , each zero, the I.B.F.S. to the artificial system is
𝑦 = 𝑦 = 𝑦 = 𝑦 = 𝑠 = 𝑠 = 𝑠 = 0 and 𝐴 = 5, 𝐴 = 8, 𝐴 = 10 𝑎𝑛𝑑 𝑊 = 23𝑀

Now, we write the zeroth iterate and represent this solution
𝑐 120 -90 100 -100 0 0 0 M M M b 𝑏

𝑎
𝑐 Basis 𝑦 𝑦 𝑦 𝑦 𝑠 𝑠 𝑠 𝐴 𝐴 𝐴
M 𝐴 1 -3 2 -2 -1 0 0 1 0 0 5 5/2
M 𝐴 1 2 (4) -4 0 -1 0 0 1 0 8 2
M 𝐴 2 1 2 -2 0 0 -1 0 0 0 10 5
𝑍 4M 0 8M -8M -M -M -M M M M 23M
𝑐 120- -90 100- - M M M 0 0 0

−𝑍 4M 8M 100+8M
Zeroth iterate(initial solution)

Step III: Perform optimality test
Since 𝑐 − 𝑍 is negative under some column, in above table solution is not optimal.
Step IV: Iterate towards an optimal solution:
Performing iteration to get first iteration result in the following table :
𝑐 120 -90 100 -100 0 0 0 M M M b 𝑏

𝑎
𝑐 Basi 𝑦 𝑦 𝑦 𝑦 𝑠 𝑠 𝑠 𝐴 𝐴 𝐴
s
M 𝐴 (1/2) -4 0 0 - 1/2 0 1 -1/2 0 1 2

1
10 𝑦 1/4 1/2 1 -1 0 -1/4 0 0 1/4 0 2 8

0
M 𝐴 3/2 0 1 0 0 1/2 - 0 -1/2 0 6 4

1
𝑍 25+2 50- 100 -100 - - - M 25-M M 200

M 4M M 25+ M
+7
M
M
𝑐 95-2M - 0 0 M 25-M M 0 - 0
−𝑍 140+4 25+
M M
Above table is the first iterate and yields the following solution to the dual problem:
𝐴 = 1, 𝑦 = 2, 𝐴 = 6 𝑊 = 200 + 7𝑀.

SAQ 01-04-06: Prove that the dual of dual is primal.
Solution: Let the primal problem in the canonical form be
Maximize 𝑍 = 𝑐 𝑥 + 𝑐 𝑥 + ⋯+ 𝑐 𝑥
Subject to 𝑎 𝑥 + 𝑎 𝑥 + ⋯+ 𝑎 𝑥 ≤𝑏 ,
𝑎 𝑥 + 𝑎 𝑥 + ⋯+ 𝑎 𝑥 ≤𝑏 , ….(1)
……………………………………...
𝑎 𝑥 + 𝑎 𝑥 + ⋯+ 𝑎 𝑥 ≤𝑏 ,
where 𝑥 , 𝑥 , … . , 𝑥 ≥ 0;
then the dual of this problem is
Minimize 𝑊 = 𝑏 𝑦 + 𝑏 𝑦 + ⋯+ 𝑏 𝑦
Subject to 𝑎 𝑦 + 𝑎 𝑦 + ⋯+ 𝑎 𝑦 ≥ 𝑐 ,
𝑎 𝑦 + 𝑎 𝑦 + ⋯+ 𝑎 𝑦 ≥ 2, ….. (2)
……………………………………...
𝑎 𝑦 + 𝑎 𝑦 + ⋯+ 𝑎 𝑦 ≥ 𝑐 ,
where 𝑦 , 𝑦 , … . , 𝑦 ≥ 0 are the associated dual variables.
Multiply equation (2) by -1, we get
Minimize −𝑊 = −𝑏 𝑦 − 𝑏 𝑦 − ⋯ − 𝑏 𝑦
Subject to −𝑎 𝑦 − 𝑎 𝑦 − ⋯ − 𝑎 𝑦 ≤−𝑐 ,
−𝑎 𝑦 − 𝑎 𝑦 − ⋯ − 𝑎 𝑦 ≤ − 2,
……………………………………... ..(3)
−𝑎 𝑦 − 𝑎 𝑦 − ⋯ − 𝑎 𝑦 ≤ −𝑐 ,
where 𝑦 , 𝑦 , … . , 𝑦 ≥ 0 are the associated dual variables.
By construction dual of the equation (3) and multiplying by -1 we obtain the system
identical to relation (1). Hence, the dual of dual is primal.
04-04: THE DUAL SIMPLEX METHOD

In this Section we have to study simplex algorithm which is related to the primal
and dual problem. It is used to solve the problem which start dual feasible i.e. whose
primal is optimal but infeasible. The dual simplex method was first discovered by C.E.
Lemke, a student of Charnes. The dual simplex method is quite similar to the regular

simplex method. It eliminates the introduction of artificial variable in the L .P. problem.
This is the advantage over the regular simplex method.
The dual simplex algorithm to solve L.P. problem consists of the following steps:
Step I: Convert the problem into maximization problem if it is in the minimization

form.
Step II: Convert ≥ type constraint, if any into ≤ type by multiplying both sides of
such constraint by −1.
Step III: Then convert the L.P.P. in the standard form introducing slack variables.
Then, express the information in the dual simplex table.
Step IV: Compute 𝑐 − 𝑍 for every column. There are three cases
a) If all 𝑐 − 𝑍 ≤ 0 and all 𝑏 are nonnegative, the solution obtained above is

the optimal basic feasible solution.
b) If all 𝑐 − 𝑍 ≤ 0 and at least one 𝑏 are negative, then proceed to step-5.
c) If any 𝑐 − 𝑍 is positive, then methods fail.
Step V: Select the row that contain the most negative 𝑏 . This row is called key row
or pivot row. The corresponding basic variable leaves the basis. This is called the
dual feasible condition.
Step VI: Observe the key row.
i) If all the elements are non-negative, then the problem does not have a
feasible solution.
ii) If at least one element is negative, then find the ratio of the corresponding
element of 𝑐 − 𝑍 row to these elements. Chose the smallest of this ratio.
The corresponding column is the key column and associated variable is the
entering variable. this is the dual optimality condition. Mark the key element
or the pivot element.
Step VII: Make the key element unity. Perform the row operation as in regular
simplex method and repeat iterations until the optimal solution is obtained in a
finite number of steps or there is a condition of nonexistence of a feasible solution.
Flow chart is shown in figure:

SOLVED PROBLEMS 04
Problem 01-04-07: Use the dual simplex method to Maximize 𝑍 = −3𝑥 − 2𝑥 subject
to 𝑥 + 𝑥 ≥ 1, 𝑥 + 𝑥 ≤ 7, 𝑥 + 2𝑥 ≥ 10, 𝑥 ≤ 3, 𝑥 , 𝑥 ≥ 0.
Solution: Multiply first and third constraint by −1,The primal is in the form
Maximize 𝑍 = −3𝑥 − 2𝑥 subject to
−𝑥 − 𝑥 ≤ −1, 𝑥 + 𝑥 ≤ 7, −𝑥 − 2𝑥 ≤ −10, 𝑥 ≤ 3, 𝑥 , 𝑥 ≥ 0.
Now, introduce the slack variable for standard L.P.P., we get
Maximize 𝑍 = −3𝑥 − 2𝑥 + 0𝑠 + 0𝑠 + 0𝑠 + 0𝑠
subject to −𝑥 − 𝑥 + 𝑠 = −1,
𝑥 + 𝑥 + 𝑠 = 7,

− 𝑥 − 2𝑥 + 𝑠 = −10,
𝑥 + 𝑥 = 3,
𝑥 , 𝑥 , 𝑠 , 𝑠 , 𝑠 , 𝑠 ≥ 0.
Put 𝑥 = 0, 𝑥 = 0, we get 𝑠 = −1, 𝑠 = 7, 𝑠 = −10, 𝑠 = 3 initial basic feasible

solution. We express this information in the table , we get dual simplex table.
𝑐 -3 -2 0 0 0 0 𝑏
𝑐 Basis 𝑥 𝑥 𝑠 𝑠 𝑠 𝑠
𝑥
0 𝑠 -1 -1 1 0 0 0 -1
0 𝑠 1 1 0 1 0 0 7
0 𝑠 -1 (-2) 0 0 1 0 -10
0 𝑠 0 1 0 0 0 1 3
𝑍 = ∑𝑐 𝑎 0 0 0 0 0 0 0
𝑐 −𝑍 -3 -2 0 0 0 0
𝑐 −𝑍 1 I.B.F.S.
3 - - - -
𝑎
From the table, we see that 𝑥 is incoming and 𝑠 is outgoing variable. −2 is the key
element. Make key element is unity and reaming entries in this column zero
performing suitable row operations. We get new table
𝑐 -3 -2 0 0 0 0 𝑏
𝑥
0 𝑠 -1/2 0 1 0 -1/2 0 4
0 𝑠 1/2 0 0 1 1/2 0 2
-2 𝑥 1/2 1 0 0 -1/2 0 5
0 𝑠 (-1/2) 0 0 0 1/2 1 -2
𝑍 = ∑𝑐 𝑎 -1 -2 0 0 1 0 -10
𝑐 −𝑍 -2 0 0 0 -1 0
𝑐 −𝑍 4 Second F.S.
- - - - -
𝑎
Replace 𝑠 by 𝑥 , we get
𝑐 -3 -2 0 0 0 0 𝑏
𝑥
0 𝑠 0 0 1 0 -1 -1 6
0 𝑠 0 0 0 1 1 1 0
-2 𝑥 0 1 0 0 0 1 3

-3 𝑥 1 0 0 0 1 -2 4
𝑍 = ∑𝑐 𝑎 -3 -2 0 0 3 4 -18
𝑐 −𝑍 0 0 0 0 -3 -4
Above table gives optimal solution, 𝑥 = 4, 𝑥 = 3 𝑎𝑛𝑑 𝑍 = −18. optimal F.S.
Problem 01-04-08: Use the dual simplex method to Maximize 𝑍 = 5𝑥 + 6𝑥 subject to

𝑥 + 𝑥 ≥ 2, 4 𝑥 + 𝑥 ≥ 4, 𝑥 , 𝑥 ≥ 0.
Solution: Now, we convert the minimization problem in to maximization problem.
Maximize 𝑍 = −5𝑥 − 6𝑥 subject to −𝑥 − 𝑥 ≤ −2, −4 𝑥 − 𝑥 ≤ −4, 𝑥 , 𝑥 ≥ 0.
Now write the above problem in standard L.P. problem
Maximize 𝑍 = −5𝑥 − 6𝑥 + 0𝑥 + 0𝑥
subject to −𝑥 − 𝑥 + 𝑠 = −2, −4 𝑥 − 𝑥 + 𝑠 = −4, 𝑥 , 𝑥 , 𝑠 , 𝑠 ≥ 0.
Initial basic solution is given by 𝑥 = 0, 𝑥 = 0, 𝑠 = −2, 𝑠 = −4 .
Now, write the dual simplex table.
𝑐 -5 -6 0 0 𝑏
𝑥
0 𝑠 -1 -1 1 0 -2
0 𝑠 (-4) -1 0 1 -4
𝑍 = ∑𝑐 𝑎 0 0 0 0 0
𝑐 −𝑍 5 6 0 0
𝑐 −𝑍 5/4 I.B.F.S.
6 - -
𝑎
Replace 𝑠 by 𝑥
𝑐 -5 -6 0 0 𝑏
𝑥
0 𝑠 0 (-3/4) 1 -1/4 -1
-5 𝑥 1 1/4 0 -1/4 1
𝑍 = ∑𝑐 𝑎 -5 -5/4 0 5/4 -5
𝑐 −𝑍 0 -19/4 0 -5/4
𝑐 −𝑍 5 𝟐𝒏𝒅 FS
- 19/3 -
𝑎
Replace 𝑠 by 𝑠 .
𝑐 -5 -6 0 0 𝑏
𝑥

0 𝑠 0 3 -4 1 4
-5 𝑥 1 1 -1 0 2
𝑍 = ∑𝑐 𝑎 -5 -5 5 0 -10
𝑐 −𝑍 0 -1 -5 0
Opt. S.
Since, all 𝑐 − 𝑍 are non-positive, we get optimal feasible solution.
𝑥 = 2, 𝑥 = 0, 𝑍 = −10
Thus, 𝑍 = −𝑍 = −(10) = 10.
SELF-TEST 04
MCQ 04-07: Dual constrain for the minimization L.P. problem can be written as
a) ∑ 𝑎 𝑦 ≥ 𝑐 .
b) ∑ 𝑎 𝑦 ≤ 𝑐 .
c) ∑ 𝑎 𝑦 ≥ 𝑐 .
d) ∑ 𝑎 𝑦 ≤ 𝑐 .
MCQ 04-08: In the dual simplex table if 𝑐 − 𝑍 > 0, then L.P. problem have
a) Basic feasible solution.
b) No basic feasible solution.
c) Feasible solution.
d) Method fails.

SAQ 01-04-07: Use the dual simplex method to Minimize 𝑍 = 𝑥 + 𝑥
subject to 2𝑥 + 𝑥 ≥ 2, − 𝑥 − 𝑥 ≥ 1, 𝑥 , 𝑥 ≥ 0.
Solution: We express the given problem in the following way
Maximize 𝑍 = −𝑍 = −𝑥 − 𝑥
subject to −2𝑥 − 𝑥 ≤ −2, 𝑥 + 𝑥 ≤ −1, 𝑥 , 𝑥 ≥ 0.
Adding slack variables 𝑠 , 𝑠 , the problem can be expressed as
Maximize 𝑍 = −𝑍 = −𝑥 − 𝑥 + 0𝑠
subject to −2𝑥 − 𝑥 + 𝑠 = −2, 𝑥 + 𝑥 + 𝑠 = −1, 𝑥 , 𝑥 , 𝑠 , 𝑠 ≥ 0.
Hence I.B.F.S. is 𝑥 = 0, 𝑥 = 0, 𝑠 = −2, 𝑠 = −1 , 𝑍 = 0, which is an infeasible solution

This can be expressed in the following table.
𝑐 -1 -1 0 0
𝑐 Basis 𝑥 𝑥 𝑠 𝑠 𝑏
0 𝑠 (-2) -1 1 0 -2
0 𝑠 1 1 0 1 -1
𝑍 0 0 0 0 0
𝑐 −𝑍 -1 -1 0 0
𝑐 −𝑍
1/2 1
𝑎
𝑥 − 𝑐𝑜𝑙𝑢𝑚𝑛 is the key column and (-2) is the key element, 𝑠 replaced by 𝑥 in table
below:
𝑐 -1 -1 0 0
𝑐 Basis 𝑥 𝑥 𝑠 𝑠 𝑏
𝑠 1 1/2 -1/2 0 1
𝑠 0 1/2 1/2 1 -2
𝑍 -1 -1/2 1/2 0 -1
𝑐 −𝑍 0 -1/2 -1/2 0
Second basic infeasible solution.
Compute 𝑐 − 𝑍 row. Since 𝑐 − 𝑍 coefficients are either negative or zero under all
columns and 𝑏 is negative, the second basic solution is optimal but infeasible. Thus,
the given L.P.P. has no feasible solution.
SAQ 01-04-08: Draw the flow chart for the dual simplex method.
Solution: The flow chart for the dual simplex method is as below

SUMMARY
In this unit, we study the duality in linear programming. If the primal is
maximization the then its dual is minimization type. If the dual contains the m variable
and m constraint then its dual contains m variable and n constraint. Also, we discussed
about solution comparison of primal and dual solution. If both primal and dual have
feasible solution then both have equal optimal solution. Also, if ether primal or dual
have an unbounded solution the other have infeasible solution. The dual simplex
method is useful to solve dual feasible, i.e. who’s primal is optimal but infeasible. It is
similar to the regular simplex method just change is that to select key -row first and then
key-column.

1. Define the dual of a linear programming problem. Explain the primal-dual relationship.
2. Discuss briefly correspondence between solution of primal and its dual.
3. What is duality? Explain the dual simplex algorithm briefly.
4. Write the dual of the primal L.P. problem
𝑀𝑎𝑥 𝑍 = 2𝑥 + 5𝑥 + 6𝑥
subject to constraints
5𝑥 + 6𝑥 – 𝑥 ≤ 3,
– 2𝑥 + 3𝑥 + 4𝑥 3 ≤ 4,
𝑥 – 5𝑥 + 3𝑥 ≤ 1,
– 3𝑥 – 3𝑥 + 7𝑥 ≤ 6
𝑎𝑛𝑑 𝑥 , 𝑥 , 𝑥 ≥ 0.
5. Write the dual of the primal 𝑀𝑖𝑛 𝑍 = 2𝑥 + 3𝑥 + 4𝑥
subject to 2𝑥 + 3𝑥 + 5𝑥 ≥ 2, 3𝑥 + 3𝑥 + 7𝑥 = 3, 𝑥 + 4𝑥 + 6𝑥 ≤ 5
and 𝑥 , 𝑥 , ≥ 0, 𝑥 is unrestricted.
6. Use the dual simplex method to Minimize 𝑍 = 20𝑥 + 2𝑥 subject to 𝑥 + 𝑥 ≥

12, 2𝑥 + 𝑥 ≥ 17, 𝑥 ≥ 2.5, 𝑥 ≥ 6, 𝑥 , 𝑥 ≥ 0.
7. One unit of product A contributes Rs. 7 and requires 3 units of raw material and
2 hours of labour. One unit of product B contributes Rs. 5 and requires one unit
of raw material and one hour of labour. Availability of raw material at present is
48 units and there are 40 hours of labour.

(a) Formulate this problem as a linear programming problem.
(b) Write its dual.
(c) Solve the dual by the simplex method and find the optimal product mix and
the shadow prices of the raw material and labour .
8. What do you understand duality? Write some important features of pri mal and
dual problem.
KEY WORDS
Primal, Dual, Dual Simplex Method.
REFERENCES
BOOKS
& COMPANY LTD

PRAGATI EDITION
MOOCS
YOUTUBE VIDEOS
i) https://youtu.be/H0sBZdjuego
ii) https://youtu.be/KLHWtBpPbEc
iii) https://youtu.be/c1V6zbjAc5I
iv) https://youtu.be/W2rYOpvkSMQ
WIKIPEDIA
https://en.wikipedia.org/wiki/Duality_(optimization)
OER


:
C REDIT 02

UNIT 02-01: GAME MODELS AND RELATED THEORY
LEARNING OBJECTIVES
 Understand the basic concepts of competitive games
 Understand the characteristics of two-person zero sum game
 Understand the types of strategies chosen by player in the game
 Construct the pay-off matrix for a two-person zero sum game
 Understand the optimal strategies for both the players in two -person zero sum
game
INTRODUCTION
“Game Theory”, a branch of applied mathematics was originally developed by
the American mathematician John Von Neumann called Father of Game Theory and his
collogue Morgenstern, a German born American economist, to solve problems in
economics. It provides tools for analyzing situations in which parties called Players
make decisions that are independent. This interdependence causes each player to
consider other player’s possible decisions or strategies in formulating his own strategy.
Game Theory does not describe how a game should be played but it describes
only the procedure and principles by which plays should be selected so that each
competitor will act so as to minimize his maximum loss (or maximize his minimum
gain).
01-01: COMPETITIVE GAME

A competitive situation will be called “A Competitive Game” if it has following
six properties:
1. It has finite number of participants (i.e., No. of Participants(n) ≥2)
2. Each participant has finite number of possible courses of action
3. Each participant must know all the courses of action available to other but
unaware of which of these will be chosen
4. A play of game occurs when each player chooses one of his courses of actio n.
The choices must be made simultaneously, so that none of them knows choice
of other until he has decided his own

5. After all participants have chosen a course of action, their respective gains are
finite
6. The gain of participant depends upon his own action s as well as those of others.
Remark: We are considering the games which involve competition, actions and
counteractions will be dealt with. Hence, we will use simple term “game” in place of
“competitive games”.
Definition: Player
Each participant (interested party) in game is called player.
Definition: Strategy
The decision rule by which a player determines his course of action is called a
Strategy.
Note: To reach the decision regarding which strategy to use, neither player needs to
know the other’s strategy.
Types of Strategies:
There are two types of strategies a player can go for:
1. Pure Strategy
2. Mixed Strategy
Let us study each of these in detail.
Definition: Pure Strategy
If a player decides to use only one particular course of action during every play,
then he is said to use pure strategy.
Notation: A pure strategy is usually represented by a number with which the course
action is associated.
Definition: Mixed Strategy
If a player decides in advance, to use all or some of his available courses of

action in some fixed proportion, he is said to use mixed strategy.
Remark: Mixed strategy is a selection among pure strategies with some fixed
probabilities (proportions).
Notation: A mixed strategy of a player with m-possible courses of action is denoted by

a set X of m non-negative numbers. The sum of these numbers is unity and each number
represents the probability with which each course of action is chosen . Thus, if

𝑥 = 𝑇ℎ𝑒 𝑃𝑟𝑜𝑏𝑎𝑏𝑖𝑙𝑖𝑡𝑦 𝑜𝑓 𝑐ℎ𝑜𝑜𝑠𝑖𝑛𝑔 𝑐𝑜𝑢𝑟𝑠𝑒 𝑖
Then, we have𝑋 = (𝑥 , 𝑥 , … , 𝑥 ), where ∑ 𝑥 = 1 and 𝑥 ≥ 0; 𝑖 = 1, 2, 3, … , 𝑚
Note:
(i) Clearly, we can say that pure strategy is a special case of mixed strategy, where
all but one 𝑥 is zero.
(ii) Thus, a player may be able to choose only m-pure strategies, but he has an
infinite number of mixed strategies to choose them.
Definition: n-Person Game
If the number of participants (n) in the game is greater than 2, then it is call ed n-
Person game.
In particular, for n=2, the game is called two -person game.
Definition: Two-Person Zero-Sum Game
A game with two players, where gain of one player equals the loss to the other is
known as a Two-Person Zero-Sum game.
Remark: In two-person zero-sum game, the interests of the two players are opposed so
that the sum of their net gains is zero.
Definition: n-Person Zero-Sum Game
If there are n-players in a game and sum of the game (i. e. sum of their net
gains) is zero, then it is called n-Person Zero-Sum game.
Definition: Pay-off
Pay-off is the outcome of playing the game.
Definition: Pay-off Matrix
A Pay-off matrix of a game is a table in which strategies of one player are listed
in rows and those of other player in columns and the cells show pay-offs to each player
such that the pay-off of the row player is listed first.
In other words, a pay-off matrix of a game is a table showing the amounts

received by the player named at the left-hand side after all possible plays of the game.
The payment is made by the player named at the top of the table.
SOLVED PROBLEMS 01
Problem 02-01-01: Consider a two-person zero sum game represented by the following
table in which we have two competitors A and B. Both competitors assumed to be eq ual

in ability and intelligence. A has a choice of strategy 1 or 2 while B can select strategy
3 or 4.
Competitor B
3 4 Min of row
1 4 6 4
Competitor A [ ]
2 3 5 3
Max of column 4 6
Give the interpretation of table and hence find value of game.
Solution: Both the competitors A and B know the pay-offs for every possible strategy.
As, all the values in pay-off matrix are positive, we can say the game favors A (since
values that favors B are negative). So, game is biased against B.
However, since B must play game, the attempt of B will be selection of

strategy to minimize his losses.
Interpretation of table:
If A selects strategy 1 and B selects strategy 3, then A receives amount 4 from B (i. e. B
has to pay 4 points to A) and on similar lines for other strategies.
The various possible strategies for the two competitors are;
1) A wins highest game value if he plays strategy 1 all the time; since it has
higher values than strategy 2.
2) B realizes this situation and plays strategy 3 to minimize his lo ss. Since, the
value of 4 in strategy 3 is less than value of 6 in strategy 4.
Thus, here game value=04, since A wins 4 points while B loses 4 points each
time the game is played.
Problem 02-01-02: Players A and B play a game in which each has three coins, a 1p, 5p
and 10p. Each selects a coin without the knowledge of other’s choice. If the sum of
coins is an odd amount, then A wins B’s coin. But if the sum is even, then B wins A’s
coin. Write pay-off matrix of this game.
Solution: In this game, we have two players both having same set of coins viz. 1p, 5p
and 10p. The conditions for playing the game are as follows:
1) If the sum of coins is odd then A wins B’s coin.
2) If the sum of coins is even then B wins A’s coin.

Then, using the given conditions. The pay-off matrix for player A can be constructed as
follows:
If both players have 1p then sum of the amount is 2p which is even number an d hence B
wins A’s amount i. e.1p. As this will represent loss of 1p to A, in pay -off matrix we
write -1. On the similar lines, if player A has 5p and player B has 10p then, sum of the
amount is 15p which is odd number and hence A wins B’s coin i. e. 10p. As this will
represent gain of 10p to A, in pay-off matrix we write 10. Continuing in this fashion,
we have the following pay-off matrix for the given game.
Player B
1𝑝 5𝑝 10𝑝
1𝑝 −1 −1 10
Player A [ ]
5𝑝 −5 −5 10
10𝑝 1 5 −10
SELF-TEST 01
MCQ01-01:If a player decides to use only one particular course of action during every
play, then he is said to use _________.
a) Mixed Strategy
b) Pure Strategy
c) Both
d) None of these
MCQ 01-02:A game with n-players is known as n-person zero-sum game, if the sum of
their net gains is _________.
a) One
b) Zero
c) Two
d) Negative

SAQ 02-01-01: What is meant by competitive game?
Solution: A competitive situation will be called “A Competitive Game” if it has

following six properties:
1. It has finite number of participants (i.e., No. of Participants(n) ≥2)
2. Each participant has finite number of possible courses of action

3. Each participant must know all the courses of action available to other but
unaware of which of these will be chosen
4. A play of game occurs when each player chooses one of his courses of action.
The choices must be made simultaneously, so that none of them knows choice
of other until he has decided his own
5. After all participants have chosen a course of action, their respective gains are
finite
6. The gain of participant depends upon his own actions as well as those of others.
SAQ 02-01-02: Write a short note on strategies for a game.
Solution: The decision rule by which a player determines his course of action is called
a Strategy. There are two types of strategies a player can go for:
1) Pure Strategy: If a player decides to use only one particular course of action during
every play, then he is said to use pure strategy. A pure strategy is usually
represented by a number with which the course action is associated.
2) Mixed Strategy: If a player decides in advance, to use all or some of his available
courses of action in some fixed proportion, he is said to use mixed strategy. Mixed
strategy is a selection among pure strategies with some fixed probabilities
(proportions).
A mixed strategy of a player with m-possible courses of action is denoted by a

set X of m non-negative numbers. The sum of these numbers is unity and each
number represents the probability with which each course of action is chosen. Thus,
if
𝑥 = 𝑇ℎ𝑒 𝑃𝑟𝑜𝑏𝑎𝑏𝑖𝑙𝑖𝑡𝑦 𝑜𝑓 𝑐ℎ𝑜𝑜𝑠𝑖𝑛𝑔 𝑐𝑜𝑢𝑟𝑠𝑒 𝑖
Then, we have 𝑋 = (𝑥 , 𝑥 , … , 𝑥 ), where ∑ 𝑥 = 1 and 𝑥 ≥ 0; 𝑖 = 1, 2, 3, … , 𝑚
Clearly, we can say that pure strategy is a special case of mixed strategy, where all
but one 𝑥 are zero.
01-02: PAY-OFF MATRIX

We have already seen that pay-off matrix of game is a table in which strategies
of one player are listed in rows and those of the other player in columns and the cells
show payoffs to each player such that the payoff of the row player is listed first. In this
section we will study the representation of payoff matrix.
REPRESENTATION OF PAY-OFF MATRIX:

Consider a game with two players viz. A and B. If a player A has m -courses of
action and a player B has n-courses of action then pay-off matrix can be constructed by
following steps:
a) Row destinations for each matrix are the courses of action available to A
b) Column destinations for each matrix are the courses of action available to B
c) With a two-person zero-sum game, the cell entries in B’s pay-off matrix will be
the negative of the corresponding entries in A’s pay -off matrix.
The pay-off matrices for A and B are represented as follows:
Player B
1 2 3 … 𝑗 … 𝑛
1 𝑎 𝑎 𝑎 … 𝑎 … 𝑎
2 𝑎 𝑎 𝑎 … 𝑎 … 𝑎
3 𝑎 𝑎 𝑎 … 𝑎 … 𝑎
Player A
⋮ ⋮ ⋮ ⋮ ⋱ ⋮ ⋱ ⋮
𝑖 𝑎 𝑎 𝑎 … 𝑎 … 𝑎
⋮ ⋮ ⋮ ⋮ ⋱ ⋮ ⋱ ⋮
[𝑚 𝑎 𝑎 𝑎 … 𝑎 … 𝑎 ]
A's Pay-off Matrix
Player B
1 2 3 … 𝑗 … 𝑛
1 −𝑎 −𝑎 −𝑎 … −𝑎 … −𝑎
2 −𝑎 −𝑎 −𝑎 … −𝑎 … −𝑎
3 −𝑎 −𝑎 −𝑎 … −𝑎 … −𝑎
Player A
⋮ ⋮ ⋮ ⋮ ⋱ ⋮ ⋱ ⋮
𝑖 −𝑎 −𝑎 −𝑎 … −𝑎 … −𝑎
⋮ ⋮ ⋮ ⋮ ⋱ ⋮ ⋱ ⋮
[ 𝑚 −𝑎 −𝑎 −𝑎 … −𝑎 … −𝑎 ]
B's Pay-off Matrix
Remark:
1) The sum of pay-off matrices for A and B is null (zero) matrix.
2) We usually omit B’s pay-off matrix, keeping in mind that it is just the negative
of A’s pay-off matrix.

3) The objective is to determine optimum strategies of both players that results in
optimum pay-off to each, irrespective of the strategy used by the other.
SOLVED PROBLEMS 02
Problem 02-01-03: Two players are there. Each secretly thinking an integer between 1
and 3 inclusive. Both players reveal their numbers simultaneously. If the sum of
numbers is even, then R player wins. A number of unit(s) from the C player to R player
equals to the difference of the numbers provided that the numbers are distinct. If the
numbers are same, then R player wins the number of unit(s) from the C player equal to
the sum of the numbers. If the sum of numbers is odd then, C player wins 3 units from
the R player.
Formulate the pay-off matrix for this game.
Solution: There are two players in this game viz. player R and player C. Both players
secretly thinking an integer between 1 and 3 inclusive and both players reveal their
numbers simultaneously. Now, the no. of units to be given from one player to another
player depends on the sum of the numbers revealed by both the players as follows:
(i) If the sum of numbers is even, then R wins an amount equal to d ifference of
two numbers, if they are distinct otherwise R wins an amount equal to the sum
of two numbers.
(ii) If the sum of numbers is odd, then C wins 3 units from R.
Hence, the pay-off matrix (w. r. t. row player R) is given by;
Player S
𝑆𝑡𝑟𝑎𝑡𝑒𝑔𝑖𝑒𝑠 1 2 3
1 2 −3 2
Player R [ ]
2 −3 4 −3
3 2 −3 6
Problem 02-01-04: If the pay-off matrix for row-player of two-person zero-sum game
is given by
Player B
𝐼 𝐼𝐼 𝐼𝐼𝐼 𝐼𝑉
𝐼 5 4 1 0
Player A 𝐼𝐼 4 3 2 −1
𝐼𝐼𝐼 0 −1 4 3
[ 𝐼𝑉 1 −2 1 2]
Find the pay-off matrix for column player.

Solution: We know that with a two-person zero-sum game, the cell entries in column
player’s pay-off matrix will be the negative of the corresponding entries in row player’s
pay-off matrix. Thus, pay-off matrix for player B is given by
Player B
𝐼 𝐼𝐼 𝐼𝐼𝐼 𝐼𝑉
𝐼 −5 −4 −1 0
Player A 𝐼𝐼 −4 −3 −2 1
𝐼𝐼𝐼 0 1 −4 −3
[ 𝐼𝑉 −1 2 −1 −2]
SELF-TEST 02
MCQ01-03: Consider a game with two persons viz. A and B (where A is row player and
B is column player). If player A has 3 courses of action and player B has 4 courses of
action then the payoff matrix has order ________.
a) 3×4
b) 4×3
c) 3×3
d) 4×4
MCQ01-04: Consider a two-person zero-sum game with the pay-off matrix given by
1 −1
[ ]. If row player chooses strategy 2 and column player chooses strategy 1, then
−2 3
a) Row player gains 2 units
b) Column player gains 2 units
c) Column player loses 2 units
d) Row player loses 3 units

SAQ 02-01-03: Write a note on Pay-off matrix for a game.
Solution: pay-off matrix of game is a table in which strategies of one player are listed
in rows and those of the other player in columns and the cells show payoffs to each
player such that the payoff of the row player is listed first.
Consider a game with two players viz. A and B. If a player A has m-courses of action
and a player B has n-courses of action then pay-off matrix can be constructed by
following steps:
a) Row destinations for each matrix are the courses of action available to A

b) Column destinations for each matrix are the courses of action available to B
c) With a two-person zero-sum game, the cell entries in B’s pay-off matrix will be
the negative of the corresponding entries in A’s pay -off matrix.
The pay-off matrices for A and B are represented as follows:
Player B
1 2 3 … 𝑗 … 𝑛
1 𝑎 𝑎 𝑎 … 𝑎 … 𝑎
2 𝑎 𝑎 𝑎 … 𝑎 … 𝑎
3 𝑎 𝑎 𝑎 … 𝑎 … 𝑎
Player A
⋮ ⋮ ⋮ ⋮ ⋱ ⋮ ⋱ ⋮
⋮ ⋮ ⋮ ⋮ ⋱ ⋮ ⋱ ⋮
[ 𝑚 𝑎 𝑎 𝑎 … 𝑎 … 𝑎 ]
A's Pay-off Matrix
Player B
1 2 3 … 𝑗 … 𝑛
1 −𝑎 −𝑎 −𝑎 … −𝑎 … −𝑎
2 −𝑎 −𝑎 −𝑎 … −𝑎 … −𝑎
3 −𝑎 −𝑎 −𝑎 … −𝑎 … −𝑎
Player A
⋮ ⋮ ⋮ ⋮ ⋱ ⋮ ⋱ ⋮
𝑖 −𝑎 −𝑎 −𝑎 … −𝑎 … −𝑎
⋮ ⋮ ⋮ ⋮ ⋱ ⋮ ⋱ ⋮
[𝑚 −𝑎 −𝑎 −𝑎 … −𝑎 … −𝑎 ]
B's Pay-off Matrix
SAQ 02-01-04: Construct a pay-off matrix for the following Rock Paper Scissors game:
In the game of Rock Paper Scissors, the players face each other and simultaneously
display their hands in one of the following three shapes: a fist denoting rock, the
forefinger and middle finger extended and spread so as to suggest scissors, or a
downward facing palm denoting a sheet of paper. The rock wins over the scissors since
it can shatter them, the scissors win over the paper since they can cut it, and the paper
wins over the rock since it can be wrapped around it. The winner collects a penny from
the opponent and no money changes hands in case of a tie.
Solution: The pay-off matrix for game of Rock Paper Scissors is given by;

Rock Paper Scissors
Rock 0 -1 +1
Paper +1 0 -1
Scissors -1 +1 0
01-03: TWO-PERSON ZERO SUM GAME

Two-person zero-sum game is a game with two players, where gain of one
player equals the loss to the other.
CHARACTERISTICS OF TWO-PERSON ZERO-SUM GAME:
1) Only two players can participate
2) Each player has finite number of strategies to use
3) Each specific strategy results in a pay-off
4) Total pay-off to the two players at the end of each play is zero
SOLVED PROBLEMS 03
Problem 02-01-05: Consider the following two-person zero-sum game
Player 2 (She)
𝐴 𝐵 𝐶
Player 1 (He)[ 1 2 −1 −3]
2 0 −5 6
3 4 3 5
Find: i) Optimal Strategies for Player 1 and Player 2.
ii) Game Value
Solution: The row minimum and column maximum for the given game are obtained as
follows:
Player 2 (She) Row Min
𝐴 𝐵 𝐶
−3
Player 1 (He)[ 1 2 −1 −3
] −5
2 0 −5 6
3 4 3 5
Column Max 4 6
If player 1 choose strategy 1, then player 2 will select strategy C so as to pay minimum
amount. Similarly, if player 1 selects strategy 2, then player 2 will go for strategy B. On
similar fashion, we have to find row minimum for each row and then out of ro w

minimum values player 1 will select that row for which value of row minimum is
largest so that player 1 will receive maximum amount from player 2. i. e. max {-3, -5,
3} =3
Thus, Optimal Strategy for player 1 is 3.
Now, for player 2,
If she will select strategy A, then player 1 will select strategy 3 for maximum gain.
Therefore, we need to find column maximum for each column. Then, from these column
maxima, player 2 will go for minimum value so as to reduce the loss i.e. min {4, 3, 6} =
3.
Therefore, Optimal Strategy for player 2 is B.
Thus, Game Value=03.
Problem 02-01-06:Consider the following two-person zero-sum game:
8 6 -5 5
2 -4 -1 6
8 7 -4 3
-5 6 1 8
5 -3 -1 8
a) Find the optimal strategies for both the row and column players.
b) Which player is favored by the game?
Solution:
i) Let us denote row player by ‘A’, with five strategies as 1, 2, 3, 4, 5 and the
column player by ‘B’ with four strategies 1, 2, 3, 4. The pay-off matrix along
with row minimum and column maximum is given by
Player B
1 2 3 4 𝑅𝑜𝑤 𝑀𝑖𝑛
1 8 6 −5 5 −5
2 2 −4 −1 6 −4
Player A
3 8 7 −4 3 −4
4 −5 6 1 8 −5
[5 5 −3 −1 8] −3
𝐶𝑜𝑙𝑢𝑚𝑛 𝑀𝑎𝑥 8 7 8
Player B will choose the strategy which will give minimum pay -off to A. Thus,
for each strategy chosen by A, B will choose strategy with minimum value.
Hence, we find row minima. Among these values, A will select the maximum
one so as to get maximum pay-off.

Thus, max {row minimum} = max {-5, -4, -4, -5, -3} = -3
Therefore, Optimal Strategy for A is 5.
Now, for each strategy chosen by B, A will select t he strategy for which he will
get maximum pay-off. Therefore, we find column maxima. Among these values B will
select the one with minimum so as to reduce loss.
Therefore, min {column maximum} = min {8, 7, 1, 8} =1
Thus, Optimal Strategy for B is 3.
ii) From the pay-off matrix, we can observe that for each strategy chosen by A, he will
always get negative pay-off i. e. no gain is obtained by A for any strategy.
Thus, it means player B (column player) is favored by the game.
SELF-TEST 03
MCQ01-05:If in a two-person zero-sum game, the pay-off matrix for row player is
+1 −1
[ ], then the pay-off matrix for column player is
−1 +2
+1 −1
a) [ ]
−1 +2
−1 +1
b) [ ]
+1 −2
+1 −1
c) [ ]
+1 +2
+1 +1
d) [ ]
+1 +2
MCQ01-06:In a two-person zero-sum game with pay-off matrix for row player given by
1 2
1 1 2
2 −1 3
[ ]
3 4 5
4 2 3
Then, which of the following is optimal strategy for row player?
a) 1
b) 2
c) 3
d) 4

SAQ 02-01-05: What is meant by two-person zero-sum game? Write down its
characteristics.
Solution: Two-person zero-sum game is a game with two players, where gain of one
player equals the loss to the other.
Characteristics of two-person zero-sum game are as follows:
2) Each player has finite number of strategies t o use
4) Total pay-off to the two players at the end of each play is zero
SAQ 02-01-06:Give an example of two-person zero-sum game. Also, write down the
pay-off matrices for both the players.
Solution: Use any one example from the given set of examples and write the pay -off
matrix for row player. To obtain pay-off matrix for column player multiply all the
entries in pay-off matrix of row player by -1.
01-04: GAME VALUE

The game value is the average winnings per play over a long number of plays.
Remarks:
1. The average amount per play that A will win in long run if A and B use their
best strategies (it is nothing but the value of game).
2. The strategy that A should use to ensure that his average gain per play is at least
equal to the value of the game i.e.
Average gain of A per play ≥ Game Value
3. The strategy that B should use to ensure that his average loss per play is no
more than the value of game i.e.
Average loss of B per play ≤Game Value
4. The optimal value of game will occur somewhere between the max {row
minimum} and min {column maximum} i. e.
Max {Row minimum}≤Value of the game ≤ Min {Column maximum}

SOLVED PROBLEMS 04
Problem 02-01-07: Consider the game given by the following pay-off matrix
1 2 Min of Row
1 2 2 2
A [ ]
2 3 4 3
Max of Column 3 4
Find the value of game. Also, predict about the average gain of A per play and
average loss of B per play.
Solution: Here, A will have maximum gain for strategy 2. Therefore, B will select the
strategy so that he will have minimum loss.
Thus, Game Value=03 with
Average gain of A per play ≥ 3 and
Average loss of B per play ≤ 3
Problem 02-01-08: Find the value of following two-person zero-sum game.
𝑆𝑡𝑟𝑎𝑡𝑒𝑔𝑖𝑒𝑠 1 2
1 2 3
A [ ]
2 −1 4
3 1 2
Solution: Here, we will find out row minimum and column maximum for each row and
column. Then, we will find out the maximum among row minima so that A will have
maximum gain. We will also find out minimum of column maxima so that B will have
minimum loss.
1 2 𝑅𝑜𝑤 𝑀𝑖𝑛
1 2 3 2
A 2 [−1 4] −1
3 1 2 1
𝐶𝑜𝑙𝑢𝑚𝑛 𝑀𝑎𝑥 2 4
Max {row minimum} =max {2, -1, 1} = 2
Min {column maximum} = min {2, 4} = 2

Thus, Optimal strategies for A and B are 1 and 1 respectively.
Game Value= 2.
SELF-TEST 04
MCQ 01-07: The average winnings per play over a long number of plays is known as
________.
a) Pay-off
b) Game Value
c) Net Gain
d) None of these
MCQ 01-08: Which of the following is true?
a) Game value is total number of winnings per play
b) Average gain per play of row player ≤ Game Value
c) Average gain per play of row player ≥ Game Value
d) Average loss per play of column player ≥ Game Value

SAQ 02-01-07: Write a short note on value of game.
Solution: The game value is the average winnings per play over a long number of
plays.
1. The average amount per play that A will win in long run if A and B use their
best strategies (it is nothing but the value of game).
2. The strategy that A should use to ensure that hi s average gain per play is at least
equal to the value of the game i.e.
Average gain of A per play ≥ Game Value
3. The strategy that B should use to ensure that his average loss per play is no
more than the value of game i.e.
Average loss of B per play ≤ Game Value
4. The optimal value of game will occur somewhere between the max {row
minimum} and min {column maximum} i. e.
Max {Row minimum} ≤ Value of the game ≤ Min {Column maximum}.
SAQ 02-01-08: Find the value of game represented by the following pay -off matrix:

Player B
1 2 3
1 12 40 20
Player A 2 [28 15 22]
3 35 27 25
Solution: Initially, we find maximum of row minimum and minimum of column

maximum;
Player B
1 2 3 Row min
1 12 40 20 12
Player A 2 [28 15 22] 15
3 35 27 25 25
Column Max 35 40 25
Here, Max {12, 15, 25} = 25; Min {35, 40, 25} = 25.
Therefore, Game Value = 25.
SUMMARY
In this unit, we have studied the competitive games. In particular, two -person
zero-sum games. We have also learned the way to construct pay-off matrix for a game
and to determine optimal strategies for players in two -person zero-sum game.

1. Two companies A and B, sell two brands of flu medicine. Company A advertises
in radio (𝐴 ), television (𝐴 ), and newspapers (𝐴 ). Company B, in addition to
using radio (𝐵 ), television (𝐵 ), and newspapers (𝐵 ), also mails brochures (𝐵 ).
Depending on the effectiveness of each advertising campaign, one company can
capture a portion of the market from the other. The following matrix summarizes
the percentage of the market captured or lost by company A.
𝐵 𝐵 𝐵 𝐵
𝐴 8 −2 9 −3
𝐴 [6 5 6 8]
𝐴 −2 4 −9 5
Find the value of game.

2. Two players A and B, play the coin-tossing game. Each player, unbeknownst to
each other, chooses a head (H) or a tail (T). Both players would reveal their
choices simultaneously. If they match (HH or TT), player A receives $1 from B.
Otherwise, A pays B $1. Prepare a pay-off matrix and also find row minimum
and column maximum values. Hence, specify the range of value of game.
3. Following matrix represents a pay-off matrix for player A (row player). Find the
pay-off matrix for player B (column player).
𝐵 𝐵 𝐵 𝐵
𝐴 8 6 2 8
𝐴 [8 9 4 5]
𝐴 7 5 3 5
4. Find the value of game by determining the row minima and column maxima for
the pay-off matrix given by
𝐵 𝐵 𝐵 𝐵
𝐴 4 −4 −5 6
𝐴 −3 −4 −9 −2
[ ]
𝐴 6 7 −8 −9
𝐴 7 3 −9 5
5. In the game below, the pay-off is for player A. Specify the range for the value of
the game.
𝐵 𝐵 𝐵 𝐵
𝐴 1 9 6 0
𝐴 2 3 8 4
[ ]
𝐴 −5 −2 10 −3
𝐴 7 4 −2 −5
6. In the game below, the pay-off is for player A. Specify the range for the value of
the game.
𝐵 𝐵 𝐵
𝐴 3 6 1
𝐴 [5 2 3]
𝐴 4 2 −5
7. Find the value of the game with pay-off matrix for row player given by:
𝐵 𝐵 𝐵 𝐵
𝐴 3 7 1 3
𝐴 [4 8 0 −6]
𝐴 6 −9 −2 4

8. Explain the following terms in detail:
(1) Strategy
(2) Pay-off matrix
(3) Value of game
KEY WORDS
Player, Strategy, Pay-off, n-person zero-sum game, game value, courses of action
REFERENCES
BOOKS
& COMPANY LTD
MOOCS
YOUTUBE VIDEOS
1) https://youtu.be/tXCBPgdn4xk
2) https://youtu.be/lpKY85HCXmA
3) https://youtu.be/2u-gm2mMjg8
WIKIPEDIA
Coin matching game:
Two Person Games (Setting up the Pay-off Matrix) Mathematical Game …
https://www3.nd.edu/~apilking/Math10120/Lectures/Topic%2025. pdf
OER
Visit Here for Course Companion website for this course:

UNIT 02-02: TWO PERSON ZERO SUM GAME
LEARNING OBJECTIVES
 Understand maximin and minimax principles for a game
 Understand the concepts of fair and strictly determinable game
 Understand the concept of saddle point for a game and use of it to find the value
of game
INTRODUCTION
Two-person zero-sum game is a particular case of n-person zero-sum game in which
n=2 i. e. which involves 2 players. In such a game, the interests of 2 players are
opposed so that the sum of their net gains are zero. Tw o-person zero-sum games are
also called ‘Rectangular Games’ because their pay-off matrix is in the rectangular form.
In this unit we are primarily concerned with the two -person zero-sum games
only. Let us recall the definition of two-person zero-sum game which we have studied
in previous unit.
02-01: TWO-PERSON ZERO-SUM GAME

A game with two players, where a gain of one player equals the loss to the other
is known as ‘Two-person zero-sum game’.
SOLVED PROBLEMS 01
Problem 02-02-01(Two-finger Morra game): Suppose A and B play the game, called
“Two-finger Morra”. In this game, both the players simultaneously show either one or
two fingers. If the number of fingers of one coincides with the number of other then
player A wins and get Rs. 1.00 from B. If they don’t coincide, then B wins and takes
Rs. 1.00 from A. Construct the pay-off matrix for this game.
Solution: As the game involves only two players and winnings of one player causes
loss to the other player, this is an example of two -person zero-sum game.
The number of rows of the matrix corresponds to the number of courses of

action of the player A and the number of columns corresponds to the number of courses
of action of player B. The table shows the payments or gains to A for each possible
outcome of the game. The negative entriesin the table denote the payments from A to B.
The pay-off matrix of the player B is just the negative of the pay -off matrix written for
player A, because we know that in a zero-sum game the gains of one are the negatives

of the gains of the other. Thus, we generally write the pay-off matrix of a player which
is written on the left to the matrix.
For the player B we may consider this as a matrix which represents the losses.
The player A is called the maximizing player and B the minimiz ing player. The pay-off
matrix for the game is given by
1 finger 2 fingers
1 finger 1 -1
A
2 fingers -1 1
Problem 02-02-02: Robin travels between two cities and may use one of the two
routes: Route A is a fast four-lane highway, and route B is a long winding road. She has
the habit of driving “superfast”. The highway patrol has a limited police force. If the
full force is allocated to the route driven by Robin, she is certain to receive a $100
speeding fine. If the force is split 50-50 between the two routes, there is a 50% chance
of getting a $100 fine on route A, and only a 30% chance of getting the same fine on
route B. Develop a pay-off matrix for the police patrol.
Solution: We consider police as a row player and robin as a column player. Depending
on the route chosen by robin and the allocation of police force on both the routes, robin
has to pay fine. This is a two-person zero-sum game which favors police.
The pay-off matrix for police patrol is:
100% A 50% A- 50% B 100% B
A 100 50 0
B 0 30 100
SELF-TEST 01
MCQ 02-01: The row player in two-person zero-sum game is also known as ________.
a) Maximizing Player
b) Minimizing Player
c) Winner
d) Looser

MCQ 02-02: The column player in two-person zero-sum game is also known as
________.
a) Maximizing Player
b) Minimizing Player
c) Winner
d) Looser
MCQ02-03:The two-person zero-sum game is also known as ________.
a) Square game
b) N-player game
c) Rectangular game
d) None of these

SAQ 02-02-01: Write a short note on properties of two-person zero-sum game.
Solution: A game with two players, where a gain of one player equals the loss to the
other is known as ‘Two-person zero-sum game’. Two-person zero-sum game is a
particular case of n-person zero-sum game in which n=2 i. e. which involves 2 players.
In such a game, the interests of 2 players are opposed so that the sum of their net gains
are zero. Two-person zero-sum games are also called ‘Rectangular Games’ because their
pay-off matrix is in the rectangular form. Properties of two -person zero-sum game are
as follows:
2) Each player has finite number of strategies to use
4) Total pay-off to the two players at the end of each play is zero .
SAQ 02-02-02:Explain the two-person zero-sum game with the help of one example.
Solution: A game with two players, where a gain of one player equals the loss to the
other is known as ‘Two-person zero-sum game’. Consider the following example:
I II
I 1 -1
A
II -1 1
As the game involves only two players and winnings of one player causes loss to the
other player, this is an example of two -person zero-sum game.
The number of rows of the matrix corresponds to the number of courses of

action of the player A and the number of columns corresponds to the number of courses
of action of player B. The table shows the payments or gains to A for each possible
outcome of the game. The negative entries in the table denote the payments from A to
B. The pay-off matrix of the player B is just the negative of the pay -off matrix written
for player A, because we know that in a zero-sum game the gains of one are the
negatives of the gains of the other. Thus, we generally write the pay -off matrix of a
player which is written on the left to the matrix.
02-02: MAXIMIN AND MINIMAX PRINCIPLES

In game theory, we determine the best strategies for each player on the basis of
maximin and minimax criterion of optimality. Mini max, as the name suggests, is a
method in decision theory for minimizing the maximum loss. On the other hand,
maximizing the minimum gain, is known as Maximin. Let us discuss these criterions in
detail.
Definition: Maximin Value
In a game, the maximin value is the highest value that the player can be sure to
get without knowing the actions of the other players.
Equivalently, it is the lowest value the other players can force the player to
receive, when they know the player’s action.
Remark: “Maximin” is a term commonly used for non-zero sum games, to describe the
strategy which maximizes one’s own minimum pay -off.
Definition: Minimax Value
In a game, the minimax value of a player is the smallest value that the other
players can force the player to receive, without knowing the other player’s actions.
Equivalently, it is the largest value the player can be sure to g et when they know
the actions of the other players.
Remark: “Minimax” is used in zero-sum games, to determine minimizing the

opponent’s maximum pay-off. This is identical to minimizing one’s own maximum loss,
and to maximizing one’s own minimum gain.

From the conditions governing minimax and maximin values, we can say that
Minimax Value ≥ Maximin Value
Thus, we can state the two principles as follows:
MAXIMIN PRINCIPLE:
“Maximize the player’s minimum gains” i. e. select the strategy that gives the
maximum gains among the row minimum value.
MINIMAX PRINCIPLE:
“Minimize the player’s maximum loss” i. e. select the strategy that gives the minimum
loss among the column maximum values.
SOLVED PROBLEMS 02
Problem 02-02-03: Consider the game given by the pay-off matrix
3 2
A[−2 −3]
−4 −5
Find the maximin and minimax values for this game.
Solution: For any two-person zero-sum game, the minimax value is obtained by finding
minimum value for each row (Row minimum) followed by identifying maximum value
among them. On the similar lines, to obtain minimax value, first find the maximum
value for each column (Column Maxima) and then identify minimum amongst them.
For the given pay-off matrix, the row minima and column maxima are given by,
B Row Min
A 3 2 2(Maximin)
-2 -3 -3
-4 -5 -5
Column Max 3 2(Minimax)
Here, Maximin value= max {2, -3, -5} =2
Minimax value= min {3, 2} =2
Problem 02-02-04: For the two-person game represented by the following pay-off
matrix, check whether maximin value and minimax values are same or not.

2 5
P[ ]
3 −3
Solution: For the given pay-off matrix, the row minima and column maxima are given
by,
B Row Min
A 2 5 2(Maximin)
3 -3 -3
Column Max 3(Minimax) 5
Here, Maximin value= max {2, -3} =2; Minimax value= min {3, 5} =3
So, we can say that Maximin Value ≠ Minimax Value
SELF-TEST 02
MCQ 02-04: Selection of the strategy that gives the maximum gains among the row
minimum value is
a) Minimax Principle
b) Maximin Principle
c) Mixed Strategy
d) None of these
MCQ 02-05: Which of the following is minimax principle?
a) Maximize the player’s minimum gains
b) Minimize the player’s maximum loss
c) Both
d) None of these

SAQ 02-02-03: Write a short note on “Maximin” and “Minimax” principles of a game.
Solution: The maximin and minimax principles of a game are as follows:
MAXIMIN PRINCIPLE:
“Maximize the player’s minimum gains” i. e. select the strategy that gives the
maximum gains among the row minimum value.
MINIMAX PRINCIPLE:

“Minimize the player’s maximum loss” i. e. select the strategy that gives the minimum
loss among the column maximum values.
The relation so obtained is: Minimax Value ≥ Maximin Value.
SAQ 02-02-04: Explain with the help of example: Maximin and Minimax value of a
game.
Solution: Consider the following pay-off matrix for two-person zero-sum game:
Q Row min
1 5 1
P[ ]
3 2 2
Column Max 3 5
Maximin value is obtained by taking maximum of row minimum entries and Minimax
value is obtained by taking minimum of column maximum entries. Here,
Maximin = Max {1, 2} = 2 and Minimax = Min {3, 5} = 3.
02-03: SADDLE POINT

Definition: Saddle Point (Equilibrium Point)
When maximin value and minimax value are equal, then the corresponding pure
strategies are called “Optimal Strategies” and the game is said to have a “Saddle Point”.
The saddle point is the point of intersection of the two courses of action and the
gain at this point is the value of the game.
Remarks:
I. Saddle point is the number which is lowest in its row and highest in its column.
II. Neither player can improve his position by selecting any other strategy .
Steps required to detect Saddle point:
1) At the right of each row, write the row minimum and ring the largest of them.
2) At the bottom of each column, write the column maximum and ring the smallest
of them.
3) If these two elements are same, the cell where the corresponding row and
column meet is a saddle point and the element in that cell is the value of the
game.
4) If there are more than one saddle points then there will be more than one
solution, each solution corresponding to each saddle point.

5) If the two ringed elements are unequal, then there is no saddle point and in such
case the value of the game lies between these two values.
SOLVED PROBLEMS 03
Problem 02-02-05: For the following two-person zero-sum game
ii) Determine the saddle point
iii) Find the optimal strategies for both the players
iv) Find the value of game
16 4 0 14 −2
10 8 6 10 12
A [ ]
2 6 4 8 14
8 10 2 2 0
Solution: To determine the saddle points, we will find out maximin and minimax values
for the above pay-off matrix.
Players B Row Min

16 4 0 14 -2 -2
10 8 6 10 12 6
2 6 4 8 14 2
A
8 10 2 2 0 0
Column 16 10 6 14 14
Max
Here, Maximin= Max {Row Minimum} = Max {-2, 6, 2, 0} = 6
Minimax= Min {Column Maximum} = Min {16, 10, 6, 14, 4} = 6
We have, Maximin= 6 =Minimax
⇒ Saddle point exists and it is (2,3).
Thus, we have following conclusion for the strategies for both the players:
The optimal strategy for player A is 2 and the optimal strategy for player B is 3.
Value of Game= 6.
Problem 02-02-06:Find the ranges of the values of p and q which will render the entry
(2, 2) a saddle point for the game.
Players B

2 4 5
10 7 q
A
4 p 6
Solution:Initially ignoring the values of p and q, we determine the maximin and

minimax values of payoff matrix as follows:
Let us denote the strategies for player A by 𝐴 , 𝐴 , 𝐴 and that for player B by
𝐵,𝐵 ,𝐵 .
Players B Row
Min
Strategies 𝐵 𝐵 𝐵
𝐴 2 4 5 2
A
𝐴 10 7 q 7
𝐴 4 p 6 4
Column Max 10 7 6
Thus, Maximin value= 7 and Minimax value= 7
Therefore, for 7 to be minimum value in 2 nd row, q≥7 and for 7 to be maximum value in
2 nd column p≤7.
Hence, the range of p and q will be p≤7, q≥7.
SELF-TEST 03
MCQ 02-06: Which of the following is true for a game with pay -off matrix given by
−4 3
[ ]
−3 −7
a) Maximin value= -4
b) Minimax value= -3
c) -4 < v <-3, where v is value of game
d) All of these
MCQ02-07: The number which is lowest in its row and highest in its column is known
as
a) Minimax value
b) Maximin value
c) Saddle point

d) Game value

SAQ 02-02-05: What is meant by saddle point? Describe the steps to find saddle point.
Solution: When maximin value and minimax value are equal, then the corresponding
pure strategies are called “Optimal Strategies” and the game is said to have a “Saddle
Point”.
The saddle point is the point of intersection of the two courses of action and the
gain at this point is the value of the game.
Steps required to detect Saddle point:
1) At the right of each row, write the row minimum and ring the largest of them.
2) At the bottom of each column, write the column maximum and ring the smalles t
of them.
3) If these two elements are same, the cell where the corresponding row and
column meet is a saddle point and the element in that cell is the value of the
game.
4) If there are more than one saddle points then there will be more than one
solution, each solution corresponding to each saddle point.
5) If the two ringed elements are unequal, then there is no saddle point and in such
case the value of the game lies between these two values.
SAQ 02-02-06: Define: Value of game. How to find value of game if i) it has a saddle
point ii) it does not have a saddle point .
Solution: The game value is the average winnings per play over a long number of
plays.
The saddle point of the game is the point for which maximin value = minimax value. If
not, the game is said to be without saddle point.
For a game with saddle point; Maximin value = Minimax value = Value of game.
For a game without saddle point, Maximin value ≤ value of game ≤ Minimax value.
02-04: CLASSIFICATION OF GAMES

Depending upon the maximin and minimax values for a game, we can classify the game
as follows:
FAIR GAME:

The game is said to be fair if, Maximin value= Minimax value= 0.
STRICTLY DETERMINABLE GAME:
The game is said to be strictly determinable if, Maximin value= Minimax value ≠0.
SOLVED PROBLEMS 04
Problem 02-02-07:In a certain game, player A has three possible choices L , M and N,
while the player B has two possible choices P and Q. Payments are to be made
according to the choices made.
Choices Payment
L, P A pays B Rs. 3
L, Q B pays A Rs. 3
M, P A pays B Rs. 2
M, Q B pays A Rs. 4
N, P B pays A Rs. 2
N, Q B pays A Rs. 3
What are the best strategies for players A and B in this game? What is the value of
game for A and B?
Solution: Let, positive number represent payment from B to A and negative number
represent payment from A to B.
Then, the payments can be easily arranged in the form of a table as follows:
Players Player B
Strategies P Q Row Min
L -3 3 -3
Player A M -2 4 -2
N 2 3 2(Maximin)
Column Max 2(Minimax) 4
To find the maximin value, find out the minimum value in each row and then identify
maximum value among them. i. e. max { -3, -2, 2} = 2.
Thus, Maximin= 2
Similarly, to find minimax value, find out the maximum value corresponding to each
column and then identify smallest value among them i.e. min {2, 4} = 2
Thus, Minimax= 2.
Therefore, in this game, Maximin= Minimax= 2 (Saddle point exists at (3, 1)).
⇒ The game is strictly determinable.

We can conclude, best strategy for player A is the strategy corresponding to maxim in
value i.e. N whereas best strategy for player B is the strategy corresponding to minimax
value i.e. P. and value of the game= 2.
3 2
Problem 02-02-08: Consider the game given by the pay-off matrix [−2 −3]. Is it a
−4 −5
fair game? What are the optimal strategies for row player and column player? Also find
the value of game.
Solution: Let us denote row player by ‘A’ and column player by ‘B’. To determine the
type of game, we need to find the maximin and minimax values. Let us obtain row
minima and column maxima.
Players B Row Min

3 2 2(Maximin)
-2 -3 -3
A
-4 -5 -5
Column Max 3 2(Minimax)
Here, Maximin value= Max {2, -3, -5} = 2
Minimax value= Min {3, 2} = 2
⇒ Maximin = Minimax = 2 ≠ 0
⇒ The game is not fair. Moreover, it is strictly determinable game which has saddle
point (1, 2).
Thus, we have the following conclusions:
Optimal Strategy for A is ‘1’ and that for B is ‘2’ withthe value of game= 2.
SELF-TEST 04
MCQ 02-08: If the value of the game is V=5. Then, the game is ________.
a) Fair
b) Strictly determinable
c) Complex
d) Simple
MCQ 02-09: The game is said to be fair if
a) Maximin value = Minimax value ≠ 0
b) Maximin value = Minimax value = 0

c) Maximin value ≠ Minimax value
d) Maximin value ≥ Minimax value
MCQ 02-10: The game is said to be strictly determinable if
a) Maximin value = Minimax value ≠ 0
b) Maximin value = Minimax value = 0
c) Maximin value ≠ Minimax value
d) Maximin value ≥ Minimax value

SAQ 02-02-07: For the following two-person zero-sum game, find saddle point,
optimal strategies for both the players and game value.
1 13 11
A [−9 5 −11]
0 −3 13
Solution: Maximin Value = Max {Row min} = Max {1, -11, -3} = 1;
Minimax Value = Min {Column Max} = Min {1, 13, 13} = 1.
As, maximin value = minimax value; the game has saddle point at (1, 1) and value of
game = 1 with (1, 1) as optimal strategies for (A, B).
SAQ 02-02-08: Write a short note on characterization of games using the maximin
value and minimax value. Also, give one example of each type of game.
Solution: Depending upon the maximin and minimax values for a game, we can
classify the game as follows:
FAIR GAME:
The game is said to be fair if, Maximin value= Minimax value= 0.
−3 −2 6
For example: [ 2 0 2 ] is pay-off matrix for the game which is fair as maximin
5 −2 −4
and minimax values are 0.
STRICTLY DETERMINABLE GAME:
The game is said to be strictly determinable if, Maximin value= Minimax value ≠0.

−1 2 −2
For example: [ ] is pay-off matrix for the game which is strictly determinable
6 4 −6
as maximin and minimax values are -2.
SUMMARY
In this unit, we have studied two-person zero-sum game with saddle point in detail. We
have also studied the maximin and minimax principles so as t o detect the saddle point,
characterize the type of game and to find value of game.

1. Write a note on two-person zero-sum game with saddle point.
2. Define: Fair game and justify with the help of example.
3. Define: Strictly determinable game and justify with the help of example.
4. Solve the game whose pay-off matrix is
Strategies I II III
I -3 -2 6
II 2 0 2
III 5 -2 -4
Also write down the type of game and optimal strategies for both the players.
5. Find the range of values for p and q which will render the entry (2, 2) a saddle
point in the game with the following pay-off matrix:
6.
Players Player 1
Strategies 1 2 3
1 1 q 3
2 p 5 10
Player 3 6 2 3
2
7. Check whether the following game has saddle point or not. Also, find the game
value and optimal strategies for both the players.
16 4 0 14 -2
10 8 6 10 12
2 6 4 8 14

8 10 2 2 0
8. Check whether the following game has saddle point or not. Also, find the game
value and optimal strategies for both the players.
1 13 11
-9 5 -11
0 -3 13
9. Check whether the following game is fairor strictly determinable game.
2 -1 -3
0 -5 6
4 3 6
KEY WORDS
Two-person zero-sum game, maximin, minimax, saddle point, fair game, strictly
determinable game.
REFERENCES
BOOKS
& COMPANY LTD

PRAGATI EDITION
MOOCS
example of saddle point game and mixed strategy pdf
http://people.maths.ox.ac.uk/griffit4/Math_Alive/3/game_theory_notes.pdf
YOUTUBE VIDEOS
1) https://youtu.be/YgqC8_WOZyY
2) https://youtu.be/Hw2XkWIv7Ww

3) https://youtu.be/N71pP5XHez4
4) https://youtu.be/lpKY85HCXmA
WIKIPEDIA
(Wikipedia-
https://en.m.wikipedia.org/wiki/Minimax#:~:text=Frequently%2C%20in%20game
%20theory%2C%20maximin,minimizing%20the%20opponent's%20maximum%2
0payoff.&text=%22Maximin%22%20is%20a%20term%20commonly,maximizes%
20one's%20own%20minimum%20payoff.)
https://getmyuni.azureedge.net › ...PDF
GAME THEORY INTRODUCTION
OER

UNIT 02-03: DOMINANCE IN GAMES
LEARNING OBJECTIVES
 Understand the use of dominance principle in reducing the games with pure and
mixed strategies
 Understand the arithmetic and algebraic method in solving 2 ×2 game with mixed
strategy
 Understand the generalized formula to obtain the optimal mixed strategies and
game value for 2×2 two-person zero-sum game without any saddle point
INTRODUCTION
In game theory, strategic dominance (commonly called dominance) occurs when
one strategy is better than another strategy for one player, no matter how that player’s
opponents may play. Many simple games can be solved by using dominance.
Dominance method is applicable to both pure and mixed strategy problems. In

case of pure strategy, the solution is obtained by itself whereas in case of mixed
strategy, it can be used to simplify the problem.
03-01: PRINCIPLE OF DOMINANCE

The principle of dominance states that “If the strategy of a player dominates
over the other strategy in all condition, then the later strategy can be ignored” because
it will not affect the solution in any way.
The strategy dominates over the other only if it is preferable over other in all
conditions. The concept of dominance is especially useful for the evaluation of two -
person zero-sum games where a saddle point does not exist.
THE DOMINANCE RULE FOR COLUMNS:
“Every value in the dominating column(s) must be less than or equal to the
corresponding value of the dominated column”.
More precisely, for columns X and Y, if X ≥ Y i. e. if all the elements of a particular

column X are greater than equal to the corresponding elements of another column Y,
then delete column X (in this situation, we say that column X is dominated by column
Y). The elements of the column X can also be compared with an average of two or more
columns and if the elements of column X are greater than the corresponding elements
after taking the average, then delete column X.

THE DOMINANCE RULE FOR ROWS:
“Every value in the dominating row(s) must be greater than or equal to the
corresponding value of the dominated row”.
More precisely, for rows X and Y, if X ≤ Y i. e. if all the elements of a particular row X
are less than or equal to the corresponding elements of another row Y, then delete row
X (in this situation, we say that row X is dominated by row Y). The elements of a
particular X can also be compared with an average of two or more other rows and if the
elements of the row X are less than or equal to the corresponding elements after taking
the average, then delete row X.
SOLVED PROBLEMS 01
Problem 02-03-01:Two players P and Q play a game. Each of them has to choose one
of the three colors, white (W), black (B) and red (R) independently of the other.
Thereafter, the colors are compared. If both P and Q have chosen white (W, W), neither
wins anything. If player P selects white and player Q black (W, B), player P loses Rs. 2
or player Q wins the same amount and so on. The complete pay-off table is shown
below. Find the optimum strategies for P and Q and the value of the game.
Colors chosen by Q
colours W B R
W 0 -2 7
Colors B 2 5 6
chosen by P R 3 -3 8
Solution: Consider the pay-off matrix and let us check the game for saddle point.
Players Q
colours W B R Row Min
W 0 -2 7 -2
B 2 5 6 2
R 3 -3 8 -3
Column 3 5 8
P
Max
Here, Maximin= 2 and Minimax= 3
⇒ Maximin ≠ Minimax

⇒ The game is without saddle point.
Observing strategies for player Q, we can definitely say that player Q will not play
strategy R since this will result in heaviest losses to him and highest gains to player P.
He can do better by playing columns W or B. Thus, column R is said to be deleted and
strategy R is called the dominated strategy.
Now, we have the resultant matrix as;
Players Q
colours W B
W 0 -2
B 2 5
P R 3 -3
Now, from above table, it is clear that player P will not play row W since it will give
him returns lower than given by row B. Hence, row W is dominated by row B and can
be deleted.
Players Q
colours W B
B 2 5
P
R 3 -3
This is reduced matrix by dominance principle which can be easily solved as discussed
later.
Remark:
Dominance need not be based on the superiority of pure strategies only. A given
strategy can be dominated if it is inferior to an average of two or more other pure
strategies.
The above remark is illustrated in the following example.
Problem 02-03-02: Reduce the following game by dominance.
Players B
Strategies 1 2 3
1 6 1 3
2 0 9 7
A
3 2 3 4

Solution: Initially, we check the game for saddle point.
Players B
Strategies 1 2 3 Row
min
1 6 1 3 1
2 0 9 7 0
A
3 2 3 4 2
Column 6 9 7
max
Here, Maximin= 2 and Minimax= 6
⇒ The game does not have saddle point.
Now, we can observe, none of the pure strategies of A is inferior to any of his other
pure strategies. However, average of A’s first and second pure strategies gives us
6+0 1+9 3+7

( , , ) = (3, 5, 5)
2 2 2
Clearly, this is superior to A’s 3 rd pure strategy (as every element in 3 rd strategy of A is
less than or equal to corresponding element in (3,5,5)).
Therefore, the third strategy may be deleted from matrix. Thus, the resultant matrix
becomes,
Strategies 1 2 3
1 6 1 3
A 2 0 9 7
Note: A game reduced by dominance may disclose a saddle point which was not found
in original matrix. This is not necessarily a true saddle point. Since, it may not be the
least value in its row and the highest value in its column as per the original matrix.
Thus, it is known as pseudo-saddle point and is ignored.
SELF-TEST 01
MCQ03-01:According to the dominance rule for columns, if X ≥ Y i. e. if all elements
of column X are greater than or equal to the corresponding elements of another column
Y, then delete
a) Column X

b) Column Y
c) Average of X and Y
d) Both X and Y
MCQ 03-02:According to the dominance rule for rows, if X ≤ Y i. e. if all elements of

row X are less than or equal to the corresponding elements of another row Y, then
delete
a) RowY
b) Row X
c) Average of X and Y
d) Both X and Y
MCQ 03-03:Dominance method is used to reduce the pay -off matrix of a game with
a) Pure strategies
b) Mixed strategies
c) Both
d) None of these

SAQ 02-03-01: Write a note on dominance method to solve games.
Solution: The principle of dominance states that “If the strategy of a player dominates
over the other strategy in all condition, then the later strategy can be ignored” because
it will not affect the solution in any way. Dominance method is applicable to both pure
and mixed strategy problems. In case of pure strategy, the solution is obtained by itself
whereas in case of mixed strategy, it can be used to simplify the problem.The concept
of dominance is especially useful for the evaluation of two -person zero-sum games
where a saddle point does not exist. The dominance rules for columns and rows is given
by:
THE DOMINANCE RULE FOR COLUMNS:
“Every value in the dominating column(s) must be less than or equal to the
corresponding value of the dominated column”.
The elements of the column X can also be compared with an average of two or more
columns and if the elements of column X are greater than the corresponding elements
after taking the average, then delete column X.

THE DOMINANCE RULE FOR ROWS:
“Every value in the dominating row(s) must be greater than or equal to the
corresponding value of the dominated row”.
The elements of a particular X can also be compared with an average of two or more
other rows and if the elements of the row X are less than or equal to the corr esponding
elements after taking the average, then delete row X.
SAQ 02-03-02: Reduce the following game by using dominance method , where A is
row player and B is column player.
Strategies 𝐵 𝐵 𝐵 𝐵
𝐴 3 5 4 2
𝐴 5 6 2 4
𝐴 2 1 4 0
𝐴 3 3 5 2
Solution:
As Row 3 ≤ Row 4, so remove Row 3.
Strategies 𝐵 𝐵 𝐵 𝐵
𝐴 3 5 4 2
𝐴 5 6 2 4
𝐴 3 3 5 2
Now, column 2 ≥ column 4, so remove column 2.
𝐴 3 4 2
𝐴 5 2 4
𝐴 3 5 2
Column 1 ≥ column 3, so remove column 1.
Strategies 𝐵 𝐵
𝐴 4 2
𝐴 2 4
𝐴 5 2
Row 1 ≤ row 3, so remove row 1.
𝐴 2 4
𝐴 5 2
This is the reduced game.

03-02: MIXED STRATEGIES
In the games where there is no saddle point and dominance has been used to
reduce the game matrix, players will resort to mixed strategies. A few different methods
will be described to optimize the winning of each player and to solve the game. One of
the players must determine what proportion of time to play each row while the other
must know what portion of the time to play each column. The pay-offs obtained will be
the expected pay-offs and the value of the game will be the expected value of the game.
Fundamental theorem of game theory:
“Every matrix game can be solved in terms of mixed strategies, i. e. if mixed strategies
are adopted, there always exists a value of the game.”
SOLUTION OF RECTANGULAR GAMES IN TERMS OF MIXED STRATEGIES:
If a game does not involve a saddle point, then the best strategies are mixed
strategies. Thus, the players cannot use their pure maximin and minimax strategies as
the optimal strategies. They both play their plays according to a predetermined set
which consists of probabilities corresponding to each of their pure strategies.
Consider a rectangular game played by players A and B, A being maximizing

player, with pay-off matrix P=[𝑎 ] × . Obviously, A has m pure strategies and B has n
pure strategies. Suppose, A’s mixed strategy is X= [𝑥 , 𝑥 , … , 𝑥 , … , 𝑥 ] and B’s mixed
strategy is Y= [ 𝑦 , 𝑦 , … , 𝑦 , … , 𝑦 ]. Therefore, A and B choose their ith and jth pure
strategies with probabilities 𝑥 and 𝑦 respectively. Then, the mathematical expectation,
expected gain to A (pay-off function for A) in the game is given by
𝐸(𝑋, 𝑌) = 𝑋 𝑃𝑌 = ∑ ∑ 𝑥 𝑎 𝑦
Now, (max min) and (min max) criterion are applied to this expected gain. If we denote
𝑚𝑎𝑥𝑚𝑖𝑛 𝑚𝑖𝑛𝑚𝑎𝑥
𝐸(𝑋, 𝑌) for player A and 𝐸(𝑋, 𝑌) for player B by 𝑣 and 𝑣 respectively,
𝑋 𝑌 𝑌 𝑋
then
𝑚𝑎𝑥 𝑚𝑖𝑛 𝑚𝑎𝑥 𝑚𝑖𝑛

𝑣= 𝐸(𝑋, 𝑌) = [ (∑ 𝑥 𝑎 )]
𝑋 𝑌 𝑋 𝑗
𝑚𝑎𝑥
= [min{∑ 𝑥 𝑎 , ∑ 𝑥 𝑎 , … , ∑ 𝑥 𝑎 }]
𝑋
and

𝑚𝑖𝑛𝑚𝑎𝑥 𝑚𝑖𝑛 𝑚𝑎𝑥
𝑣= 𝐸(𝑋, 𝑌) = [ (∑ 𝑎 𝑦 )]
𝑌 𝑋 𝑌 𝑖
𝑚𝑖𝑛
= [max{∑ 𝑎 𝑦 , ∑ 𝑎 𝑦 , … , ∑ 𝑎 𝑦 }]
𝑌
It can be shown that as in case of pure strategies,
𝑣 ≤ 𝑣 or 𝐸(𝑋, 𝑌) ≤ 𝐸(𝑋, 𝑌)
𝑋 𝑌 𝑌 𝑋
Thus, according to maximin criterion, for A the best strategy is that which maximizes
𝑚𝑖𝑛 𝑚𝑎𝑥
the ∑ 𝑥 𝑎 and for B best strategy is that which minimizes the 𝑗 ∑ 𝑦 𝑎 . A
𝑖
𝑚𝑎𝑥 𝑚𝑖𝑛 𝑚𝑖𝑛𝑚𝑎𝑥
pair of strategies (𝑋 , 𝑌 ) for which 𝑣 = 𝑣 , i. e. 𝐸(𝑋, 𝑌) = 𝐸(𝑋, 𝑌) =
𝑋 𝑌 𝑋 𝑌
𝐸(𝑋 , 𝑌 ) is called the Strategic saddle point of expectation function 𝐸(𝑋, 𝑌) or game and
(𝑋 , 𝑌 ) form the equilibrium pair of optimal strategies and 𝐸(𝑋 , 𝑌 ) gives the value of
the game. Fundamental theorem assures us that there always exist optimum strategies
such that 𝑣 = 𝑣 i. e. there always exists a strategic saddle point solution. Optimum
strategies say 𝑋 and 𝑌 have the following remarkable properties:
1) If one of the players adheres to his optimal mixed strategy and the other player
deviates from his optimal strategy, then the deviating player can only decrease
his yield and cannot increase in any case (at most may be equal).
2) If one of the players adheres to his optimal strategy, then the value of the game
does not alter if the opponent uses his supporting strategies only either singly or
in any mixture.
To show this, suppose 𝑣 is the value of the game and 𝑣 , 𝑣 , … , 𝑣 are the gains to A
when A uses his optimal mixed strategy and B uses his p ure supporting 1 st ,2 nd , …, nth
strategy respectively. By the above first property as B deviates from his optimal
strategy, B will lose more than 𝑣. Therefore, we have
𝑣 ≥ 𝑣, 𝑣 ≥ 𝑣, … , 𝑣 ≥ 𝑣
But we can show that

𝑣 ≯ 𝑣, 𝑣 ≯ 𝑣, … , 𝑣 ≯ 𝑣
Let, B’s optimal strategy be 𝑌 = [𝑦 , 𝑦 , … , 𝑦 ].
Note that as B uses his pure strategies with probabilities 𝑦 , 𝑦 , … , 𝑦 , we get
𝑣 = 𝑣 𝑦 + 𝑣 𝑦 + ⋯ + 𝑣 𝑦 , (𝑦 + 𝑦 + ⋯ + 𝑦 = 1, 𝑦 ≥ 0)

Obviously if any of 𝑣 ′𝑠 is greater than 𝑣, then the expression on the right-hand side is
also greater than 𝑣 which is contradiction. Therefore, neither of 𝑣 ′𝑠 is greater than 𝑣
and hence necessarily each 𝑣 is equal to 𝑣.
Also, note that since corresponding to each supporting strategy ga me value is 𝑣,

if B uses his supporting strategies in accordance with some other probability
distribution say 𝑦 ′, 𝑦 ′, … , 𝑦 ′, where ∑ 𝑦 ′ = 1, 𝑦 ′ ≥ 0 , then expected yield is given
by
= 𝑣𝑦 + 𝑣𝑦 + ⋯ + 𝑣𝑦 ′
= 𝑣(𝑦 + 𝑦 + ⋯ + 𝑦 ′) = 𝑣 ⋅ 1 = 𝑣
It shows that even if B uses his supporting strategies in any mixture, the value of the
game remains unchanged.
Similarly, we can show that if B adheres to his optima strategy.
Furthermore, a game is said to be symmetric if pay-off matrix remains unchanged on

interchanging the roles of the two players. Symmetric games have skew symmetric
matrices i. e. square matrix in which 𝑎 = −𝑎 and 𝑎 = 0 (diagonal elements are
zero).
It can be shown that a symmetric game has the value zero and the two optimal
strategies are identical.
For, we know that the pay-off function to A is
𝐸(𝑋, 𝑌) = ∑ ∑ 𝑥 𝑎 𝑦
It can be shown easily that in case of a skew-symmetric pay-off matrix if 𝑋 = 𝑌 then

𝐸(𝑋, 𝑌) = 0, i. e. if both the players choose the same mixed strategy, then expected pay-
off is zero. Suppose 𝑋 and 𝑌 are the optimal strategies of A and B respectively and 𝑣
is the value of the game i. e.
𝐸(𝑋 , 𝑌 ) = 𝑣
𝑚𝑎𝑥𝑚𝑖𝑛 𝑚𝑖𝑛
Now we know that, 𝑣 = 𝐸(𝑋, 𝑌) = 𝐸(𝑋 , 𝑌)
𝑋 𝑌 𝑌
Therefore, 𝐸(𝑋 , 𝑌) ≥ 𝑣.
Putting 𝑌 = 𝑋 , we get, 𝐸(𝑋 , 𝑋 ) ≥ 𝑣.
But 𝐸(𝑋 , 𝑋 ) = 0, as the game is symmetric,
Therefore, 0 ≥ 𝑣 𝑜𝑟 𝑣 ≤ 0 … (1)

𝑚𝑖𝑛𝑚𝑎𝑥 𝑚𝑎𝑥
Again, 𝑣 = 𝐸(𝑋, 𝑌) = 𝐸(𝑋, 𝑌 )
𝑌 𝑋 𝑋
Therefore, 𝐸(𝑋, 𝑌 ) ≤ 𝑣.
Putting 𝑋 = 𝑌 , we get 𝐸(𝑌 , 𝑌 ) ≤ 𝑣.
But 𝐸(𝑌 , 𝑌 ) = 0, therefore 0 ≤ 𝑣 … (2)
From (1) and (2), we get that 𝑣 = 0 and hence 𝑋 = 𝑌 .
Effect of addition (or subtraction) of a fixed number:
If we add (subtract) a fixed number 𝑙 to (from) each element of the pay-off

matrix then the optimal strategies remain unchanged while the value of the game
increases (or decreases) by 𝑙.
Note that for a pay-off matrix, element 𝑎 becomes 𝑎 + 𝑙. If we denote the

original pay-off function by 𝐸(𝑋, 𝑌) and the new pay-off function by 𝐸 (𝑋, 𝑌), then
𝐸(𝑋, 𝑌) = ∑ ∑ 𝑥 𝑎 𝑦
𝐸 (𝑋, 𝑌) = ∑ ∑ 𝑥 (𝑎 + 𝑙)𝑦
= 𝐸(𝑋, 𝑌) + 𝑙 (as∑ 𝑥 = 1, ∑ 𝑦 = 1)
Obviously, operation of adding 𝑙 does not alter the nature of 𝐸 (𝑋, 𝑌) and the
new pay-off function can be obtained from the old pay -off function by simply adding
the constant 𝑙 to it. Therefore, the optimal strategies are same for both the games and if
they are 𝑋 , 𝑌 , then by the above relation, we get
𝐸 (𝑋 , 𝑌 ) = 𝐸(𝑋 , 𝑌 ) + 𝑙
i.e., value of the new game= 𝑣 + 𝑙.
Similarly, the result can be proved for subtraction of 𝑙.
The game can be interpreted as the original one if before starting the game B pays to A
in advance an amount of 𝑙 units. Similarly, it can be shown that on multiplying the pay-
off matrix by a constant 𝑙, the optimal strategies remain unaltered while the value of the
game becomes 𝑙 times the value of the original game.
MIXED STRATEGIES (2×2 GAMES):
There are two methods used for finding optimum strate gies as well as game
value for a 2×2 game. These methods are:

1) Arithmetic method
2) Algebraic method
We will discuss these methods in detail in next two units.
SOLVED PROBLEMS 02
Problem 02-03-03: If all the elements of the pay-off matrix of a game are non-negative
and every column of this matrix has at least one positive element. Show that the value
of the corresponding game is positive.
Solution: Since each column has at least one positive element, the column maximum of
each column will be positive. Hence,
𝑚𝑖𝑛𝑚𝑎𝑥
𝑣= (𝑎 )= some non-zero number= e (say)
𝑗 𝑖
Further if each row contains at least one zero element, then the row minimums
will be zero for each row and hence
𝑚𝑎𝑥 𝑚𝑖𝑛
𝑣= (𝑎 ) = 0
𝑖 𝑗
In this case since 𝑣 < 𝑣, the game has no saddle point. Hence
𝑣 < 𝑣 < 𝑣 or 0 < 𝑣 < 𝑒.
It gives that 𝑣 (value of the game) is positive.
Again, if any one of the rows contains all the elements positive (no
zero), then in this row minimum will also be positive. Hence
𝑚𝑎𝑥 𝑚𝑖𝑛
𝑣= (𝑎 )= some positive number = k (say)
𝑖 𝑗
Then, 𝑣 ≤ 𝑣 ≤ 𝑣 or 𝑘 ≤ 𝑣 ≤ 𝑒.
In this case also ‘𝑣’ is positive.
2 −1
Problem 02-03-04: Consider a game with pay-off matrix [ ]. If value of the
−1 0
game is v=-1/4, then what will be the effect on the value of game if we i) add 2 to pay-
off matrix ii) multiply 2 to pay-off matrix?
Solution:
i) We know that, if 𝑣 is the value of game and we add some positive constant 𝑙 to
pay-off matrix, then value of the new game will be 𝑣 + 𝑙.
Thus, value of game= (− ) + 2= .

ii) We know that, if 𝑣 is the value of game and we multiply pay-off matrix by some
constant 𝑙 to, then value of the new game will be 𝑣𝑙.
Thus, value of game= (− ) × 2=− .
SELF-TEST 02
MCQ 03-04: If the pay-off matrix remains unchanged on interchanging the roles of two
players, then the game is said to be
a) Anti-symmetric
b) Symmetric
c) Two-person zero sum
d) Constant
MCQ 03-05: Symmetric games have _______ matrices.
a) Skew-symmetric
b) Symmetric
c) Scalar
d) Diagonal

SAQ 02-03-03: Write a note on: ‘Effect of addition (subtraction) of a fixed number to
(from) each element of pay-off matrix’.
Solution: If we add (subtract) a fixed number 𝑙 to (from) each element of the pay-off
matrix then the optimal strategies remain unchanged while the value of the game
increases (or decreases) by 𝑙.Note that for a pay-off matrix, element 𝑎 becomes 𝑎 + 𝑙.
If we denote the original pay-off function by 𝐸(𝑋, 𝑌) and the new pay-off function by
𝐸 (𝑋, 𝑌), then
𝐸(𝑋, 𝑌) = ∑ ∑ 𝑥 𝑎 𝑦
𝐸 (𝑋, 𝑌) = ∑ ∑ 𝑥 (𝑎 + 𝑙)𝑦
= 𝐸(𝑋, 𝑌) + 𝑙 (as ∑ 𝑥 = 1, ∑ 𝑦 = 1)
Obviously, operation of adding 𝑙 does not alter the nature of 𝐸 (𝑋, 𝑌) and the
new pay-off function can be obtained from the old pay -off function by simply adding

the constant 𝑙 to it. Therefore, the optimal strategies are same for both the games and if
they are 𝑋 , 𝑌 , then by the above relation, we get
𝐸 (𝑋 , 𝑌 ) = 𝐸(𝑋 , 𝑌 ) + 𝑙
i.e., value of the new game= 𝑣 + 𝑙.
Similarly, the result can be proved for subtraction of 𝑙.
SAQ 02-03-04: Define the terms: Symmetric game, Strategic saddle point of
expectation function, Also, state ‘Fundamental theorem of game theory’.
Solution: Symmetric game: A game is said to be symmetric if pay-off matrix remains

unchanged on interchanging the roles of the two players.
Strategic Saddle point of expectation function: A pair of strategies (𝑋 , 𝑌 ) for which

𝑣 = 𝑣 , i. e. 𝐸(𝑋, 𝑌) = 𝐸(𝑋, 𝑌) = 𝐸(𝑋 , 𝑌 ) is called the Strategic saddle
𝑋 𝑌 𝑋 𝑌
point of expectation function 𝐸(𝑋, 𝑌).
Fundamental theorem of game theory:
“Every matrix game can be solved in terms of mixed strategies, i. e. if mixed strategies
are adopted, there always exists a value of the game.”
03-03: ARITHMETIC METHOD

Arithmetic method provides an easy way for finding the optimum strategies for
each player in 2×2 game. It consists of following steps:
a) Subtract the two digits in column 1 and write them under column 2, ignoring
sign.
b) Subtract the two digits in column 2 and write them under column 1, ignoring
sign.
c) Similarly, proceed for the two rows.
These values are called oddments. They are the frequencies with which the
players must use their courses of action in their optimum strategies.
SOLVED PROBLEMS 03
Problem 02-03-05(Two-person zero-sum game without saddle point): In a game of
matching coins, player A wins Rs. 2 if there are two heads, wins nothing if there are
two tails and loses Rs. 1 when there are one head and one tail. Determine the pay -off
matrix, best strategies for each player and the value of game to A.
Solution: From the given information, the pay-off matrix for A will be,

Player B
Strategies H T
H 2 -1
A
T -1 0
Since, the game has no saddle point, the optimal strategies will be mixed strategies.
Using the steps described above, we get
Player B
Strategies H T oddments
H 2 -1 1 1 1
=
A 3+1 4
= 0.25
T -1 0 3 3 3
=
3+1 4
= 0.75
oddments 1 3
1 1 3 3
= =
3+1 4 3+1 4
= 0.25 = 0.75
Thus, for optimum gains,
Player A should use strategy H for 25% of the time and strategy T for 75% of the time;
while player B should use strategy H 25% of the time and strategy T 75% of the time.
To obtain the value of the game, any one of the following expressions may be used:
Using A’s oddments:
B plays H; value of the game,
× ×
V= Rs. [ ]= Rs.( )
B plays T; value of the game,
×( ) ×
V= Rs. [ ]= Rs.( )
Using B’s oddments:
A plays H; value of the game,
× ×
V= Rs. [ ]= Rs.( )
A plays T; value of the game,
( )× ×
V= Rs. [ ]= Rs.( )
The above values of V are equal only if the sum of the oddments vertically and
horizontally is equal. Thus, the full solution of the game is given by,

A(1, 3), B(1, 3) and value of game(V)= Rs.( )i. e. A gains Rs.( ) i. e. he loses
Rs.( ) which B in turn gets.
Note: Arithmetic method is easier than the algebraic method but it cannot be applied to
larger games.
Problem 02-03-06: Reduce the game by dominance and find the game value.
Strategies I II III IV
I 3 2 4 0
II 3 4 2 4
III 4 2 4 0
IV 0 4 0 8
Solution: Consider the game,
Strategies I II III IV Row

Min
I 3 2 4 0 0
II 3 4 2 4 2
III 4 2 4 0 0
IV 0 4 0 8 0
Column 4 4 4 8
Max
Here, Maximin = 2 & Minimax = 4
⇒ The game has no saddle point. Let us try to reduce it by dominance,
Let us denote row player by ‘A’ and column player by ‘B’. Then, from player A’s point
of view, row I is dominated by row III. Thus, by using dominance rule for rows, row I
is to be deleted resulting into the following reduced matrix :
Player B
Player Strategies I II III IV
A II 3 4 2 4
III 4 2 4 0
IV 0 4 0 8
In this matrix, none of the row dominates other. So, we observe for columns i. e. from
player B’s point of view. Thus, by using dominance rule for columns, column I is
dominated by column III. So, column I is to be deleted from the pay-off matrix which
results in the following reduced matrix:

Player B
Player Strategies II III IV
A
II 4 2 4
III 2 4 0
IV 4 0 8
In this matrix, no single row (or column) dominates another row (or column). However,
column II is dominated by the average of columns III and IV, which is
3
= [2]
4
[ ]
Hence, column II is deleted, which gives the reduced matrix:
Strategies III IV
II 2 4
III 4 0
IV 0 8
Again, row II is dominated by the average of III and IV rows, which is
( , ) = (2, 4)
Therefore, row II is deleted, resulting into 2×2 matrix,
Strategies III IV
III 4 0
IV 0 8
The above 2×2 matrix has no saddle point. We will solve it by arithmetic method,
Player B
Strategies III IV oddments
III 4 0 8 =
IV 0 8 4 =
Player A
oddments 8 4
= =

Thus, the complete solution to the given problem is,
Optimal strategy for player A: (0,0, , )
Optimal strategy for player B: (0,0, , )
× ×
Value of game (for A): = .
SELF-TEST 03
MCQ 03-06:________ are the frequencies with which the players must use their
courses of action in their optimum strategies.
a) Payoffs
b) Oddments
c) Strategies
d) Gains
MCQ 03-07: In a 4×4 two-person zero-sum game, the pay-off matrix obtained after
2 3
reduction by using principle of dominance is [ ] . Then, optimal strategy for
4 −5
column player is
a) (0, , , 0)
b) (0,0, , )
c) (0, , , 0)
d) ( , , 0,0)

SAQ 02-03-05: Solve the following game by using the principle of dominance :
Strategy I II III IV V VI
1 4 2 0 2 1 1
2 4 3 1 3 2 2
3 4 3 7 -5 1 2
4 4 3 4 -1 2 2
5 4 3 3 -2 2 2

Solution: The game has no saddle point as maximin=1 ≠ minimax=2.
Row 1 is dominated by row 2 and row 5 is dominated by row 4. So, we delete row 1 and
row 5.
Strategy I II III IV V VI
2 4 3 1 3 2 2
3 4 3 7 -5 1 2
4 4 3 4 -1 2 2
Columns I and II are dominated by columns IV, V and VI, also column VI is dominated
by column V. Therefore, columns I, II and VI will be deleted.
Strategy III IV V
2 1 3 2
3 7 -5 1
4 4 -1 2
Now, we can observe that none of row (or column) dominates another row (or column).
However, column V is dominated by average of columns III and IV, which is
(1 + 3)/2 2
[(7 − 5)/2] = [ 1 ]
(4 − 1)/2 3/2
So, column V will be deleted and we get the following reduced matrix :
Strategy III IV
2 1 3
3 7 -5
4 4 -1
Row 4 is dominated by average of row 2 and 3. Hence, row 4 will be deleted. The
resulting game is:
Strategy III IV

2 1 3
3 7 -5
This game is without saddle point. Solving this game by arithmetic method, we get
Optimal strategy for A: (0, 6/7, 1/7, 0, 0);
Optimal strategy for B: (0, 0, 4/7, 3/7, 0);
Game value= 13/7.
SAQ 02-03-06: Write a short note on arithmetic method for finding optimum strategies
and game value.
Solution: Arithmetic method provides an easy way for finding the optimum strategies
for each player in 2×2 game. It consists of following steps:
a) Subtract the two digits in column 1 and write them under column 2, ignoring
sign.
b) Subtract the two digits in column 2 and write them under column 1, ignoring
sign.
c) Similarly, proceed for the two rows.
These values are called oddments. They are the frequencies with which the
players must use their courses of action in their optimum strategies. (Give one
example to show how it works).
03-04: ALGEBRAIC METHOD

While applying this method it is assumed that x represents the fraction of time
(frequency) for which player A uses strategy 1 and (1 -x) represents the fraction of time
(frequency) for which he uses strategy 2. Similarly, y and (1-y) represents the fraction
of time for which player B uses strategies 1 and 2 respectively.
SOLVED PROBLEMS 04
Problem 02-03-07(Two-person zero-sum game without saddle point): The two
armies are at war. Army A has two air-bases, one of which is thrice as valuable as the
other. Army B can destroy an undefended air-base, but it can destroy only one of them.
Army A can also defend only one of them. Find the best strategy for A to minimize its
losses.
Solution: Since both armies have only two possible courses of action, the gain matrix
for army A is,

Army B
1(Attack 2(Attack
smaller larger
airbase) airbase)
1(Defend 0 -3 →x
Army smaller Both The
A airbase) survive larger
one
destroyed
2(Defend -1 0 → 1-
larger The Both x
airbase) smaller survive
one
destroyed
↓ ↓
y 1-y
There is no saddle point.
Under this method, Army A wants to divide its plays between the two rows so that the
expected winnings by playing the first row are exactly equal to the expected winnings
by playing the second row irrespective of what army B does.
In order to arrive at the optimum strategies for army A, it is necessary to equate

its expected winnings when army B plays column 1 to its expected winnings when army
B plays column 2 i. e. when
0 ⋅ 𝑥 + (−1)(1 − 𝑥) = (−3) ⋅ 𝑥 + 0 ⋅ (1 − 𝑥)
⇒ −1 + 𝑥 = −3𝑥
⇒ 4𝑥 = 1
1
⇒𝑥=
4
th th
Thus, army A should play first row of the time and second row (=1-x) of the time.
Similarly, army B wants to divide its time between columns 1 and 2 so that the expected
winnings are same by playing each column, no matter what army A does. Optimum
strategies for army B will be found by equating its expected winnings when army A
plays row 1 to its expected winnings when army A plays row 2.
i. e. when
0 ⋅ 𝑦 + (−3)(1 − 𝑦) = (−1) ⋅ 𝑦 + 0 ⋅ (1 − 𝑦)
⇒ −3 + 3𝑦 = −𝑦
⇒ 4𝑦 = 3

3
⇒𝑦=
4
th th
Thus, army B should play first column of the time and second column
(=1-y) of the time. With these optimum strategies gain-matrix becomes,
Army B
1 2
1 0 -3 (x)
2 -1 0 (1-x)
Army A (y) (1-y)
The game value can be found either for army A or for army B.
Game value for army A:
th
While army B plays column 1, of time, army A wins zero for time and -1 for
th
time; also, while army B plays column 2, of time, army A wins -3 for time and
zero for time.
Thus, total expected winnings for army A are
Game value = [0 × − 1 × ] + [−3 × + 0 × ]
=− − =− .
Game value for army B:
While army A plays row 1, of time, army B wins zero for of time and -3 for of
time; also, while army A plays row 2, of time, army wins -1 for of time and zero for
of time.
Thus, game value for army B= [ × 0 − 3 × ] + [ × −1 + 0 × ]=− .
Thus, the full solution of the game is;
Army A: ( , )
Army B: ( , )
Game Value: =− .
Problem 02-03-08:For any 2×2 two-person zero-sum game without any saddle point,
having pay-off matrix for player A as

Player B
Player A 𝐴 𝑎 𝑎
𝐴 𝑎 𝑎
Find the optimal mixed strategies and value of the game.
Solution:
For player A,
Let, x→ be the probability of selecting strategy 𝐴
1-x → be the probability of selecting strategy 𝐴 .
For player B,
Let, y→ be the probability of selecting strategy 𝐵
1-y → be the probability of selecting strategy 𝐵 .
Then,
Player B
Strategies 𝐵 𝐵 Probability
𝐴 𝑎 𝑎 x
𝐴 𝑎 𝑎 1-x
Player A
Probability y 1-y
The expected value of the game to player A is given by,
𝐸(𝑥, 𝑦) = 𝑥𝑦𝑎 + (1 − 𝑥)𝑦𝑎 + 𝑥(1 − 𝑦)𝑎 + (1 − 𝑥)(1 − 𝑦)𝑎 … (1)
In order to determine the optimum values of x and y, we differentiate E(x,y)

partially w. r. t. x and y,
𝜕
[𝐸(𝑥, 𝑦)] = 𝑦𝑎 − 𝑦𝑎 + (1 − 𝑦)𝑎 − (1 − 𝑦)𝑎
𝜕𝑥
𝜕
[𝐸(𝑥, 𝑦)] = 𝑥𝑎 + (1 − 𝑥)𝑎 − 𝑥𝑎 − (1 − 𝑥)𝑎
𝜕𝑦
By setting, [𝐸(𝑥, 𝑦)] = 0 and [𝐸(𝑥, 𝑦)] = 0, we get
𝑦(𝑎 −𝑎 −𝑎 +𝑎 )= 𝑎 −𝑎
⇒ 𝑦[𝑎 +𝑎 − (𝑎 + 𝑎 )] = 𝑎 −𝑎

𝑎 −𝑎
⇒𝑦=
[𝑎 + 𝑎 − (𝑎 + 𝑎 )]
and
𝑥(𝑎 −𝑎 −𝑎 +𝑎 )=𝑎 −𝑎
𝑎 −𝑎
⇒𝑥=
[𝑎 + 𝑎 − (𝑎 + 𝑎 )]
Substituting these values of x, y in E(x, y), the required value of the game is
𝑎 𝑎 −𝑎 𝑎
𝑉 = 𝐸(𝑥, 𝑦) =
[(𝑎 + 𝑎 ) − (𝑎 + 𝑎 )]
SELF-TEST 04
0 2
MCQ 03-08: If a game has the pay-off matrix [ ]. Then, the value of game is
3 1
a) V=3
b) V=2
c) V=3/2
d) V=6/7
MCQ 03-09: While applying algebraic method, it is assumed that x represents fraction
of ______ for which row player uses strategy 1.
a) Time
b) Game
c) Strategy
d) Value

SAQ 02-03-07: Explain the algebraic method for finding optimum strategies and game
value.
Solution: While applying this method it is assumed that x represents the fraction of
time (frequency) for which player A uses strategy 1 and (1-x) represents the fraction of
time (frequency) for which he uses strategy 2. Similarly, y and (1 -y) represents the
fraction of time for which player B uses strategies 1 and 2 respectively. For any 2 ×2
two-person zero-sum game without any saddle point, having pay-off matrix for player A
as
Player B

Player 𝐴 𝑎 𝑎
A 𝐴 𝑎 𝑎
Let, x→ be the probability of selecting strategy 𝐴
1-x → be the probability of selecting strategy 𝐴 .
For player B,
Let, y→ be the probability of selecting strategy 𝐵
1-y → be the probability of selecting strategy 𝐵 .
Then,
Player B
Strategies 𝐵 𝐵 Probability
𝐴 𝑎 𝑎 x
𝐴 𝑎 𝑎 1-x
Probability y 1-y
Player
A
The expected value of the game to player A is given by,
𝐸(𝑥, 𝑦) = 𝑥𝑦𝑎 + (1 − 𝑥)𝑦𝑎 + 𝑥(1 − 𝑦)𝑎 + (1 − 𝑥)(1 − 𝑦)𝑎 … (1)
In order to determine the optimum values of x and y, we differentiate E(x,y)

partially w. r. t. x and y,
𝜕
[𝐸(𝑥, 𝑦)] = 𝑦𝑎 − 𝑦𝑎 + (1 − 𝑦)𝑎 − (1 − 𝑦)𝑎
𝜕𝑥
𝜕
[𝐸(𝑥, 𝑦)] = 𝑥𝑎 + (1 − 𝑥)𝑎 − 𝑥𝑎 − (1 − 𝑥)𝑎
𝜕𝑦
By setting, [𝐸(𝑥, 𝑦)] = 0 and [𝐸(𝑥, 𝑦)] = 0, we get
𝑦(𝑎 −𝑎 −𝑎 +𝑎 )= 𝑎 −𝑎
⇒ 𝑦[𝑎 +𝑎 − (𝑎 + 𝑎 )] = 𝑎 −𝑎
𝑎 −𝑎
⇒𝑦=
[𝑎 + 𝑎 − (𝑎 + 𝑎 )]
and
𝑥(𝑎 −𝑎 −𝑎 +𝑎 )=𝑎 −𝑎
𝑎 −𝑎
⇒𝑥=
[𝑎 + 𝑎 − (𝑎 + 𝑎 )]
Substituting these values of x, y in E (x, y), the required value of the game is

𝑉 = 𝐸(𝑥, 𝑦) =
[(𝑎 + 𝑎 ) − (𝑎 + 𝑎 )]
0 2
SAQ 02-03-08: A game has the pay-off matrix 𝐴 [ ]. Show that 𝐸(𝑥, 𝑦) = 1 − 2𝑥(𝑦 −
1 1
) and deduce that in the solution of the game, the first player follows a pure strategy
while the second has infinite number of mixed strategies.
Solution: Suppose, A uses his 1 st and 2 nd pure strategies with probabilities x and 1-x, B
uses mixed strategies with (y, 1-y). Then, the expected pay-off is given by
𝐸 (𝑥, 𝑦) = 0 ∙ 𝑥 ∙ 𝑦 + 2 ∙ 𝑥 ∙ (1 − 𝑦) + (1 − 𝑥) ∙ 𝑦 + 1 ∙ (1 − 𝑥) ∙ (1 − 𝑦)
= 1 + 𝑥 − 2𝑥𝑦 = 1 − 2𝑥(𝑦 − ).
Then, using the result, for any 2×2 two-person zero-sum game without any saddle point
𝑎 𝑎
with pay-off matrix as [𝑎 𝑎 ] having (x, 1-x) as probabilities of selecting strategies
1 and 2 for row player and (y, 1-y) as probabilities of selecting strategies 1 and 2 for
column player. We have,
𝑎 −𝑎
𝑥=
[𝑎 + 𝑎 − (𝑎 + 𝑎 )]
𝑎 −𝑎
𝑦=
[𝑎 + 𝑎 − (𝑎 + 𝑎 )]
𝑉 = 𝐸(𝑥, 𝑦) =
[(𝑎 + 𝑎 ) − (𝑎 + 𝑎 )]
Using these formulae, we have
𝑥 = 0, 𝑥 = 1 − 𝑥 = 1, 𝑦 = , 𝑦 = 1 − 𝑦 = 𝑎𝑛𝑑 𝑉 = 1.
Therefore, A’s optimum strategy is ( 𝑥 , 𝑥 )=(0,1) and B’s optimum mixed strategy is
(1/2,1/2). If A follows his optimum strategy, B can use his supporting strategies either
singly or in any proportion. Thus, A has only one optimum pure strategy, while B has
infinite number of optimum mixed strategies.
SUMMARY
In this unit, we studied the dominance method to reduce the given complex game into
simple one. We have also learned the arithmetic and algebraic method to solve 2 ×2 two-
person zero-sum games without saddle points (i. e. with mixed strategies).
1. Solve the following two-person zero-sum game using method of dominance.
Strategies 1 2 3 4 5
1 4 6 5 10 6
2 7 8 5 9 10
3 8 9 11 10 9

4 6 4 10 6 4
2. Reduce by using dominance principle and then find optimal strategies for both
the players. Also, find the value of game which is given by
Firms Firm B
Strategies B1 B2 B3 B4
A1 35 65 25 5
A2 30 20 15 0
A3 40 50 0 10
Firm A
A4 55 60 10 15
(Hint: Use arithmetic method to obtain the solution of reduced game.)
3. Write a note on methods used for finding optimum strategies and game value for
2×2 game with mixed strategies.
4. Solve the game whose pay-off matrix is given below by arithmetic method and
verify the results by algebraic method. Calculate the game value.
Players B
A 5 2
3 4
5. Use dominance property to solve the following game between two players A and
B:
Players B
A 6 8 6
4 12 2
6. Solve by using dominance property, the following game:
Players B
Strategies I II III
I 1 7 2
A
II 6 2 7
III 6 1 6
1 7 2
7. Solve the game whose pay-off matrix is given by [6 2 7].
5 1 6
8. Solve the following game by using principle of dominance :

A1 2 -2 -3 4
A2 4 1 2 2
KEY WORDS
Principle of dominance, dominance rule in games, mixed strategies, arithmetic method,
algebraic method.
REFERENCES
BOOKS
& COMPANY LTD

PRAGATI EDITION
MOOCS
YOUTUBE VIDEOS
1) https://youtu.be/YJvbxAvxkDc
2) https://youtu.be/h_GYxpuFVlw
3) https://youtu.be/fSuqTgnCVRg
4) https://youtu.be/oneW6SdUJPE
5) https://youtu.be/MPM_KgaXij0
WIKIPEDIA
 https://hithaldia.in › sas_facultyPDF
 https://www.geeksforgeeks.org › ...
 https://cbom.atozmath.com › CBOM
 https://getmyuni.azureedge.net/assets/main/study-
material/notes/mechanical_engineering_operation-research_game-
theory_notes.pdf
 http://people.maths.ox.ac.uk/griffit4/Math_Alive/3/game_theory_note s.pdf
OER

UNIT 02-04: MIXED STRATEGIES (2×N AND M×2 GAMES)
LEARNING OBJECTIVES
 Understand algebraic method to solve m×2 and 2×n games
 Understand graphic method for finding solution of m×2 and 2×n games
 Understand the method of subgames to solve m×2 and 2×n games
INTRODUCTION
The 2×n and m×2 games are the games in which one of the players has only two
strategies while his opponent may have any number of strategies . To solve such games,
initially we look for a saddle point; if it exists then the game is readily solved. If not,
then the next step is to reduce the given matrix to 2 ×2 size matrix by using the rules of
dominance. Once the matrix gets reduced to 2 ×2 size, we can solve it easily by using
arithmetic or algebraic methods which were discussed in previous unit 2.3. If, however,
it is not possible to reduce a matrix to 2×2 size matrix, it can be still solved by
algebraic method, method of subgames and graphic method. We will discuss each of
these methods in this unit in detail.
04-01: ALGEBRAIC METHOD FOR 2×N AND M×2 GAMES

The pay-offs for a rectangular game can always be given by m × n matrix, where
player A has m-possible courses of action and player B has n -possible courses of
action& the pay-off matrix is (𝑎 ). This is represented in tabular form as follows:
Player B
1 2 3 … 𝑗 … 𝑛
1 𝑎 𝑎 𝑎 … 𝑎 … 𝑎
2 𝑎 𝑎 𝑎 … 𝑎 … 𝑎
3 𝑎 𝑎 𝑎 … 𝑎 … 𝑎
Player A
⋮ ⋮ ⋮ ⋮ ⋱ ⋮ ⋱ ⋮
⋮ ⋮ ⋮ ⋮ ⋱ ⋮ ⋱ ⋮
[ 𝑚 𝑎 𝑎 𝑎 … 𝑎 … 𝑎 ]
It can be shown mathematically that
1. Each rectangular game has a specific value 𝑣. This value is unique.
2. For player A, there exists a best strategy X i. e. there exists frequencies

𝑥 ,𝑥 ,…,𝑥 such that X = (𝑥 , 𝑥 , … , 𝑥 ); 𝑥 + 𝑥 + ⋯ + 𝑥 = 1 i. e. ∑ 𝑥 = 1.

i.e. if he plays strategy 1 with frequency 𝑥 , strategy 2 with frequency 𝑥 ,…,
strategy m with frequency 𝑥 , then he can assure himself at least an expected
gain of 𝑣, where 𝑣 is the value of the game.
3. Similarly, for player B, there exists a best strategy, Y = ( 𝑦 , 𝑦 , … , 𝑦 ); ∑ 𝑦 =

1, such that if he plays strategies 1, 2, …, n with the frequencies 𝑦 , 𝑦 , … , 𝑦
respectively, he can assure himself at most a loss of 𝑣.
The unknowns (frequencies) 𝑥 , 𝑥 , … , 𝑥 ; 𝑦 , 𝑦 , … , 𝑦 and 𝑣 can be found from

the following relations:
𝑥 + 𝑥 + ⋯ + 𝑥 = 1, 𝑥 ≥ 0 … (i)
𝑦 + 𝑦 + ⋯ + 𝑦 = 1, 𝑦 ≥ 0 … (ii)
𝑥 𝑎 +𝑥 𝑎 + ⋯+𝑥 𝑎 ≥ 𝑣, 𝑓𝑜𝑟 𝑗 = 1, 2, … , 𝑛 … (iii)
𝑦 𝑎 + 𝑦 𝑎 + ⋯+𝑦 𝑎 ≤ 𝑣, 𝑓𝑜𝑟 𝑖 = 1, 2, … , 𝑚 … (iv)
Note that relation (iii) actually represents n-inequalities, one for each j.
On the similar lines, relation (iv) represents m-inequalities. Thus, we have
m+n+2 relations in m+n+1 unknowns (we added restrictions 𝑥 ≥ 0, 𝑦 ≥ 0 since
negative frequencies have no meaning).
The algebraic method is a direct method to solve the relations (i), (ii),
(iii) and (iv) for the unknowns. It must be kept in mind that each rectangular
game has a value viz. 𝑣, which exists and is unique. Therefore, the main idea is
to find such value 𝑣 which satisfies all the four relations. The first obvious step
is to assume that relations (iii) and (iv) are equalities.
Let us illustrate, how algebraic method can help to solve 2 ×n or m×2

games with the help of example.
SOLVED PROBLEMS 01
Problem 02-04-01:Solve the game for which the pay-off matrix is given by
Players B
Strategies 1 2
1 -2 -4
A 2 -1 3
3 1 2
Solution: From the above discussion, as A has 3 strategies and B has 2 strategies ,
let X= ( 𝑥 , 𝑥 , 𝑥 ) and Y= ( 𝑦 , 𝑦 ) be the best strategies for players A and B

respectively and V be the value of game.

Then,𝑥 , 𝑥 , 𝑥 ; 𝑦 , 𝑦 and V satisfies the relations:
𝑥 + 𝑥 + 𝑥 =1 …(i)
𝑦 + 𝑦 =1 …(ii)
𝑥 (−2) + 𝑥 (−1) + 𝑥 (1) ≥ 𝑉 …(iii)
𝑥 (−4) + 𝑥 (3) + 𝑥 (2) ≥ 𝑉 …(iv)
𝑦 (−2) + 𝑦 (−4) ≤ 𝑉 …(v)
𝑦 (−1) + 𝑦 (3) ≤ 𝑉 …(vi)
𝑦 (1) + 𝑦 (2) ≤ 𝑉 …(vii)
Hence, we have seven relations in six unknowns 𝑥 , 𝑥 , 𝑥 ; 𝑦 , 𝑦 and V.
Now, the first step towards solution is to assume all the inequalities as equalities and
we get
𝑥 + 𝑥 + 𝑥 =1 …(1)
𝑦 + 𝑦 =1 …(2)
𝑥 (−2) + 𝑥 (−1) + 𝑥 (1) = 𝑉 …(3)
𝑥 (−4) + 𝑥 (3) + 𝑥 (2) = 𝑉 …(4)
𝑦 (−2) + 𝑦 (−4) = 𝑉 …(5)
𝑦 (−1) + 𝑦 (3) = 𝑉 …(6)
𝑦 (1) + 𝑦 (2) = 𝑉 … (7)
Now, considering equations (2), (5) and (6), we have
Equation (2)⟹ 𝑦 = 1 − 𝑦 ,
With this substitution,
Equation (5)⟹ −2𝑦 − 4(1 − 𝑦 ) = 𝑉
⟹ −2𝑦 − 4 + 4𝑦 = 𝑉
⟹ 2𝑦 − 4 = 𝑉
Equation (6)⟹ −𝑦 + 3𝑦 = 𝑉
⟹ −𝑦 + 3(1 − 𝑦 ) = 𝑉
⟹ −4𝑦 + 3 = 𝑉
Therefore, we get, 2𝑦 − 4 = −4𝑦 + 3

7
⟹𝑦 =
6
Which is unacceptable, since frequency 𝑦 cannot be greater than 1 (0 ≤ 𝑦 ≤ 1).
Therefore, if we assume relations (v) and (vi) as equalities, we do not get a solution.
A general procedure is, then, to assume one inequality and the other relations as
equalities. If we again meet a contradiction, we assume other inequalities until a
solution is finally reached.
These computations can be made simpler with the help of the following theorems:
Theorem 1: states that; “If 𝑥 𝑎 +𝑥 𝑎 +⋯+ 𝑥 𝑎 > 𝑉, then 𝑦 = 0”.
Theorem 2: states that; “If 𝑦 𝑎 + 𝑦 𝑎 + ⋯ + 𝑦 𝑎 < 𝑉, then 𝑥 = 0”.
Applying these theorems to our problem, we proceed as follows:
(1) If −2𝑥 − 𝑥 + 𝑥 > 𝑉, then 𝑦 = 0.
Therefore, equations (5), (6) and (7) can be true only if 𝑦 = 0.
∴ 𝑦 + 𝑦 = 0, which is contradicted by equation (2).
(2) Similarly, a contradiction is obtained if −4𝑥 + 3𝑥 + 2𝑥 > 𝑉.
(3) If −2𝑦 − 4𝑦 < 𝑉, then 𝑥 = 0 and equations (1), (3) and (4) will lead to
contradiction. Proceeding in this manner, we finally find the only
following relations lead to a solution:
𝑥 +𝑥 +𝑥 = 1 … (1)
𝑦 +𝑦 =1 … (2)
−2𝑥 − 𝑥 + 𝑥 = 𝑉 … (3)
−4𝑥 + 3𝑥 + 2𝑥 > 𝑉… (4a)
−2𝑦 − 4𝑦 < 𝑉 … (5a)
−𝑦 + 3𝑦 < 𝑉 … (6a)
𝑦 + 2𝑦 = 𝑉 … (7)
From relation (4a), 𝑦 = 0, so that 𝑦 = 1.
Further, from relation (5a), 𝑥 = 0 and from (6a), 𝑥 = 0, so that 𝑥 = 1.
Substituting the values of 𝑦 and 𝑦 in (7), we get, V=1. Therefore, the final
solution is given by 𝑥 = 0, 𝑥 = 0, 𝑥 = 1, 𝑦 = 1, 𝑦 = 0 and V=1.

i.e. optimum strategy for A is A (0, 0, 1);optimum strategy for B is B(1, 0) and the
game value=1.
Remark: Evidently, the algebraic solution of the game is very lengthy and the
computations become much more complex as the number of possible choices for A and
B increases.
Problem 02-04-02: Solve the following game:
Players B
Strategies I II III
I -1 2 1
II 1 -2 2
III 3 4 -3
A
Solution: Suppose [ 𝑥 , 𝑥 , 𝑥 ] and [ 𝑦 , 𝑦 , 𝑦 ] are the optimal strategies of A and B

respectively and V is the value of the game. Then we are supposed to find the values of
𝑥 , 𝑥 , 𝑥 ; 𝑦 , 𝑦 , 𝑦 and V which satisfy the following relations:
(−1)𝑥 + 1𝑥 + 3 𝑥 ≥ 𝑉 … (1)
2𝑥 + (−2)𝑥 + 4 𝑥 ≥ 𝑉 … (2)
1𝑥 + 2𝑥 − 3 𝑥 ≥ 𝑉 … (3)
(−1)𝑦 + 2 𝑦 + 1𝑦 ≤ 𝑉 … (4)
1𝑦 + (−2) 𝑦 + 2𝑦 ≤ 𝑉 … (5)
3𝑦 + 4 𝑦 + (−3)𝑦 ≤ 𝑉 … (6)
𝑥 + 𝑥 + 𝑥 =1 … (7)
𝑦 + 𝑦 + 𝑦 =1 … (8)
𝑥 ,𝑥 ,𝑥 , 𝑦 ,𝑦 ,𝑦 ≥ 0 … (9)
First of all, we consider all the first six inequalities as equations. Then , we have the
following system of 8 equations in 7 unknowns:
−𝑥 + 𝑥 + 3 𝑥 = 𝑉 … (10)
2𝑥 − 2𝑥 + 4 𝑥 = 𝑉 … (11)
𝑥 + 2𝑥 − 3 𝑥 = 𝑉 … (12)
−𝑦 + 2 𝑦 + 𝑦 = 𝑉 … (13)
𝑦 − 2 𝑦 + 2𝑦 = 𝑉 … (14)

3𝑦 + 4 𝑦 − 3𝑦 = 𝑉 … (15)
𝑥 + 𝑥 + 𝑥 =1 … (16)
𝑦 + 𝑦 + 𝑦 =1 … (17)
Now we try to solve this system of equations:
Adding (10) and (12), we get 𝑥 = ( ) 𝑉 and again adding 2 times of (10) in (11), we get
𝑥 = ( ) 𝑉. Putting these values in any one of the equations (10), (11), (12), we get
𝑥 = ( ) 𝑉. Substituting all these in (16), we get 𝑉 = and hence
𝑥 = ( ), 𝑥 = ( ), 𝑥 = ( )
Now, substituting this value of V in (13), (14), (15) and solving them, we get
7 6 10
𝑦 = ( ),𝑦 = ( ),𝑦 = ( )
23 23 23
These values of 𝑦 , 𝑦 , 𝑦 also satisfy (17). Thus, we have obtained a solution of this
system of equations. Note that the solution also satisfies the non-negativity restrictions
given in (9).
Hence, the optimal strategies of A and B are [ , , ] and ( , , ) respectively and
the value of the game is V= .
SELF-TEST 01
MCQ 04-01: In rectangular game of size m × n, while solving it by algebraic method,
we have m+n+2 relations in ______ unknowns.
a) m + n
b) m+n+2
c) m+n+1
d) mn
MCQ 04-02: If X= (𝑥 , 𝑥 , … , 𝑥 ) is the best strategy for row player, where 𝑥 ,

𝑥 ,…,𝑥 are frequencies then ∑ 𝑥 =
a) 0
b) 1
c) m
d) m-1

SAQ 02-04-01: Write a short note on: “Algebraic method to solve rectangular game”.
Solution: The pay-offs for a rectangular game can always be given by m × n matrix,
where player A has m-possible courses of action and player B has n -possible courses of
action & the pay-off matrix is (𝑎 ). This is represented in tabular form as follows:
Player B
1 2 3 … 𝑗 … 𝑛
1 𝑎 𝑎 𝑎 … 𝑎 … 𝑎
2 𝑎 𝑎 𝑎 … 𝑎 … 𝑎
3 𝑎 𝑎 𝑎 … 𝑎 … 𝑎
Player A
⋮ ⋮ ⋮ ⋮ ⋱ ⋮ ⋱ ⋮
⋮ ⋮ ⋮ ⋮ ⋱ ⋮ ⋱ ⋮
[ 𝑚 𝑎 𝑎 𝑎 … 𝑎 … 𝑎 ]
It can be shown mathematically that,
1. Each rectangular game has a specific value 𝑣. This value is unique.
2. For player A, there exists a best strategy X i. e. there exists frequencies

𝑥 ,𝑥 ,…,𝑥 such that X = (𝑥 , 𝑥 , … , 𝑥 ); 𝑥 + 𝑥 + ⋯ + 𝑥 = 1 i. e. ∑ 𝑥 = 1.
i.e. if he plays strategy 1 with frequency 𝑥 , strategy 2 with frequency 𝑥 ,…,

strategy m with frequency 𝑥 , then he can assure himself at least an expected
gain of 𝑣, where 𝑣 is the value of the game.
3. Similarly, for player B, there exists a best strategy, Y = ( 𝑦 , 𝑦 , … , 𝑦 ); ∑ 𝑦 =

1, such that if he plays strategies 1, 2, …, n with the frequencies 𝑦 , 𝑦 , … , 𝑦
respectively, he can assure himself at most a loss of 𝑣.
The unknowns (frequencies) 𝑥 , 𝑥 , … , 𝑥 ; 𝑦 , 𝑦 , … , 𝑦 and 𝑣 can be found from the

following relations:
𝑥 + 𝑥 + ⋯ + 𝑥 = 1, 𝑥 ≥ 0 … (i)
𝑦 + 𝑦 + ⋯ + 𝑦 = 1, 𝑦 ≥ 0 … (ii)
𝑥 𝑎 +𝑥 𝑎 + ⋯+𝑥 𝑎 ≥ 𝑣, 𝑓𝑜𝑟 𝑗 = 1, 2, … , 𝑛 … (iii)
𝑦 𝑎 + 𝑦 𝑎 + ⋯+𝑦 𝑎 ≤ 𝑣, 𝑓𝑜𝑟 𝑖 = 1, 2, … , 𝑚 … (iv)
Note that relation (iii) actually represents n-inequalities, one for each j. On the
similar lines, relation (iv) represents m-inequalities. Thus, we have m+n+2 relations in

m+n+1 unknowns (we added restrictions 𝑥 ≥ 0, 𝑦 ≥ 0 since negative frequencies have
no meaning).
The algebraic method is a direct method to solve the relations (i), (ii), (iii) and
(iv) for the unknowns. It must be kept in mind that each rectangular game has a value
viz. 𝑣, which exists and is unique. Therefore, the main idea is to find such value 𝑣
which satisfies all the four relations.
SAQ 02-04-02: Solve the following game using algebraic method.
9 1 4
0 6 3
5 2 8
Solution: If [𝑥 , 𝑥 , 𝑥 ] and (𝑦 , 𝑦 , 𝑦 ) are optimal strategies of players A & B and V is
the value of the game, then we have following set of inequalities:
9𝑥 + 0𝑥 + 5𝑥 ≥ 𝑉 … (1)
𝑥 + 6𝑥 + 2𝑥 ≥ 𝑉 … (2)
4𝑥 + 3𝑥 + 8𝑥 ≥ 𝑉 … (3)
9𝑦 + 𝑦 + 4𝑦 ≤ 𝑉 … (4)
0𝑦 + 6𝑦 + 3𝑦 ≤ 𝑉 … (5)
5𝑦 + 2𝑦 + 8𝑦 ≤ 𝑉 … (6)
𝑥 +𝑥 +𝑥 = 1 … (7)
𝑦 +𝑦 +𝑦 =1 … (8)
𝑥 ,𝑥 ,𝑥 ,𝑦 ,𝑦 ,𝑦 ≥ 0 … (9)
Consider (1)-(6) as equations and then solve them with (7) and (8),
9𝑥 + 0𝑥 + 5𝑥 = 𝑉 … (10)
𝑥 + 6𝑥 + 2𝑥 = 𝑉 … (11)
4𝑥 + 3𝑥 + 8𝑥 = 𝑉 … (12)
9𝑦 + 𝑦 + 4𝑦 = 𝑉 … (13)
0𝑦 + 6𝑦 + 3𝑦 = 𝑉 … (14)
5𝑦 + 2𝑦 + 8𝑦 = 𝑉 … (15)
The system is solvable as such and the values of unknowns are;
𝑥 = ,𝑥 = ,𝑥 = ,𝑦 = ,𝑦 = ,𝑦 = 𝑎𝑛𝑑 𝑉 = .

04-02: METHOD OF SUBGAMES FOR 2×N AND M×2 GAMES
The subgame method sub-divides the given 2 × n or m × 2 game into a number
of 2 × 2 games, each of which is then solved and then the optimal strategies are
selected.
Rules for subgame method:
1) 2 × n games:
Select the lowest game value among all different game values. Determine the
optimal strategy according to the subgame of lowest game value.
2) m × 2 games:
Select the highest game value among all different game values. Determine
optimal strategy according to the subgame of highest game value.
Let us illustrate this method with the help of following example:
SOLVED PROBLEMS 02
Problem 02-04-03(2 × 3 game, No Dominance): Two airlines operate the same air-
route, both trying to get as large a market as possible. Based on a certain market, daily
gains and losses in rupees are shown in table given below, in which positive values
favor airline B. Find the solution for the game.
Airlines Airline B
Strategies Does Advertises Advertises special
Nothin special rates features (like movies,
g fine food)
Advertises special 275 -50 -75
Airline
rates
A
Advertises special
features (like 125 130 150
movies, fine
food)
Solution: For airline A, we have two strategies. Let us denote them by 1 and 2
respectively. Also, for airline B, we have three strategies. Let us denote them by 1, 2
and 3 respectively.
The pay-off matrix for a game is,
Players B
Strategies 1 2 3 Row min
1 275 -50 -75 -75
2 125 130 150 125(Maximin)

A Column 275 130 150
max
(Minimax)
As, Maximin = 125 ≠ Minimax = 130,
⟹ The game has no saddle point.
Also, we are unable to reduce it by dominance. This 2 × 3 game can be thought of as

three 2 × 2 subgames.
Subgame 01: (Ignoring column 3)
Strategies 1 2
A 1 275 -50
2 125 130
B
Strategies 1 3
A 1 275 -75
2 125 150
B
Strategies 2 3
A
1 -50 -75
2 130 150
Airline B which has a greater number of columns (than the number of rows for A), has
more flexibility, generally resulting in a better strategy. In order to find optimum
strategy for airline B, all the above three 2 × 2 subgames must be solved for their
strategies and game values. We shall solve them by arithmetic method.
Subgame 01: (There is no saddle point)
B
Strategies 1 2 oddments Probability
1 275 -50 5 1/66

A 2 125 130 325 65/66
oddments 180 150
Probability 36/66 30/66
=6/11 =5/11
Therefore,
The strategy for A is, A (1/66, 65/66); the strategy for B is, B (36/66, 30/66, 0) and
game value= Rs. [275 × + 125 × ] = Rs. 127.30
Subgame 02: (There is no saddle point)
B
1 275 -75 25 1/15
A 2 125 150 350 14/15
oddments 225 150
Probability 9/15=3/5 6/15=2/5
Therefore,
The strategy for A is A (1/15, 14/15); the strategy for B is B (9/15, 0, 6/15) and game
value= Rs. [275 × − 75 × ]= Rs. 135
Subgame 03: (There is a saddle point)
Strategies 2 3 Row min

A
1 -50 -75 -75
2 130 150 130(maximin)
Column 130 150

max
(minimax)
Thus, maximin = minimax = 130
⟹ The subgame has saddle point at (2, 2).
Therefore, the solution is: A (0, 1), B (0, 1, 0) and V= Rs. 130
Since, airline B has the flexibility to play any two out of the courses of action available
to it, it will play those strategies for which the loss is minimum. We can observe that all
the values for the subgames are positive, means airline A is winner. Hence, airline B
will play subgame 1 for which the loss is minimum i.e. Rs. 127.30.Hence, the complete
solution to problem is:

Strategies: A (1/66, 65/66), B (36/66, 30/66, 0)
Game value(V) = Rs. 127.30
Remark: Subgame with a saddle point cannot be solved by using arithmetic method. If
it is applied, the resultant solution will be incorrect.
Problem 02-04-04: Solve the following 3 × 2 game by method of subgames.
Strategies 1 2
1 8 1
2 5 3
3 -3 7
Solution: Let us denote row player by ‘A’ and column player by ‘B’.
We consider the subgames as follows:
Subgame 01:

1 8 1 1
2 5 3 3(maximin)
Column 8 3
max (minimax)
Here, maximin = minimax =3 means that the game has a saddle point.
Therefore, the solution of the subgame 01 is;
A (0, 1, 0), B (0, 1), Value of game (V)= 3.
Subgame 02:

2 5 3 3(maximin)
3 -3 7 -3
Column 5 7
max (minimax)
Here, maximin ≠ minimax means that the game has no saddle point .
So, we will solve it by using arithmetic method.

2 5 3 10 10/12=5/6
3 -3 7 2 2/12=1/6
oddments 4 8
Probability 4/12=1/3 8/12=2/3

Therefore, the solution for subgame 02 is:
A (0, 5/6, 1/6), B (1/3, 2/3), value of game (V)= 5 × + 3 × = .
Subgame 03:

1 8 1 1(maximin)
3 -3 7 -3
Column 8 7
max (minimax)
Here, maximin ≠ minimax means that the game has no saddle point.
Let us solve it by arithmetic method:

1 8 1 10 10/17
3 -3 7 7 7/17
oddments 6 11
Therefore, the solution for subgame 03 is:
A (10/17, 0, 7/17), B (6/17, 11/17), value of game (V)= 8 × −3× = .
The subgame with highest game value is subgame 02. Thus, the optimal solution to
given 3 × 2 game is: A (0, 5/6, 1/6), B (1/3, 2/3) and value of game (V)= 11/3.
SELF-TEST 02
MCQ 04-03: While solving 2 × n game with subgame method, the subgame with _____
gives optimal solution.
a) Lowest value
b) Highest value
c) Zero value
d) Saddle point
MCQ 04-04: While solving m × 2 game with subgame method, the subgame with _____
gives optimal solution.
a) Lowest value
b) Highest value
c) Zero value
d) Saddle point

SAQ 02-04-03: Solve the following 2 × n game by the method of subgames.
𝐴 1 3 11
𝐴 8 5 2
Solution: We have 3 subgames whose payoff matrices are as follows:
1 3 3 11 1 11
[ ], [ ], [ ].
8 5 5 2 8 2
Subgame 01: It has saddle point. The optimal strategies are A (0,1), B (0,1,0) and V=5.
Subgame 02: Without saddle point. So, the solution obtained by arithmetic method is;
A (3/11, 8/11), B (0, 9/11, 2/11) and V=49/11.
Subgame 03: Without saddle point. So, the solution obtained by arithmetic method is;
A (3/8, 5/8), B (9/16, 0, 7/16) and V=43/8.
Therefore, the required solution for given 2 × 3 game is:
The optimal strategies for players: A (3/11, 8/11), B (0, 9/11, 2/11) and
Value of the game: V=49/11.
(Students are expected to write complete arithme tic method for each subgame).
SAQ 02-04-04: Write a short note on: “Method of subgames”.
Solution: The subgame method sub-divides the given 2 × n or m × 2 game into a

number of 2 × 2 games, each of which is then solved by known method depending upon
the existence of saddle point and then the optimal strategies are selected.
Rules for subgame method:
1) 2 × n games:
Select the lowest game value among all different game values. Determine the
optimal strategy according to the subgame of lowest game value.
2) m × 2 games:
Select the highest game value among all different game values. Determine
optimal strategy according to the subgame of highest game value.

04-03: GRAPHIC METHOD FOR 2×N GAMES
This method is applicable to only those games in which one of the players has
two strategies only. The advantage of graphic method over other method is that it is
relatively fast.
Consider the 2 × n game represented by following pay-off matrix:
Players B
A Strategies 𝑦 𝑦 … 𝑦
𝑥 𝑎 𝑎 … 𝑎
= (1
−𝑥 )
It is assumed that the game has no saddle point .
Player A has two strategies 𝑥 and 𝑥 (= 1 − 𝑥 ) , where 𝑥 ≥ 0, 𝑥 ≥ 0 . The

expected pay-offs for A corresponding to the pure strategies of B are given in the
following table:
B’s pure strategies A’s expected pay-off
1 𝑎 𝑥 + 𝑎 (1 − 𝑥 ) = (𝑎 − 𝑎 )𝑥 + 𝑎
2 𝑎 𝑥 + 𝑎 (1 − 𝑥 ) = (𝑎 − 𝑎 )𝑥 + 𝑎
⋮ ⋮
n 𝑎 𝑥 + 𝑎 (1 − 𝑥 ) = (𝑎 − 𝑎 )𝑥 + 𝑎
So, from the above table, we can observe that A’s expected pay -off varies
linearly with 𝑥 . Thus, according to the maximin criterion for mixed strategies, A
should select that value of 𝑥 which will maximize his minimum expected pay-off. This
can be done as follows:
Plot the above lines as a function of 𝑥 . A number is allotted to each line

corresponding to B’s pure strategy. The lower boundary of these lines (shown by heavy
lines) gives the minimum expected pay-off to A as a function of 𝑥 . The highest point
on this lower boundary (shown by a dot), gives the maximin expected pay -off to A and
hence the optimum value 𝑥 = (𝑥 ). (see the fig. below).
The two optimum strategies for B are given by two lines which pass through this
maximin point. If more than two lines pass through the maximin point, then identify
two lines with opposite slopes. Thus, the 2 × n game is reduced to 2 × 2 game which
can be easily solved by methods discussed in previous units.

Algorithm to solve 2 × n game graphically:
Step 1: Reduce the size of payoff matrix of player A by applying dominance property, if
it exists.
Step 2: Let,
x = be the probability of selection of alternative 1 by player A and
1-x = be the probability of selection of alternative 2 by play er A.
Derive the expected gain function of player A with respect to each of the alternatives of
player B.
Step 3: Find the value of gain, when x=0 and x=1.
Step 4: Plot the gain functions on a graph by assuming a suitable scale [Keep x on X-
axis and the gain on Y-axis].
Step 5: Find the highest intersection point in the lower boundary of the graph →
Maximin point.
Step 6: If the no. of lines passing through the maximin point is only two, form a 2 × 2
payoff matrix and go to step 8. Otherwise go to step 7.
Step 7: Identify any two lines with opposite slopes passing through that point. Then,
form a 2 × 2 payoff matrix.
Step 8: Solve the 2 × 2 game using oddments and find the strategies for players A and B
and also value of the game.

SOLVED PROBLEMS 03
Problem 02-04-05: Solve the following 2 × 5 game by graphic method.
Players B
A Strategies 1 2 3 4 5
1 -5 5 0 -1 8
2 8 -4 -1 6 -5
Solution: For the pay-off matrix, initially we check saddle point.
Players B
A Strategies 1 2 3 4 5 Row min
1 -5 5 0 -1 8 -5
(maximin)
2 8 -4 -1 6 -5 -5
Column 8 5 0 6 8
max (minimax)
So, maximin= -5 ≠ minimax =0.
⟹ The game has no saddle point. Also, it cannot be reduced by dominance.
Let us find A’s expected pay-offs corresponding to B’s pure strategies.
`Let, 𝑥 = be the probability of selection of strategy 1 by player A.
1-𝑥 = be the probability of selection of strategy 2 by player A.
Thus, we have the following table:
B’s pure A’s expected pay-off

strategies
1 −5𝑥 + 8(1 − 𝑥 ) = −13𝑥 + 8
2 5𝑥 − 4(1 − 𝑥 ) = 9𝑥 − 4
3 0𝑥 − 1(1 − 𝑥 ) = 𝑥 − 1
4 −1𝑥 + 6(1 − 𝑥 ) = −7𝑥 + 6
5 8𝑥 − 5(1 − 𝑥 ) = 13𝑥 − 5
Let us plot these lines on graph. Draw two parallel lines 𝐴 and 𝐴 one unit apart and
mark a scale on each of them. These two lines represent the two strategies available to
A.
To represent B’s first strategy, join -5 on 𝐴 to 8 on 𝐴 ; to represent B’s second

strategy, join 5 on 𝐴 with -4 on 𝐴 and so on (for the remaining strategies).

The highest intersection point in the lower boundary is maximin point. It is
intersection of 𝐵 and 𝐵 . Thus, the 2 × 5 game can be easily reduced to following 2 × 2
game:
Players B
A Strategies 1 3 oddments Probability
1 -5 0 9 9/14
2 8 -1 5 5/14
oddments 1 13
Thus, optimum strategies are A (9/14, 5/14), B (1/14, 0, 13/14, 0, 0) and value of
the game is, V= −5 × +0× = .
Problem 02-04-06: Solve the following game by graphical method.
Players B
A Strategies 1 2 3
1 -4 2 -6
2 3 -9 4
Solution: This is 2 × 3 game.
Let, x= probability of selection of strategy 1 by player A
1-x= probability of selection of strategy 2 by player A
Therefore, the expected gain function for player A w. r. t. strategies of player B

is given in the following table:

B’s pure A’s expected A’s expected gain
strategy payoff function x=0 x=1
1 -4x+3(1-x)= - 3 -4
7x+3
2 2x-9(1-x) =11x- -9 2
9
3 -6x+4(1-x) =- 4 -6
10x+4
Let us plot the gain function of A on a graph by assuming suitable scale. Keep x on x -
axis and the gain on y-axis. When x=0, A’s expected gain is 3 and when x=1, A’s
expected gain is -4. Join them and denote it by 𝐵 . On similar lines, draw the lines for
A’s expected payoff function for B’s 2 nd and 3 rd strategy and denote them by 𝐵 and 𝐵
respectively.
Therefore, the graph obtained is represented by following figure in which maximin

point is intersection of 𝐵 and 𝐵 . Thus, the given 2 × 3 game gets reduced to the 2 × 2
game.
The reduced 2 × 2 game is,
Players B
A Strategies 2 3 Row min
1 2 -6 -
6(maximin)

2 -9 4 -9
Column 2(minimax) 4
max
As, maximin= 6 ≠ minimax= 2, the game is without saddle point. So, we solve it by
using arithmetic method.
Players B
A Strategies 2 3 oddments Probability
1 2 -6 13 13/21
2 -9 4 8 8/21
oddments 10 11
Therefore, the solution of game is,
A (13/21, 8/21), B (0, 10/21, 11/21) and the value of game (V)= 2 × −6× = − .
SELF-TEST 03
MCQ 04-05: While solving 2 × n game graphically, the highest intersection point in the
lower boundary of the graph is
a) Maximin point
b) Minimax point
c) Value of game
d) None of these
MCQ 04-06: In the graph of 2 × n game, if no of lines passing through the maximin
point are more than 2, then to obtain a reduced 2 × 2 pay-off matrix
a) Identify any two lines passing through maximin point with same slopes
b) Identify any two lines passing through maximin point with opposite slopes
c) Select any two lines at random
d) None of these

SAQ 02-04-05: Explain the graphical method to solve 2 × n game.
Solution: Consider the 2 × n game represented by following pay-off matrix:
Players B
A Strategies 𝑦 𝑦 … 𝑦

= (1 − 𝑥 )
It is assumed that the game has no saddle point.
Player A has two strategies 𝑥 and 𝑥 (= 1 − 𝑥 ) , where 𝑥 ≥ 0, 𝑥 ≥ 0 . The

expected pay-offs for A corresponding to the pure strategies of B are given in the
following table:
B’s pure A’s expected pay-off

strategies
1 𝑎 𝑥 + 𝑎 (1 − 𝑥 ) = (𝑎 − 𝑎 )𝑥 + 𝑎
2 𝑎 𝑥 + 𝑎 (1 − 𝑥 ) = (𝑎 − 𝑎 )𝑥 + 𝑎
⋮ ⋮
n 𝑎 𝑥 + 𝑎 (1 − 𝑥 ) = (𝑎 − 𝑎 )𝑥 + 𝑎
So, from the above table, we can observe that A’s expected pay -off varies linearly with
𝑥 . Thus, according to the maximin criterion for mixed strategies, A should select that
value of 𝑥 which will maximize his minimum expected pay-off. This can be done as
follows:
Plot the above lines as a function of 𝑥 . A number is allotted to each line

corresponding to B’s pure strategy. The lower boundary of these lines (shown by heavy
lines) gives the minimum expected pay-off to A as a function of 𝑥 . The highest point
on this lower boundary (shown by a dot), gives the maximin expected pay -off to A and
hence the optimum value 𝑥 = (𝑥 ). (see the fig. below)

The two optimum strategies for B are given by two lines which pass through this
maximin point. If more than two lines pass through the maximin point, then identify
two lines with opposite slopes. Thus, the 2 × n game is reduced to 2 × 2 game which
can be easily solved by methods discussed in previous units.
SAQ 02-04-06: Solve the following game graphically.
I 1 4 -2 -3
II 2 1 4 5
Solution: After drawing the graph, the given game gets reduced to following 2 ×2 game:
1 4
[ ]
2 1
After solving this game by arithmetic method, we obtain the following solution:
Optimal strategy for row player: [1/4, 3/4]
Optimal strategy for column player: [3/4, 1/4, 0, 0] and value of game V= 7/4.
04-04: GRAPHIC METHOD FOR M×2 GAMES

We can treat m × 2 games similar to 2 × n games, In case of m × 2 games, we
get minimax point, which will be the lowest point on the upper boundary. Thus, it can
be reduced to 2 × 2 game and then can be solved by one of the known methods.
Algorithm to solve m × 2 game graphically:
Step 1: Reduce the size of the pay-off matrix of player A by applying the dominance
property, if it exists.
Step 2: Let, y= be the probability of selection of alternative 1 by player B and
1-y= be the probability of selection of alternative 2 by player B.
Derive the expected gain function for player B w. r. t. each of the alternatives of player
A.
Step 3: Find the value of the gain, when y=0 and y=1.
Step 4: Plot the gain function on a graph by assuming a suitable scale. Keep y on X-
axis and the gain on Y-axis.
Step 5: Find the lowest intersection point in the upper boundary of the graph →
Minimax point.

Step 6: If the no of lines passing through the minimax point is only two, form a 2 × 2
pay-off matrix and go to step 8; otherwise go to step 7.
Step 7: Identify any two lines with opposite slopes passing through minimax point.
Then, form a 2 × 2 pay-off matrix.
and also the value of the game.
SOLVED PROBLEMS 04
Problem 02-04-07: Solve the following game by graphic method.
Players B
Strategies 𝑦 𝑦 𝑦 𝑦
A 𝑥 19 6 7 5
𝑥 7 3 14 6
𝑥 12 8 18 4
𝑥 8 7 13 -1
Solution:The given game is without saddle point. Let us try to reduce it by dominance.
As all cell values in column 2 are less than the corresponding values in columns 1 and
3, delete columns 1 and 3.
Therefore, the reduced matrix becomes;
Players B
Strategies 𝑦 𝑦
𝑥 6 5
A 𝑥 3 6
𝑥 8 4
𝑥 7 -1
Again, all the cell values for row 3 are higher than those for row 4. Hence, row 3
dominates row 4. Thus, delete row 4. With this the matrix gets reduced to,
Players B
Strategies 𝑦 𝑦 =1-𝑦
𝑥 6 5
A 𝑥 3 6
𝑥 8 4
The given game gets reduced to 3 × 2 game; let us solve this by graphic method.

B’s expected pay-offs corresponding to A’s pure strategies are given below.
A’s pure B’s expected pay-offs

strategies
1 6𝑦 + 5(1 − 𝑦 ) = 𝑦 + 5
2 3𝑦 + 6(1 − 𝑦 ) = −3𝑦 + 6
3 8𝑦 + 4(1 − 𝑦 ) = 4𝑦 + 4
These three straight lines can be plotted as functions of 𝑦 as shown in following graph:
Draw lines 𝐵 and 𝐵 parallel to each other, one unit apart and mark scaleon each of
them. These two lines represent the two strategies available to B. To represe nt A’s first
strategy, join 6 on 𝐵 with 5 on 𝐵 ; 3 on 𝐵 with 6 on 𝐵 and so on.
We take lowest intersection point on the upper boundary of graph, which is

minimax point and gives best strategies for A. The minimax point is intersection of 𝐴
and 𝐴 .
Therefore, the 3 × 2 game gets reduced to 2 × 2 game, which is given by;
Players B
1 6 5 3 3/4
2 3 6 1 1/4
oddments 1 3
A Probability 1/4 3/4

Therefore, the optimal strategies are; A (3/4, 1/4, 0, 0), B (0, 1/4, 0, 3/4) and the value
of the game is (V)= 6 × + 5 × = .
Problem 02-04-08: Solve the following 4 × 2 game graphically:
Players B
Strategies 1 2
1 6 -7
A
2 1 3
3 3 1
4 5 -1
Solution: Clearly, the game is without saddle point and cannot be further reduced by
dominance property.
Let, y= be the probability of selection of strategy 1 by player B and
1-y= be the probability of selection of strategy 2 by player B.
Now, we will find expected gain function of player B w. r. t. each of str ategies of player
A. Along with we find the value of the gain when y=0 and y=1.
A’s pure B’s expected payoff B’s expected gain

strategies function y=0 y=1
1 6𝑦 − 7(1 − 𝑦) = 13𝑦 -7 6
−7
2 𝑦 + 3(1 − 𝑦) = −2𝑦 3 1
+3
3 3𝑦 + 1(1 − 𝑦) = 2𝑦 + 1 1 3
4 5𝑦 − 1(1 − 𝑦) = 6𝑦 − 1 -1 5
Now, we plot the gain function on graph by assuming a suitable scale. Keep y on X -axis
and the gain on Y-axis.

Find the lowest intersection point in the upper boundary of the graph, which is minimax
point.
The minimax point is intersection of three lines 𝐴 , 𝐴 and 𝐴 . So, we identify

any two lines with opposite slopes passing through minimax point. Among these 3
lines, 𝐴 , 𝐴 are of opposite slopes and 𝐴 , 𝐴 are of opposite slopes. We can select any
of the obtained pairs of lines and form 2 × 2 pay-off matrix.
Suppose, we select 𝐴 , 𝐴 . Thus, the pay-off matrix will be;
Players B
A
2 1 3 1(maximin)
4 5 -1 -1
Column 5 3
max (minimax)
As, maximin ≠ minimax, the game has no saddle point.
So, we solve it by using arithmetic method.
Players B
2 1 3 6 6/8=3/4
A 4 5 -1 2 2/8=1/4
oddments 4 4
Probability 4/8=1/2 4/8=1/2
Therefore, the solution of game is;
A (0, 3/4, 0, 1/4), B (1/2, 1/2) and value of the game is (V)= 1 × + 3 × = 2.
SELF-TEST 04
MCQ 04-07: While solving m × 2 game graphically, the lowest intersection point in the
upper boundary of the graph is
a) Maximin point
b) Minimax point
c) Value of game
d) None of these
MCQ 04-08: In graphic method, if the no of lines passing through the minimax point is
more than two, then we select any two lines with
a) Same slopes

b) Opposite slopes
c) Positive slopes
d) Negative slopes

SAQ 02-04-07: Explain the steps of graphic method for solving m × 2 game
graphically.
Solution: Algorithm to solve m × 2 game graphically:
Step 1: Reduce the size of the pay-off matrix of player A by applying the dominance
property, if it exists.
Step 2: Let, y= be the probability of selection of alternative 1 by player B and
1-y= be the probability of selection of alternative 2 by player B.
Derive the expected gain function for player B w. r. t. each of the alternatives of player
A.
Step 3: Find the value of the gain, when y=0 and y=1.
Step 4: Plot the gain function on a graph by assuming a suitable scale. Keep y on X-
axis and the gain on Y-axis.
Step 5: Find the lowest intersection point in the upper boundary of the graph →
Minimax point.
Step 6: If the no of lines passing through the minimax point is only two, form a 2 × 2
pay-off matrix and go to step 8; otherwise go to step 7.
Step 7: Identify any two lines with opposite slopes passing through minimax point.
Then, form a 2 × 2 pay-off matrix.
and also the value of the game.
SAQ 02-04-08: Solve the following 5 × 2 game graphically:
𝐴 -6 7
𝐴 4 -5
𝐴 -1 -2
𝐴 -2 5
𝐴 7 -6
Solution: After drawing the graph, the given game gets reduced to following 2 ×2 game:

−2 5
[ ]
7 −6
We can solve this game by using arithmetic method and obtain the following solution:
Optimal strategy for A: [0, 0, 0, 13/20, 7/20];
Optimal strategy for B: [11/20, 9/20] and value of game V= 23/20.
SUMMARY
In this unit, we have studied the algebraic, graphic and subgame method to solve 2 × n
and m × 2 games in detail.

1. Solve the following 2 × 4 game by graphic method
Players B
Strategies 1 2 3 4
A 1 3 3 4 0
2 5 4 3 7
2. Solve the following game by using method of subgames.
Players B
A Strategies I II III IV
1 2 2 3 -1
2 4 3 2 6
3. Solve the following game by graphic method:
𝐴 -2 0
𝐴 3 -1
𝐴 -3 2
𝐴 5 -4
4. Write algorithm to solve 2 × n game graphically.
5. Write algorithm to solve m × 2 game graphically.
6. The following matrix represents the pay-off to row player A in a rectangular

game between two persons A and B
Strategies 1 2 3 4
1 8 15 -4 -2
2 19 15 17 16
3 0 20 15 5

By the notion of dominance, reduce the game to a 2 × 4 game and solve it
graphically.
7. Solve the game whose pay-off matrix is
Strategies 1 2 3
1 3 -2 4
2 -1 4 2
3 2 2 6
8. Solve the following game with row player A and column player B graphically:
Strategies 1 2
1 2 3
2 6 7
3 -6 10
4 -3 -2
5 3 2
KEY WORDS
2 × n games, m × 2 games, algebraic method, method of subgames, graphic method,
rules for subgames.
REFERENCES
BOOKS
& COMPANY LTD

PRAGATI EDITION
MOOCS
NPTEL course on operations research by IIT Roorkee by Prof.Kusumdeep
YOUTUBE VIDEOS
1) https://youtu.be/KUskbAasVCY
2) https://youtu.be/VWQIpwLmhGk
3) https://youtu.be/xCCI37IJgiI

WIKIPEDIA
https://www.geeksforgeeks.org/game-theory-normal-form-game-set-6-graphical-
method-2-x-n-game/
OER

C REDIT 03

UNIT 03-01: NETWORK ANALYSIS
LEARNING OBJECTIVES
 Understand the basic terms related to network
 Understand the method of drawing network diagram
 Understand the errors in drawing network diagram
INTRODUCTION
Government and industrial establishments always plan new projects for future
development and expansion. A project is a collection of well -defined tasks called
activities and when all these activities are carried out, the project is said to be
completed. The basic feature of all the projects is that these are non -repetitive and
involves hundreds and thousands of activities. These projects are of big-size and great
complexity. All these projects require huge expenditure and time, and management
cannot be guided by past experience in their planning and control.
Thus, there is a need of careful study and analysis of the entire project before
getting financial approval and signal to go ahead. The methods of network analysis will
help to study and analyze such projects and fulfill the objective of finding some
strategy so that project finishes in time with lowest cost.
01-01: EVENTS
Definition: Event
Event is an accomplishment that occur at a recognizable point in time. Event

occurs at an instantaneous point in time, but requires no time or resource itself.
Representation:
Events are usually identified by numbers written inside a circle or bracket viz. (2) or
represents 2 nd event.
Examples of event:
1) Design completed
2) Parts assembled
3) Pipelines laid

Types of Events:
1) Tail Event:
Event which marks the beginning of an activity is called Tail Event.
If a particular tail event represents the commencement of the project, it is known as

‘Initial Event’.
2) Head Event:
Event which represents the completion of an activity is called Head event.
If a particular head event makes the completion of the project, then it is called ‘Final
Event’ or ‘End Event’.
3) Successor and Predecessor Event:
Consider a following network;
The event which needs to be completed before the start of given event is known
as Predecessor Event, whereas the event which starts after the completion of given
event is said to be successor for given event.
In the given network diagram;
(1) is predecessor of (2)
(2) is successor of (1)
(3) is successor of (2).

SOLVED PROBLEMS 01
Problem 03-01-01: In the following network diagram, write down the predecessor and
successor events for each one.
Solution: In this network diagram,
Successor events:
(2) is successor of (1), (3) is successor of ( 1), (4) is successor of (2) and (3),
Predecessor events:
(1) is predecessor of (2) and (3), (2) is predecessor of (4), (3) is predecessor
of (4), (4) is predecessor of (5).
Problem 03-01-02: Write a short note on head event and tail event.
Solution:
Head event: Event which represents the completion of an activity is called Head event.
If a particular head event makes the completion of the project, then it is called ‘Final
Event’ or ‘End Event’.
Tail event: Event which marks the beginning of an activity is called Tail Event.

In the following figure, (1) is initial event, whereas ( 4) is end event. Also, for the
activity A: (1) is tail event and (2) is head event. On the similar lines, for activity B: (2)
is tail event and (3) is head event, for activity C: (3) is tail event and (4) is head event.

SELF-TEST 01
MCQ 01-01: Event which marks the beginning of an activity is called _______.
a) Head event
b) Tail event
c) Initial event
d) End event
MCQ 01-02: The event which starts after the completion of given event is known as
a) Successor event
b) Predecessor event
c) Head event
d) Tail event

SAQ 03-01-01: Write a short note on ‘Event’.
Solution: Event is an accomplishment that occur at a recognizable point in time. Event

occurs at an instantaneous point in time, but requires no time or resource itself.
Events are usually identified by numbers written inside a circle or bracket.
Examples of event: Design completed, Parts assembled, Pipelines laid etc. There are
three types of events viz. head event, tail event, successor and predecessor event.
Tail event: Event which marks the beginning of an activity is called Tail Event.
Head event: Event which represents the completion of an activity is called Head event.
Successor and Predecessor event: The event which needs to be completed before the
start of given event is known as Predecessor Event, whereas the event which starts after
the completion of given event is said to be successor for given event.
SAQ 03-01-02: Explain the various types of events.
Solution: There are three types of events:
1)Tail event: Event which marks the beginning of an activity is called Tail Event.


5) Head event: Event which represents the completion of an activity is called Head
event.
If a particular head event makes the completion of the project, then it is called
‘Final Event’ or ‘End Event’.
6) Successor and Predecessor event: The event which needs to be completed before the
start of given event is known as Predecessor Event, whereas the event which starts
after the completion of given event is said to be successor for given ev ent.
In the given network diagram;
(2) is successor of (1)
01-02: ACTIVITIES
Definition: Activity
An activity represents a job or a task that has to be carried out to complete a

specific event.
Representation:
Activity is represented by an arrow (⟶). The length of arrow may represent the
time taken for the completion of work and its direction will indicate the direction of

project. The event lying on the tail of the arrow must be completed before the start of
the operation on the right end (head) of the arrow. Thus, there can be only one activity
between any two events.
If (1) is event towards tail and (2) is event towards head of an arrow i.e.
(1) ⟶(2) then the activity can also be written as 1-2. Activities are associated with
intervals of time over which they are performed.
Examples of activity:
1) Mix concrete
2) Design a plan of building
3) Assemble parts
Types of Activities:
1) Parallel Activities:
The activities which can be performed simultaneo usly and independently to each
other are called parallel activities.
For example:
2) Serial Activities:
The activities which are performed one after the another in succession; are serial
activities.
For example:
3) Predecessor and Successor Activities:
The activity which occurs immediate before the given activity is said to be
predecessor activity whereas the activity which occurs immediate after the given
activity is known as the successor activity.

For example: In the network given below;
P is predecessor activity of Q.
Q is predecessor activity of R.
Q is successor activity of P.
R is successor activity of Q.
4) Dummy Activity:
A dummy is type of activity in a network which needs zero time to perform and
does not require any resources but can be easily used in the network to depict the logic
of the relationship between various activities of the project.
The concept of dummy enables all activities in a project to be shown by only one
arrow for each. In other words, a dummy is some job or task which stipulates a
necessary time orientation but the activity takes no time or resources in itself. These are
used to represent a situation where one event cannot take place until a previous event
has taken place although this requires no time or resources i.e. these are used to
maintain proper precedence relations between two events and is denoted by broken
arrow.
Representation:
For example:
Consider a situation where A and B are concurrent activities. C is dependent on

A and D is dependent on A and B both. Such situation can be handled by using a
dummy activity as shown in the fig. below:

SOLVED PROBLEMS 02
Problem 03-01-03 (More than one dummy activity): Draw the network diagram:
A < B, C, D, G; C, D < E; B < H; H, G, F < I; B, E < F
[NOTE: A < B represents activity A is preceded by activity B.]
Solution: Initially, we convert given problem into standard format as follows:
Activity Predecessor
A -
B A
C A
D A
E C, D
F B, E
G A
H B
I H, G, F
We can observe that activity A has no predecessor, which means A is initial activity.
Also, A acts as predecessor activity for B, C, D and G.
Then, B acts as predecessor activity for F and H, whereas C and D acts as predecessor
activities for E. Also, G, H and F acts as predecessor for I. So, we can draw network
diagram as follows:
As, C and D have same tail event, they cannot have same end event. So, we need to add
dummy activity D1. On similar lines, by rules of drawing network diagrams (Discussed
in detail in section 01-03), we need to add dummy activities D2 and D3.
Problem 03-01-04: Draw the network diagram for the following problem:

Activity Preceding Activity
A -
B -
C -
D A, B
E A, B
F C, D, E
G C, D, E
Solution:
SELF-TEST 02
MCQ 01-03: An ________ represents a job or a task that has to be carried out to
complete a specific event.
a) Event
b) Dummy
c) Activity
d) Network
MCQ 01-04: The activities which can be performed simultaneously and independently
to each other are known as
a) Parallel activities
b) Serial activities
c) Predecessor activities
d) Successor activities
MCQ 01-05: The type of activity in a network which needs zero time to perform and
does not require any resources but used to depict the logic of relationship between
various activities of project is _____ activity.
a) Parallel

b) Serial
c) Dummy
d) Successor

SAQ 03-01-03: Define: activity and write down the types of activities.
Solution: An activity represents a job or a task that has to be carried out to complete a
specific event.
Examples of activity: Mix concrete, design a plan of building, Assemble parts etc.
The following are the types of activities:
1) Parallel Activities: The activities which can be performed simultaneously and

independently to each other are called parallel activities.
2) Serial Activities: The activities which are performed one after the another in
succession; are serial activities.
3) Predecessor and Successor Activities: The activity which occurs immediate

before the given activity is said to be predecessor activity whereas the activity
which occurs immediate after the given activity is known as the successor
activity.
4) Dummy Activity: A dummy is type of activity in a network which needs zero

time to perform and does not require any resources but can be easily used in the
network to depict the logic of the relationship between various activities of the
project.
SAQ 03-01-04: Write a short note on ‘Dummy activity’ in the network.

Solution: A dummy is type of activity in a network which needs zero time to perform
and does not require any resources but can be easil y used in the network to depict the
logic of the relationship between various activities of the project.
The concept of dummy enables all activities in a project to be shown by only
one arrow for each. In other words, a dummy is some job or task which stip ulates a
necessary time orientation but the activity takes no time or resources in itself. These are
used to represent a situation where one event cannot take place until a previous event
has taken place although this requires no time or resources i.e. the se are used to
maintain proper precedence relations between two events and is denoted by broken
arrow.
For example:

Consider a situation where A and B are concurrent activities. C is dependent on A and D
is dependent on A and B both. Such situation can be handled by using a dummy activity
as shown in the fig. below:
01-03: DRAWING NETWORK DIAGRAMS

Network:
Network is a combination of several activities and events. A network can be

defined by a set of points or nodes that are connected by lines or arrows. The links in a
network are generally characterized in terms of time, cost or distance involved in
traversing them.
The problem of networks is to find a course of action which minimizes some

measure of performance. The measure can be time, cost, di stance travelled etc.
Some other terms related to the network:
1) Chain: It is the sequence of adjacent activities.
For example:
2) Loop: It is the connection of activity to an earlier event on the same path. It can
be avoided by introducing dummy activities.
For example:
3) Source: It is the starting event of a project. It is an event with only succeeding

but no preceding activity.
4) Sink: The last event showing the end of a project is known as sink. It is an event
with only preceding but no succeeding activity.
5) Circuit: A finite path in which the starting event connects with end event.

Rules for drawing network diagram:
There are number of rules in connection with the handling of events and
activities of a project network which are given below:
1) Each activity is represented by one and only one arrow in the network.
2) No two activities can be identified by the same end events. If occurs then such
error can be rectified by adding dummy activity.
3) The network has only one entry point called the start event and one poi nt of
emergence called end event.
4) Use dummy activity freely in rough graph but final network should have only
reluctant dummy.
5) Two events are numbered in such a way that the event of higher number can
happen only after the event of lower number is completed.
6) Use straight arrows which will be from left to right. Try to avoid arrows which
cross each other.
SOLVED PROBLEMS 03
Problem 03-01-05: Draw the network diagram for the following stated problem:
Activity Description Predecessor

Activity
A Open job order -
B Get material A
C Get tools A
D Cutting B
E Bending D
F Welding E
G Finishing F, C
H Testing G
Solution:

Problem 03-01-06: Draw network diagram for the following problem:
Activity Preceded by
A -
B -
C A
D B
E A
F B
G C, D
H G, F
I E
J H, I
K J
Solution:
SELF-TEST 03
MCQ 01-06: The connection of activity to an earlier event on the same path is nothing
but
a) Chain
b) Loop
c) Source
d) Sink
MCQ 01-07: The last event showing end of the project is known as

a) Chain
b) Loop
c) Source
d) Sink

SAQ 03-01-05: Define the terms: Chain, Loop, Source, Sink, Circuit
Solution: The definitions are as follows:
Chain: It is the sequence of adjacent activities.
Loop: It is the connection of activity to an earlier event on the same path. It can be
avoided by introducing dummy activities.
Source: It is the starting event of a project. It is an event with only succeeding but no
preceding activity.
Sink: The last event showing the end of a project is known as sink. It is an event with
only preceding but no succeeding activity.
Circuit: A finite path in which the starting event connects with end event.
SAQ 03-01-06: Write down the rules for drawing network diagram.
Solution: There are number of rules in connection with the handling of events and
activities of a project network which are given below:
1) Each activity is represented by one and only one arrow in the network.
2) No two activities can be identified by the same end events. If occurs then such
error can be rectified by adding dummy activity.
3) The network has only one entry point called the start event and one point of
emergence called end event.
4) Use dummy activity freely in rough graph but final network should have only
reluctant dummy.

5) Two events are numbered in such a way that the event of higher number can
happen only after the event of lower number is completed.
6) Use straight arrows which will be from left to right. Try to avoid arrows which
cross each other.
01-04: ERRORS IN DRAWING NETWORKS

Three types of errors are most commonly observed in drawing network
diagrams.
1) Dangling:
To disconnect an activity before the completion of all activities in a

network diagram is known as dangling.
As shown in the figure below; activity B is not la st activity in network

and has no connection with end event.
2) Looping or Cycling:
Looping error is also known as cycling error in a network diagram. Drawing an

endless loop in a network is known as error of looping.
For example: In the network diagram given below, 3-4-5 is a loop.
3) Redundancy:
Unnecessarily inserting the dummy activity in network logic is known as error

of redundancy as shown in following diagram:

Characteristics of network analysis:
The following are important characteristics in a network analysis:
1) The objective is to be finished within a specific time, otherwise there is a

penalty.
2) Various activities are to be completed in a certain order and there are some
activities which can start on the completion of other activities.
3) The cost associated with any activity is proportional to its time of completion.
4) There can be bottlenecks in the process and the resources to be allocated are
limited.
Analysis of network diagrams:
By enumerating the various sequences of activities in a network di agram and

calculating the duration of activities in each sequence, one can find critical path.
The techniques of finding critical path systematically are discussed in later units of
this section.
SOLVED PROBLEMS 04
Problem 03-01-07: Define: Dangling error and explain with the help of an example.
Solution:
Definition: To disconnect an activity before the completion of all activities in a

network diagram is known as dangling.
Consider the following network diagram:
In this network diagram, dangling error is observed in the activities 5 to 8 and 6 to 7 as

there is no connection next to these activities.

Problem 03-01-08: Consider the following network diagrams and observe the type of
error in them.
Solution: we know that, drawing an endless loop in a network is known as error of

looping. In figure 1, activities 5-6-7-8 makes an endless loop and hence the error
observed in figure 1 is looping.
Also, we know that, the dangling error is obtained when there is d isconnection
to the activities before the completion of all activities. In figure 2, activities 5 to 6 and
5 to 8 represents dangling error.
SELF-TEST 04
MCQ 01-08: The error obtained by unnecessary insertion of dummy activity in the
network logic is known as
a) Dangling
b) Looping
c) Redundancy
d) None of these
MCQ 01-09: Which of the following is not error in drawing network diagrams?
a) Dangling
b) Looping
c) Redundancy

d) Chain

SAQ 03-01-07: Write a short note on errors in drawing network diagrams.
Solution: Three types of errors are most commonly observed in drawing network
diagrams.
1) Dangling: To disconnect an activity before the completion of all activities in a

network diagram is known as dangling. As shown in the figure below; activity B is
not last activity in network and has no connection with end event.
2) Looping or Cycling: Looping error is also known as cycling error in a network

diagram. Drawing an endless loop in a network is known as error of looping.
For example: In the network diagram given below, 3-4-5 is a loop.
3) Redundancy: Unnecessarily inserting the dummy activity in network logic is

known as error of redundancy as shown in following diagram:
SAQ 03-01-08: Write down the characteristics of network analysis.
Solution: The following are the characteristics of network analysis:
1) The objective is to be finished within a specific time, otherwise there is a

penalty.

2) Various activities are to be completed in a certain order and there are some
activities which can start on the completion of other activities.
3) The cost associated with any activity is proportional to its time of completion.
4) There can be bottlenecks in the process and the resources to be allocated are
limited.
SUMMARY
In this unit, we have studied the basic terminologies relat ed to network diagram.
We have also studied the method of drawing network diagrams.

1. Draw the network diagram for the following:
Activity Predecessor
A -
B A
C A
D B
E C
F C
G D, E, F
H G
2. Write down the successor and predecessor events for each of the activity in the
following network diagram:
3. Define: Head event, Tail event, Initial event, End event and explain with the
help of example.
4. Construct a network diagram for the following:
Job Successor
Activity
S A
A B
B D
C E

D E
E F
F None
5. A motorist drives into a garage, has his tank filled with petrol, has oil checked
and topped up, his tire pressures adjusted, pays for the petrol and oil and drives
off. Prepare an activity dependence table and the network for the problem.
6. For the following network diagram, prepare a table containing activities and
predecessor activities.
7. Draw the following network diagram:
A project consists of 6 jobs A, B, C, D, E and F. Number the events if (i) B and

C depends only on A, (ii) D depends on B but not on C, (iii) Job E depends on C
and B (iv) The project is completed when D and E are done.
8. Draw the following network diagram:
A, D and G start at the origin; J follows F but precedes K; C follows A but

precedes J; H follows D but precedes L; B follows A but precedes F; K, L and E
are terminal activities; K follows G and H; E follows B ; F is independent of C;
L is independent of J.
KEY WORDS
Event, Activity, Dummy activity, Network, Dangling, Looping, Cycling,
Redundancy.
REFERENCES
BOOKS
& COMPANY LTD

PRAGATI EDITION

MOOCS
YOUTUBE VIDEOS
1) https://youtu.be/fDeCjk8EVGU
2) https://youtu.be/gEBMPP0SOnU
3) https://youtu.be/ChF6FkW4I6c
4) https://youtu.be/mHxI-3MUJ0U
5) https://youtu.be/z5-YGBn4mMw
WIKIPEDIA
https://mathstudy.home.blog/2019/05/12/error-in-drawing-network/
OER

UNIT 03-02: FULKERSON ’S RULE
LEARNING OBJECTIVES
 Understand the application of Fulkerson’s rule to number the events in network

diagram
 Understand the basics of Project management
 Understand various aspects of Time-cost analysis
 Understand various time and cost estimates
 Understand crashing of project network
INTRODUCTION
Network diagrams are drawn according to sequence of execution of activities.
For network representation, it is necessary that various nodes be properly labelled.
Nodes are introduced indicating completion of one or more activities and starting of
one or more activities. If network diagram is complex, it seems difficult to number the
events. A standard procedure for numbering the events developed by D. R. Fulkerson is
most commonly used for this purpose.
02-01: FULKERSON’S RULE

Following is step by step procedure of applying Fulkerson’s 𝑖 − 𝑗 rule: -
Step 1: Find the start event and label it as number ‘1’. A start event is one which has
arrows emerging from it but not entering it.
Step 2: Delete all arrows emerging from all numbered events. This will create at least
one new start event out of the preceding events.
Step 3: Number all new start events ‘2’, ‘3’ and so on. No definite rule is necessary but
numbering from top to bottom may facilitate other u sers using the network when there
is more than one new start event.
Step 4: Go on step no. 2 & 3, repeating until the end reaches.
SOLVED PROBLEMS 01
Problem 03-02-01: Number the events of the network shown in the following fig. with
the help of Fulkerson’s rule.

Solution: As event ‘a’ is starting event (initial event); hence number it as ‘1’. Due to
activity K emerging out of a and ending at event b, the end of activity will be the new
initial event and number it as 2.
There are two arrows L and M emerging out of event 2. So, we have two more
events 3 and 4. Repeating the same procedure and neglecting the ends, we obtain next
numbering for given events. Therefore, the required network is,
Problem 03-02-02: A project consists of seven activities. Activities P, Q, R run

simultaneously. The relationship among the various activities is as follows:
Activity Immediate
Successor
P S
Q T
R U
Activity V is the last operation of project and it is also immediate successor to S. T and
U. Draw the network of project and hence number the events by using Fulkerson’s rule.
Solution: As activities P, Q, R run simultaneously, these are concurrent activities

initiating from starting event. Then, S, T and U are immediate successors to activities P,
Q and R respectively. Also, V is the last operation or we can say immediate successor to
S, T and U.
Therefore, the required network diagram obtained is as follows:

SELF-TEST 01
MCQ 02-01: A standard procedure for numbering the events in network diagram was
developed by
a) Watson
b) Fulkerson
c) Neumann
d) None of these
MCQ 02-02: Network diagrams are drawn according to sequence of execution of

______.
a) Activities
b) Events
c) Nodes
d) None of these
MCQ 02-03: New start event can be created by _____ all arrows emerging from all
numbered events.
a) Adding
b) Multiplying
c) Deleting
d) Counting

SAQ 03-02-01: Explain the ‘Fulkerson’s Rule’ for numbering the events in a network
diagram.
Solution: If network diagram is complex, it seems difficult to number the events. A

standard procedure for numbering the events developed by D. R. F. Fulkerson is most
commonly used for this purpose.
Following is step by step procedure of applying Fulkerson’s 𝑖 − 𝑗 rule: -
Step 1: Find the start event and label it as number ‘1’. A start event is one which has
arrows emerging from it but not entering it.
Step 2: Delete all arrows emerging from all numbered events. This will create at least
one new start event out of the preceding events.

Step 3: Number all new start events ‘2’, ‘3’ and so on. No definite rule is necessary but
numbering from top to bottom may facilitate other users using the network when there
is more than one new start event.
Step 4: Go on repeating step no 2 and 3 until the end reached.
SAQ 03-02-02: According to Fulkerson’s rule, what are the correct event numbers
corresponding to events a, b, c, d in the following network diagram:
Solution: a=1, b=2, c=3, d=4.
02-02: PROJECT MANAGEMENT

1) A project consists of interrelated activities which are to be executed in certain order
before the entire task is completed.
2) The activities are interrelated in a logical sequence which is known as precedence

relationship.
3) Project is represented in form of network for the purpose of analytical treatment to

get solutions for scheduling and controlling its activities.
There are two basic techniques of project management:
i) Critical Path Method (CPM)
ii) Project Evaluation and Review Technique (PERT)
Phases of Project Management:
1) Planning: Divide the project into distinct activities and estimate the time
requirement for each activity. Then, establish the precedence relationships among
the activities followed by the construction of an arrow (network) diagram.
2) Scheduling: Determine the start and end time of each and every activity.
3) Controlling: Continuous monitoring and progress reporting is done by making use

of the arrow diagram and time chart.
Critical Path:
The path forming an unbroken chain of critical activities from the start event to
the end event is called the critical path.

The critical path of project network is the longest path in the network. It can be
identified by simply listing out all the possible paths from the start node to end node of
the project and then selecting the path with the maximum sum of activity times on that
path. All the activities lying on the critical path are called critical activities; as any
delay in their execution will lead to delay in the completion of the entire project.
SOLVED PROBLEMS 02
Problem 03-02-03: Determine the critical path for the following network:
Solution: There are following three paths in the given network;
Path I: 1→2→3→6→7
Path II: 1→2→4→6→7
Path III: 1→2→5→7
The project duration along path I is 3+4+4+3=14, Path II is 3+3+6+3=15 and

Path III is 3+8+5=16. Hence, the longest path is Path III with project duration of
16 weeks. Thus, 1→2→5→7 is critical path.
Problem 03-02-04: Determine the critical path for the following network:
Solution: There are following three paths in the given network;
Path I: 1→2→3→6→7
Path II: 1→2→4→6→7
Path III: 1→2→5→6→7

The project duration along path I is 5+4+8+4=21, Path II is 5+3+6+4=18 and Path
III is 5+8+7+4=24. Hence, the longest path is Path III with project duration of 24
weeks. Thus, 1→2→5→6→7 is critical path.
SELF-TEST 02
MCQ 02-04: Which of the following is not the phase of project management?
a) Planning
b) Scheduling
c) Controlling
d) Breaking
MCQ 02-05: The path forming an unbroken chain of critical activities from the start
event to the end event is called
a) Critical path
b) Continuous path
c) Shortest path
d) Tree

SAQ 03-02-03: Write a short note on project management.
Solution: A project consists of interrelated activities which are to be executed in

certain order before the entire task is completed. The activities are interrelated in a
logical sequence which is known as precedence relationship. Project is represented in
form of network for the purpose of analytical treatment to get solutions for scheduling
and controlling its activities.
There are two basic techniques of project management:
i) Critical Path Method (CPM)
ii) Project Evaluation and Review Technique (PERT).
SAQ 03-02-04: What are the various phases of project management? Explain.
Solution: There are three different phases of project management:
1) Planning: Divide the project into distinct activities and estimate the time
requirement for each activity. Then, establish the pre cedence relationships among
the activities followed by the construction of an arrow (network) diagram.
2) Scheduling: Determine the start and end time of each and every activity.

3) Controlling: Continuous monitoring and progress reporting is done by making use
of the arrow diagram and time chart.
02-03: TIME-COST TRADE -OFF ANALYSIS

Due to limited availability of the resources and the funds to complete some
work/project, the focus of organization is on use of these resources and money in most
efficient way. In network analysis, the total project duration is evaluated on the basis of
normal level of resources and efficient work. Project involves number of activities and
there is a direct relationship between cost and the time of completion of an activity.
The reduction in target time cannot be done arbitrarily. The time given by the
critical path activities provides the total project time and any reduction in target time
can be possible only in reduction with the activities lying on the critical path.
The appropriate measures of time reduction can be:
(i) Increasing manpower
(ii) Introducing shift system or overtime
(iii) Installing more machinery or equipment
(iv) Use of special material and improved methods
(v) Reducing safety margins and lowering standards
(vi) Selecting vendors or sub-contractors who charge extra but expedite

deliveries
(vii) Revising drawings, altering the sequence of operations etc.
All these measures involve additional expenditure and it is always desirable to

analyze the increase in cost associated with each measure before finding some final
strategy. The analysis in such situation requires some decision criteria that will decide
which activity’s duration is to be minimized and what will be additional expenditure.
Definitely, one would like to formulate a least expensive time reduction strategy. The
procedure is known as time-cost trade-off analysis.
Aspects of Time-Cost Analysis:
(1) To find the total duration of a project for which the cost is minimum
(2) To find the least expenditure strategy to mi nimize the duration by some
desired/specified period
(3) To determine the time-cost risk relationships of the network technique

Time and Cost Estimates:
1) Normal cost and time:
The minimum cost of carrying out the activity in minimum time without
allocating extra resources is termed as Normal cost. Similarly, the minimum
time required at the normal cost is known as Normal time.
2) Crash cost and time:
Crash cost is the minimum cost of carrying on any activity using additional
resources and the corresponding minimum time to complete an activity is known
as Crash time. It is the minimum time beyond which an activity time cannot be
crashed even by incurring additional cost.
3) Project normal cost/time:
The sum of the normal costs (normal times) of all activities is known as Project
normal cost (Project normal time). Similarly, the project crash cost/time can be
defined.
4) Direct cost:
Direct costs are the costs that can be identified with the activity. These includes
the costs incurred for a project i.e. include expenditure on plant, machineries,
land, building, engineering charges etc. These expenditures are termed as direct
as it is possible to allocate these costs directly to the activities. As we go on
reducing the activity time, cost goes on increasing.
5) Indirect cost:
These are the costs, which cannot be identified with the activity. The longer the
duration of project, the higher is the indirect cost.
For example: Administration and establishment charges, expenditure on

inventories, loss of revenue, etc.
Indirect costs can be further divided into two parts:
(i) Fixed indirect costs: which does not depend upon the progress of the project
(due to general and administrative expenses etc.)
(ii) Variable indirect costs: depends upon the duration of the project (like
expenditure, supervision etc.)
The indirect cost decreases with reduction in the duration of project.
6) Total cost:

It is the sum of direct and indirect costs associated with a project. As the direct
cost decreases with time and indirect cost increases with time, th e total project
cost curve will have a point where the total cost will be minimum. The time
corresponding to this point is called the optimum duration time and the cost, the
optimum cost.
7) Cost slope:
Cost slope indicates additional cost incurred per unit o f time saved in reducing
the duration of an activity. In other words, cost slope is the slope of the direct
cost curve, approximated as a straight line. It is given by
𝐶𝑟𝑎𝑠ℎ 𝑐𝑜𝑠𝑡 − 𝑁𝑜𝑟𝑚𝑎𝑙 𝑐𝑜𝑠𝑡

𝐶𝑜𝑠𝑡 𝑆𝑙𝑜𝑝𝑒 =
𝑁𝑜𝑟𝑚𝑎𝑙 𝑡𝑖𝑚𝑒 − 𝐶𝑟𝑎𝑠ℎ 𝑡𝑖𝑚𝑒
SOLVED PROBLEMS 03
Problem 03-02-05: Find out the cost slopes for each of the activities in the following
problem:
Activity Normal Time Normal Cost Crash Time Crash Cost

(in weeks) (in Rs.) (in weeks) (in Rs.)
1-2 7 700 4 850
1-3 5 500 3 700
1-4 8 600 5 1200
2-5 9 800 7 1250
3-5 5 700 3 1000
3-6 6 1100 5 1300
4-6 7 1200 5 1450
5-7 2 400 1 500
6-7 3 500 2 850
Solution: We know that, the cost slope is given by;

Activity Normal Time Normal Cost Crash Time Crash Cost Cost
(in weeks) (in Rs.) (in weeks) (in Rs.) Slope
1-2 7 700 4 850 50
1-3 5 500 3 700 100
1-4 8 600 5 1200 200
2-5 9 800 7 1250 225
3-5 5 700 3 1000 150
3-6 6 1100 5 1300 200
4-6 7 1200 5 1450 125
5-7 2 400 1 500 100
6-7 3 500 2 850 350

Problem 03-02-06: Find out the cost slopes for each activity and project normal cost
for the following data:
Activity Normal Time Normal Cost Crash Time Crash Cost

(in weeks) (in Rs.) (in weeks) (in Rs.)
1-2 9 640 6 700
1-3 8 500 5 575
1-4 15 400 10 550
2-4 5 100 3 120
3-4 10 200 6 260
4-5 2 100 1 140
Solution: We know that, the cost slope is given by;

Project normal cost = Sum of normal costs of all the activities.
= Rs. (640+500+400+100+200+100) = Rs. 1940.
Activity Normal Time Normal Cost Crash Time Crash Cost Cost
(in weeks) (in Rs.) (in weeks) (in Rs.) Slope
1-2 9 640 6 700 20
1-3 8 500 5 575 25
1-4 15 400 10 550 30
2-4 5 100 3 120 10
3-4 10 200 6 260 15
4-5 2 100 1 140 40
SELF-TEST 03
MCQ 02-06: _________ is the sum of direct and indirect costs associated with a
project.
a) Total cost
b) Cost slope
c) Crash cost
d) Normal cost
MCQ 02-07: _________ does not depend upon the progress of the project.
a) Variable indirect cost
b) Fixed indirect cost
c) Direct cost

d) None of these

SAQ 03-02-05: What are the appropriate measures of time reduction ?
Solution: The appropriate measures of time reduction are:
(i) Increasing manpower
(ii) Introducing shift system or overtime
(iii) Installing more machinery or equipment
(iv) Use of special material and improved methods
(v) Reducing safety margins and lowering standards
(vi) Selecting vendors or sub-contractors who charge extra but expedite

deliveries
(vii) Revising drawings, altering the sequence of operations etc.
SAQ 03-02-06: Explain: Normal cost and time, Crash cost and time, Project normal
cost and time.
Solution:
1) Normal cost and time: The minimum cost of carrying out the activity in
minimum time without allocating extra resources is termed as Normal cost.
Similarly, the minimum time required at the normal cost is known as Normal
time.
2) Crash cost and time: Crash cost is the minimum cost of carrying on any activity
using additional resources and the corresponding minimum time to complete an
activity is known as Crash time. It is the minimum time beyond which an
activity time cannot be crashed even by incurring additional cost.
3) Project normal cost/time:
The sum of the normal costs (normal times) of all activities is known as Project
normal cost (Project normal time). Similarly, the project crash cost/time can be
defined.
02-04: TIME-COST OPTIMIZATION/NETWORK CRASHING

The process of shortening the duration of a project is known as Crashing. It is
implemented when project manager want to shorten the project completion time by
spending extra resources (more money). The analysis of network and the individual

activities for crashing the time up to some desired level can be classified in the
following categories:
(i) The target time is to be crashed to some desired level without any cost
consideration
(ii) The target time is to be crashed to some desired level with cost
consideration
(iii) The target time is crashed to a level where increase in profits due to
additional production remains more than the additional cost of time crashing.
Objectives of Project Crashing:
The following are some of the main objectives of project crashing:
(i) To minimize the normal duration of the project.
(ii) To evolve a strategy of time crashing with minimum possible cost of

crashing.
(iii) If delayed completion of some activities at any stage introduces the

possibility of enhancing project duration, then time crashing by expediting one
or more remaining activities on the critical path is to be explored.
(iv) To spare the resources at an early time for allocation to more profitable
jobs/activities.
(v) To ensure balanced and uniform requirement of resources.
Time Crashing procedure:
Step 1: Identify the critical path activities and other non -critical path activities which
are likely to become critical during the course of crashing.
Step 2: Calculate the cost slope for the activities identified in step 1.
Step 3: Rank the activities lying on the critical path in ascending order of cost slope.
Step 4: Crash those activities first which have the smallest cost slope to the maximum
possible extent and calculate the new direct cost by cumulative addition of the cost of
crashing to the normal cost.
Step 5: Parallel Crashing: During the course of crashing the activities lying on the
critical path, there is always a likelihood that other paths which were non -critical

before crashing may now become critical. Then the project duration can be reduced by
crashing the activities of parallel critical paths simultaneously.
Thus, the lowest cost solution can be obtained by sequentially crashing the
critical path activity and parallel critical path activity for which cost slope is lowest.
The procedure of time crashing is explained with the help of following example:
SOLVED PROBLEMS 04
Problem 03-02-07: For the following data, find the critical path, the target time and
cost of the project, if target time is reduced to 21 weeks; keeping into consideration the
cost of crashing for various activities at each stage.
Activity Normal Time Normal cost Possible crash time Crash cost per week
in weeks in Rs. in weeks (cost slope)
1-2 4 3000 1 300
2-4 4 3600 1 200
4-7 6 8000 2 1250
7-9 8 12800 3 1667
1-5 5 7500 2 850
5-6 10 14400 4 450
6-7 7 8400 3 467
1-3 3 2400 0 0
3-6 6 2600 2 125
6-8 9 8600 5 960
8-9 2 1100 1 150
9-10 3 1600 1 250
Solution: Initially, we draw the network diagram in order to determine the critical path.
Using the given data, the network can be drawn as follows:
We can observe, there are five different paths from node 1 to node 10. The table
containing all these five paths along with the corresponding target time is as follows:
Path Sequence of Sum of the normal times of various Target time in

activities activities weeks
𝑃 1-2-4-7-9-10 4+4+6+8+3 25
𝑃 1-5-6-8-9-10 5+10+9+2+3 29
𝑃 1-5-6-7-9-10 5+10+7+8+3 33
𝑃 1-3-6-8-9-10 3+6+9+2+3 23
𝑃 1-3-6-7-9-10 3+6+7+8+3 27
Here, 𝑃 is the path with maximum target time of 33 weeks, so 𝑃 is the critical path.

Now, to prepare a table containing the crashed activities at various stages along with
the target times for various paths after crashing, we will follow the following
procedure:
i) As, 𝑃 is critical path, we will consider the following table for activities
of𝑃 :
Activities on path 𝑃 Crash time in weeks Crash cost in Rs. /week

1-5 2 850
5-6 4 450
6-7 3 467
7-9 3 1667
9-10 1 250
In the above table, activity 9-10 is with minimum cost slope, which is 250 per week.
So, we select this activity for crashing by one week and get the following target times
for various paths.
𝑃 𝑃 𝑃 𝑃 𝑃
24 28 (32) 22 26
Crash cost = 250 × 1 = Rs. 250. (1)
ii) Again, 𝑃 is critical path with next activity with lowest cost slope is activity
5-6 with crash cost of Rs. 450/week. So, we crash the time of this activity by
4 weeks and get the following target times for various paths.
24 24 (28) 22 26
Crash cost = 450 × 4 = Rs. 1800. (2)
iii) Again, 𝑃 is critical path with next activity with lowest cost slope is activity
3 weeks and get the following target times for various paths.
24 24 (25) 22 23
Crash cost = 467 × 3 = Rs. 1401. (3)
iv) Again, 𝑃 is critical path with next activity with lowest crash cost is activity
1 week so that we can make it parallel to the path with next maximum target
time (24) and get the following target times for various paths.

(24) 23 (24) 22 23
Crash cost = 850 × 1 = Rs. 850. (4)

v) Now, we can observe that 𝑃 and 𝑃 becomes critical paths at this stage with
maximum crash time of 24 weeks. So, we prepare a table of crash cost and
crash time for both the paths.
Activities Possible Cost Activities Possible Cost
available for crash slope available for crash slope
crashing on 𝑃 time crashing on 𝑃 time
1-2 1 300 1-5 1 850
2-4 1 200 7-9 3 1667
4-7 2 1250
7-9 3 1667
We need to crash one activity from each path in such a way that crash cost will be
minimum. This can be decided by selecting the choice with minimum crash cost from
following:
Common activity method:
Activity 7-9 is common to both the critical paths. So, if we crash activity 7 -9 by 3
weeks then crash cost = 1667 × 3 = Rs. 5001.
Minimum cost method:
In this method, we select the activity with minimum cost slope from each of the critical
path and then crash both the activities simultaneously. Here,
along path 𝑃 → 2-4 is with minimum cost slope (200)
If we crash these two activities simultaneously by 1 week, then
crash cost = 200 + 850 = Rs. 1050
We can observe that crashing activities 2-4 and 1-5 simultaneously gives least crash
cost, which is Rs. 1050. Hence, crashing the activities 2-4 and 1-5 by one week, we get
the following crashed times:
23 22 23 22 23
In this step, crash cost = Rs. 1050. (5)
vi) Now, 𝑃 , 𝑃 and 𝑃 are critical paths. We will crash activities by 1 week so as
to make these paths parallel to the path with next maximum target time (22
weeks).

Consider the following table for activities along these paths:
Path Activities available Possible Cost slope

for crashing crash time
1-2 1 300
4-7 2 1250
𝑃 7-9 3 1667
𝑃 7-9 3 1667
𝑃 3-6 2 125
7-9 3 1667
Again, we will use common activity method and minimum cost method to decide the
activity to be crashed.
Common activity method:
If we crash common activity 7-9 by 1 week, then crash cost = Rs. 1667
Minimum cost method:
So, if we crash these activities by 1 week then, crash time = Rs. 2092
We can observe that minimum crash cost is obtained by crashing common activity 7 -9
by 1 week. Thus, we have
22 22 22 22 22
In this step, crash cost = Rs. 1667 (6)
vii) Now, all paths are critical with duration of 22 weeks. We want to reduce
target time to 21 weeks. So, we must crash activities by 1 week so that crash
cost will be minimum.
Path Activities available Possible crash Cost slope

for crashing time
1-2 1 300
𝑃 4-7 2 1250
7-9 2 1667
𝑃 6-8 5 960
8-9 1 150
𝑃 7-9 2 1667
3-6 2 125
𝑃 6-8 5 960
8-9 1 150

3-6 2 125
𝑃 7-9 2 1667
If we select the activity 7-9 which is common to 𝑃 , 𝑃 and 𝑃 along with activity 8-9
which is common to 𝑃 and 𝑃 for crashing, then we have the least crash cost. So,
crashing activities 7-9 and 8-9 by 1 week, we have
21 21 21 21 21
Crash cost = Rs. 1667 + Rs. 150 = Rs. 1817. (7)
Thus, we have the following table;
Path Target Time crashed in weeks at various stages

time (9-10) (5-6) (6-7) (1-5) (2-4, (7-9) (7-9,
1 1-5) 8-9)
week 4weeks 3weeks 1 week 1week 1week 1week
𝑃 25 24 24 24 24 23 22 21
𝑃 29 28 24 24 23 22 22 21
𝑃 33 32 28 25 24 23 22 21
𝑃 23 22 22 22 22 22 22 21
𝑃 27 26 26 23 23 23 22 21
Thus, Project duration= 21 weeks
Normal cost of project
= (3000+3600+8000+12800+7500+14400+8400+2400+2600+8600+1100+ 1600)
= Rs. 74,000.
Additional crash cost = 250+1800+1401+850+1050+1667+1817 = Rs. 8835.
The total cost of project with duration of 21 weeks
= Normal cost of project + Additional crash cost
= Rs. 74000+ Rs. 8835 = Rs. 82,835
Problem 03-02-08: The following table gives data on normal time/cost and crash
time/cost for a project.
Activity Normal Crash

Time in weeks Cost in Rs. Time in weeks Cost in Rs.
1-2 3 300 2 400
2-3 3 30 3 30
2-4 7 420 5 580
2-5 9 720 7 810
3-5 5 250 4 300
4-5 0 0 0 0
5-6 6 320 4 410

6-7 4 400 3 470
6-8 13 780 10 900
7-8 10 1000 9 1200
𝑇𝑜𝑡𝑎𝑙 4220
Indirect cost= Rs. 50 per week.
(i) What are the normal project duration and cost?
(ii) Crash the relevant activities systematically and determine the optimal project
completion time and cost.
Solution: The network for the given problem can be drawn as follows:
The different paths and the corresponding durations are;
Path Sequence of activities Duration

𝑃 1→2→3→5→6→7→8 31
𝑃 1→2→3→5→6→8 30
𝑃 1→2→4→5→6→7→8 30
𝑃 1→2→4→5→6→8 29
𝑃 1→2→5→6→8 31
𝑃 1→2→5→6→7→8 32
(i) The duration is maximum for path 𝑃 and hence 𝑃 i.e.1→2→5→6→7→8 is

critical path.
The normal project duration is 3+9+6+4+10=32 weeks.
Direct cost of the project= Rs. 4220
Indirect normal cost= 50×32 = Rs. 1600.
Hence, the normal cost of the project = Rs. 4220 + Rs. 1600 = Rs. 5820.
(ii) Let us calculate cost slopes for the activities on the critical path.
Activities 1-2 2-5 5-6 6-7 7-8

Possible 1 2 2 1 1
crash

Cost 400 − 300 810 − 720 410 − 320 470 − 400 1200 − 1000
Slope 3−2 9−7 6−4 4−3 10 − 9
= 100 = 45 = 45 = 70 = 200
Crashing cost per week less than indirect cost per week is only observed in case
of activities 2-5 and 5-6. Hence, crashing will be economic only if these two
activities are considered for crashing.
Thus, crashing 5-6 by two weeks at a crashing cost of Rs. (45×2) = Rs. 90. The
duration of various paths at this stage becomes
𝑃 𝑃 𝑃 𝑃 𝑃 𝑃
29 28 28 27 29 30
Again, 𝑃 remains critical path. The next activity to be crashed is 2 -5 with a

crash cost of Rs. 45 per week. We will crash this activity by 1 week so as to
make the duration equal to next maximum value (29). Thus, we have
29 28 28 27 28 29
Thus, Crash cost = 45 × 1 = Rs. 45.
Now, at this stage 𝑃 and 𝑃 becomes critical path. If we crash the common
activity 6-7 by 1 week, then we get minimum crash cost = Rs. 70 > Indirect cost.
It means we cannot crash the activities as further crashing will be more
expensive than the saving in indirect cost,
Thus, we have
29 28 28 27 28 29
Therefore, the optimum crash time is 29 weeks.
Total normal cost of project = Rs. 4220.
Total indirect cost = (Indirect cost/week × duration) + total crash cost
= Rs. [ (50 × 29) + (45 + 45 + 45)] = Rs. 1585.
Hence, the optimal cost is given by,
Optimal Cost = Total normal cost + Total indirect cost
= Rs. 4220 + Rs. 1585 = Rs. 5805.
SELF-TEST 04
MCQ 02-08: The process of shortening the duration of a project is known as
a) Crashing

b) Levelling
c) Shortening
d) Smoothing
MCQ 02-09: Which of the following is objective of project crashing?
a) To minimize the normal duration of the project
b) To evolve a strategy of time crashing with minimum possible cost of crashing
c) To spare the resources at an early time for allocation to more profitable jobs
d) All of these

SAQ 03-02-07: Write down the objectives of project crashing.
Solution: The objectives of project crashing are as follows:
(i) To minimize the normal duration of the project.
(ii) To evolve a strategy of time crashing with minimum possible cost of crashing.
(iii) If delayed completion of some activities at any stage introduces the possibility of
enhancing project duration, then time crashing by expediting one or more remaining
activities on the critical path is to be explored.
(iv) To spare the resources at an early time for allocation to more profitable
jobs/activities.
(v) To ensure balanced and uniform requirement of resources.
SAQ 03-02-08: Explain the steps involved in time crashing procedure.
Solution: The following are steps involved in time crashing procedure:
Step 1: Identify the critical path activities and other non -critical path activities which
are likely to become critical during the course of crashing.
Step 2: Calculate the cost slope for the activities identified in step 1.
Step 3: Rank the activities lying on the critical path in ascending order of cost slope.
Step 4: Crash those activities first which have the smallest cost slope to the maximum
possible extent and calculate the new direct cost by cumulative addition of the cost of
crashing to the normal cost.
Step 5: Parallel Crashing: During the course of crashing the activi ties lying on the
critical path, there is always a likelihood that other paths which were non -critical

before crashing may now become critical. Then the project duration can be reduced by
crashing the activities of parallel critical paths simultaneously.
Thus, the lowest cost solution can be obtained by sequentially crashing the
critical path activity and parallel critical path activity for which cost slope is lowest.
SUMMARY
In this unit, we have studied Fulkerson’s Rule for numbering the events in the
network diagram. We had short introduction to project management and its aspects. We
have also studied the Time-cost analysis, various time and cost estimates and crashing
of a project network in detail.

1. Number the events by using Fulkerson’s Rule for the network diagram given
below:
2. Draw network diagram and hence number the events using Fulkerson’s rule for
Activities Predecessor
A -
B A
C A
D A
E C, D
F B, E
G A
H B
I G, H, F
3. Number the events in following network diagram with the help of Fulkerson’s
rule:

4. Explain the concept of critical path with the help of example.
5. For the following data find the optimal project completion time and cost.
Activity Normal Crash

Time in Cost in Rs. Time in Cost in Rs.
days days
1-2 8 100 6 200
1-3 4 150 2 350
2-4 2 50 1 90
2-5 10 100 5 400
3-4 5 100 1 200
4-5 3 80 1 100
Indirect cost = Rs. 70 per day
6. For the project network, the normal and crash durations of the various activities
along with the associated costs are given. Determine the least cost of 36 days
schedule.
Activity Normal duration Normal cost Crash duration Crash cost

1-2 8 1800 6 2200
2-3 16 1500 11 2200
3-5 14 1800 9 2400
1-4 12 2400 9 3000
4-5 15 800 14 2000
5-6 10 2000 8 4000
7. Find the optimal project completion time and cost for the following problem:
Activity Normal Time Crash Time Normal Cost Crash cost

1-2 2 1 300 320
1-3 8 6 400 460
1-4 10 5 500 750
2-5 5 3 100 150
3-5 7 6 250 260
4-6 15 10 700 1000
5-6 6 4 150 230
Indirect cost = Rs. 100 per day.
8. Explain the terms: Direct cost, Indirect cost, Total cost.

KEY WORDS
Fulkerson’s rule, project management, critical path, time -cost trade-off analysis, time
and cost estimates, normal cost, normal time, crash cost, crash time, project normal
cost, direct cost, indirect cost, total cost, cost slope, network crashing, time -cost
optimization.
REFERENCES
BOOKS
& COMPANY LTD

PRAGATI EDITION
MOOCS
YOUTUBE VIDEOS
1) https://youtu.be/vutuWFm6FNI
2) https://youtu.be/Sq7leBP8oNE
3) https://youtu.be/U7kXfNnAjcE
4) https://youtu.be/GkoQ2UK4-j8
5) https://youtu.be/kYgW_OWDZx0
6) https://youtu.be/Ly1x78Or3pY
7) https://youtu.be/bLkkTHpm2Wk
WIKIPEDIA
Drawing network diagram and Fulkerson’s rule:
https://www.yourarticlelibrary.com/industrial-engineering-2/fulkersons-rule-for-
numbering-the-events-with-diagram/90595
OER

UNIT 03-03: CRITICAL PATH METHOD (C.P.M.)
LEARNING OBJECTIVES
 Understand the main objectives and applications of CPM in project management
 Understand the methods to find critical path and total project time
 Understand the various time estimates used to determine critical path
INTRODUCTION
Critical path method is one of the two basic techniques of project manageme nt.
CPM works on basic assumption that duration is inversely proportional to the cost of
resources allotted to the activity. The method of critical path analysis is commonly used
with all forms of project including construction, aerospace and defense, soft ware
development, engineering and plant maintenance, research projects, product
development etc. Any project with interdependent activities can apply this method of
mathematical analysis.
03-01: AN INTRODUCTION TO CPM

CPM is an effective tool for scheduling and planning. A CPM is a route between
two or more operations, which minimize (or maximize) some measure of performance.
This can also be defined as the sequence of activities which will require greatest normal
time to accomplish i.e. in a critical path the sequence of activities with largest durations
are singled out. It is called critical path because any delay in activities lying on this
path would cause a delay in the whole project. A critical path is that sequence of
activities that determines the total project time. To quicken the process the activities
lying on the critical path should be taken first.
The non-critical path activities have some slack or float associated with them
and their extension or delay is not likely to affect the final completion data. The slack
activities should be given priority in order of their float values.
Remark:
The critical path of a project network is the longest path in the network. It can
be identified by simply listing out all the possible paths from the st art node to end node
of the project and then selecting the path with the maximum sum of activity times on
that path.

Main objectives of CPM:
The main objectives of CPM are as follows:
1) To find the obstacles and difficulties involved in a production proces s.
2) To assign time for each operation or activity.
3) To decide about the starting and finishing times of the work.
4) One of the main objectives of network analysis is to identify the critical path so
that resources from outside the system or from the non -critical activities can be
diverted to activities on critical path, if required.
In CPM, initially the network diagram is drawn. Then, after the network diagram
is logically made, the required time to do each operation (activity) is posted
above and to the left of each operation circle. These times are then combined to
develop a schedule which minimize or maximize the measure of performance for
each operation. The schedule thus obtained, can be used to determine the critical
path.
Let us consider a route or path (i, j) defined as a connected set of links

which can be traversed to get from one stated node (i) to another node (j). Then,
the critical path is the path from node (i) to node (j) such that the sum of the
values of the links traversed is minimum (maximum).
Some situations where critical path method can be effectively used are:
a) Location of and deliveries from a warehouse.
b) Production planning.
c) Road systems and traffic schedules
d) Communication networks.
SOLVED PROBLEMS 01
Problem 03-03-01: Find out the critical path in the following network diagram in
which the duration (in weeks) is represented on arrow:
Solution: In the given network diagram, we have the following two paths;

A→B→D→E and A→C→D→E with durations written on the arrow.
For path A→B→D→E; the duration is = 5+2+3= 10
For path A→C→D→E; the duration is = 4+7+3= 14
So, the duration is higher for the second path. Hence, the critical path for the given
network is A→C→D→E with duration of project 14 weeks.
Problem 03-03-02: Find out the critical path in the following network diagram in
which the duration (in weeks) is represented on arrow:
Solution: In the given network diagram, we have the following three paths;
A→D→E→F, A→B→E→F and A→C→E→F.
For path A→D→E→F; the duration is = 7+3+8= 18;
For path A→B→E→F; the duration is = 5+2+8= 15;
For path A→C→E→F; the duration is = 4+7+8= 19.
So, the path with the longest duration is A→C→E→F. Hence, it is the critical path for
given network with duration of project 19 weeks.
SELF-TEST 01
MCQ 03-01: Critical path of a project network is the ________ path in the network.
a) Longest
b) Shortest
c) Simplest
d) Hardest
MCQ 03-02: CPM works on the basic assumption that
a) Duration is proportional to the cost of resources allocated to activity
b) Duration is inversely proportional to the cost of resources allocated to

activity

c) Duration is equal to the cost of resources allocated to activity
d) Duration is not proportional to the cost of resources allocated to activity

SAQ 03-03-01: Write a short note on ‘Critical Path Method’.
Solution: CPM is an effective tool for scheduling and planning. A CPM is a route
between two or more operations, which minimize (or maximize) some measure of
performance. This can also be defined as the sequence of activities which will requi re
greatest normal time to accomplish i.e., in a critical path the sequence of activities with
largest durations are singled out. It is called critical path because any delay in activities
lying on this path would cause a delay in the whole project. A crit ical path is that
sequence of activities that determines the total project time. To quicken the process the
activities lying on the critical path should be taken first.
The non-critical path activities have some slack or float associated with them
and their extension or delay is not likely to affect the final completion data. The slack
activities should be given priority in order of their float values.
SAQ 03-03-02: What are the main objectives of CPM?
Solution: The main objectives of CPM are as follows:
1) To find the obstacles and difficulties involved in a production process.
2) To assign time for each operation or activity.
3) To decide about the starting and finishing times of the work.
4) One of the main objectives of network analysis is to identify the critical path so
that resources from outside the system or from the non -critical activities can be
diverted to activities on critical path, if required.
03-02: LABELLING METHOD FOR DETERMINATION OF CRITICAL PATH

A labelling method consists of building up the shorte st route tree, one link at a
time, working outwards from the origin. A set of links connecting every node in a
network which does not contain any loop is called a tree. A tree has a property that the
path between any two nodes is unique.
The following are the steps in labelling method:
1. Label origin with “distance zero”. Go to step 2.
2. Look for the links whose tail ends are labelled and whose heads are unlabeled.
For each such link, find the sum of the labels at the tail end and its length. Tick

the link for which this sum is minimum (or maximum) and label the head with
this sum. Return to the beginning of step 2.
Terminate the process when (I) The terminal node is labeled (II) All nodes have
been labeled (III) The minimum of the sums formed in step 2 is finit e.
The ticked links will form a shortest tree. The main drawback of labelling method is
that there may be large set of numbers at each stage.
SOLVED PROBLEMS 02
Problem 03-03-03: Find the critical path from A to I by labelling method for the
network given by:
Solution: Here, node ‘A’ is origin node ‘I’ is terminal. Therefore, label node A as zero.
The links with labelled tails and unlabeled heads are those from A are AB and AD and
the sums of their labels are;
Stage 1: AB: 0+9= 9
AD: 0+3= 3
The maximum of these sums is for AB; thus, link AB is ticked and B is labelled with 9.
Similarly, we proceed further from node B and have following stages:
Stage 2: BC: 9+4= 13
BD: 9+2= 11
So, tick BC and label C as 13.
Working in this way, we have the following stages:
Stage 3: CG: 13+5= 18
CD: 13+8= 21
Tick CD, label D as 21.
Stage 4: DE: 21+2= 23
DF: 21+7= 28

Tick DF, label F as 28.
Stage 5: FH: 28+2= 30
Tick FH, label H as 30.
Stage 6: HI: 30+6= 36
Tick HI, label I as 36.
Hence, the critical path is A→B→C→D→F→H→I with distance= 36.
The full working is summarized in following table:
Stage Additional Sum of Link Node Value

No links label ticked labelled of
considered lengths node
1 AB 0+9= 9
AD 0+3= 3 AB B 9
2 BC 9+4= 13
BD 9+2= 11 BC C 13
3 CG 13+5= 18
CD 13+8= 21 CD D 21
4 DE 21+2= 23
DF 21+7= 28 DF F 28
5 FH 28+2= 30 FH H 30
6 HI 30+6= 36 HI I 36
Problem 03-03-04: Find the critical path from A to E by labelling method for the
network given by:

Solution: Here, node A is origin and node E is terminal. Therefore, label node A as
zero.
The links with labelled tails and unlabeled heads are those from A are AB and
AC. Therefore,
Stage 1: AB: 0+7= 7
AC: 0+2= 2
The maximum of these sums is for AB. Thus, link AB is ticked and B is labelled with 7.
Similarly, we proceed further from node B and have the following stages:
Stage 2: BF: 7+4= 11
BD: 7+2= 9
BC: 7+3= 10
So, tick BF and label F as 11.
Stage 3: FD: 11+1= 12
FE: 11+5= 16
So, tick FE and label E as 16.
Hence, the critical path is A→B→F→E with distance 16.
SELF-TEST 02
MCQ 03-03: A set of links connecting every node in a network which does not contain
any loop is called
a) Path
b) Cycle
c) Loop
d) Tree
MCQ 03-04: In the labelling method, we label origin with distance ______.

a) Zero
b) One
c) Equal to project duration
d) None of these

SAQ 03-03-03: Explain ‘Labelling Method’ for determination of critical path with the
help of an example.
Solution: Consider the following network for which we are going to f ind the critical
path from A to E by using labelling method:
Here, node ‘A’ is origin and E is the terminal. So, label node A as zero. The links with
labelled tails and unlabeled heads are those from A i. e. AB, AD and the sum of their
labels are
AB: 0+9=9 AD: 0+3=3
The maximum of these sums is for AB. Thus, tick link AB and B is labelled with 9.
Similarly, we proceed further from node B and have following stages.
Stage 2: BC: 9+4=13 BD: 9+2=11
So, tick link BC and label C as 13.
Stage 3: CD: 13+5=18
Tick CD and label D as 18
Stage 4: DE: 18+2=20
So, tick link DE and label E as 20.
Hence, the critical path in this case is: A→B→C→D→E with distance equal to 20.
SAQ 03-03-04: Describe the steps involved in labelling method of finding critical path.
Solution: The following are the steps in labelling method:
1. Label origin with “distance zero”. Go to step 2.

2. Look for the links whose tail ends are labelled and whose heads are unlabeled.
For each such link, find the sum of the labels at the tail end and its length. Tick
the link for which this sum is minimum (or maximum) and label the head with
this sum. Return to the beginning of step 2.
Terminate the process when i) the terminal node is labeled ii) all nodes have been
labeled iii) the minimum of the sums formed in step 2 is finite.The ticked links will
form a shortest tree.
03-03: METHOD BASED ON TIME ESTIMATES TO FIND CRITICAL PATH

This method is based on the calculation of the following time estimates:
1] Earliest Start Time (EST) of an event:
It is the earliest possible time at which an activity can start. It is calculated from
the earliest start time of tail event. It is conventionally written inside a rectangular
frame put just above the corresponding node.
2] Earliest Finish Time (EFT) of an event:
This is the sum of the early start tail and the time of duration for the activity.
3] Latest Finish Time (LFT) of an event:
It is the latest possible time when an activity should end. It is calculated from
the latest finish time of the head event. For its calculat ion, we require the total project
time.
4] Latest Start Time (LST):
The difference between the latest finish time of head event and the duration time
of the activity is the LST of the activity. It is generally written inside a triangle just
below corresponding node.
5] Total Project Time:
It is the shortest possible time in which the project can be completed .
The procedure for the calculation of various time estimates is illustrated by the
following example:
SOLVED PROBLEMS 03
Problem 03-03-05: Calculate the various time estimates for the following network:

Solution: We use the following steps for calculation of EST of events in network
diagram: -
i] Proceed from left to right in the diagram and assign time zero to 1 st event.
ii] Proceed to the next event and if a number of paths can lead to this event, then
select the time for which EST of tail event plus the duration is maximum.
The activities cannot start until their tail end events are finished and it should be
completed before its head end event starts.
Take EST for event (1) as 0.
For event (2): EST = 0+15= 15time units.
For event (3): we have two paths viz.
1→3 for which 0+15= 15
1→2→3 for which EST (2) + Duration= 15+3 =18
Take maximum of these as EST (3).
Thus, EST (3) = 18time units.
For event (4): EST (4) = EST (3) + Duration= 18+8= 26units.
For event (5): We can approach from two nodes i.e. from event (2) or from event
(4).
Time when approached from event (2) = EST (2) + Duration from (2)
= 15+ 5 = 20
Time when approached from event (4) = EST (4) + Duration from (4)
= 26+ 1 = 27
∴ EST (5) = 27.
For event (6): We can approach from nodes (3), (4) and (5).
Time from (3) = EST (3) + Duration from (3) = 18+12 = 30

Hence, EST (6) = 40.
For event (7): We can approach only from event (6)
∴ EST (7) = EST (6) + Duration from (6)= 40+14 = 54 units
The total duration of the project = 54 units
To calculate LFT for each event in network diagram:
i] Take total duration of project as LFT of last event.
ii] LFT can be calculated by starting from right to left in network diagram and
subtracting the duration times for each possible path from the LFT of head
event.
LFT: -
For event (7): LFT (7) = 54 units
For event (6): event (6) can be approached from event (7) only.
∴ LFT (6) = LFT (7) – Duration from (6) to (7) = 54 - 14 = 40 units
For event (5): This event can be approached from event (6) only.
∴ LFT (5) = LFT (6) – Duration from (5) to (6) = 40 – 3 = 37 units
For event (4): We can approach from (5) or from (6)
Time from (5) = LFT (5) – Duration from (4) to (5) = 37 – 1 = 36
Take minimum of these two as LFT for (4).
∴ LFT (4) = 26
For event (3): can be approached from (4) and (6)
∴ LFT (3) = 18

∴ LFT (2) = 15
∴ LFT (1) = 0.
We can summarize EST and LFT of all events in following table:
Event 1 2 3 4 5 6 7
EST 0 15 18 26 27 40 54
LFT 0 15 18 26 37 40 54
Note that:
EST of any activity = EST of its tail event
LFT of any activity = LFT of its head event
EFT = EST + Duration of activity
LST = LFT – Duration of activity
Therefore, we can have the following table:
Activity Duration Start Times Finish Times

EST LST EFT LFT
1-2 15 0 0 15 15
1-3 15 0 3 15 18
2-3 3 15 15 18 18
2-5 5 15 32 20 37
3-4 8 18 18 26 26
3-6 12 18 28 30 40
4-5 1 26 36 27 37
4-6 14 26 26 40 40
5-6 3 27 37 30 40
6-7 14 40 40 54 54

Problem 03-03-06:Consider the following data and based on that draw network
diagram and prepare a table containing EST and LFT for each event.
Activity Duration (in

Days)
1-2 6
1-3 9
2-4 3
3-4 4
3-5 8
2-6 12
4-6 7
5-6 1
Solution: The network diagram for the given data is as follows:
Table containing EST and LFT for each event is as follows:
Event 1 2 3 4 5 6
EST 0 6 9 13 17 20
LFT 0 8 9 13 19 20
SELF-TEST 03
MCQ 03-05: _______ is the earliest possible time at which an activity can start.
a) EST
b) EFT
c) LST
d) LFT
MCQ 03-06:For a given activity i-j, if EST = 5 and duration = 3, then EFT =
a) 5
b) 3
c) 8
d) 2
MCQ 03-07:For a given activity i-j, if LFT = 8 and duration = 3, then LST =

a) 8
b) 3
c) 11
d) 5

SAQ 03-03-05: Explain the terms: EST, EFT, LFT, LST, Total project time.
Solution:
1] Earliest Start Time (EST) of an event: It is the earliest possible time at which an
activity can start. It is calculated from the earliest start time of tail event. It is
conventionally written inside a rectangular frame put just above the corresponding
node.
2] Earliest Finish Time (EFT) of an event: This is the sum of the early start tail and the
time of duration for the activity.
3] Latest Finish Time (LFT) of an event: It is the latest possible time when an activity
should end. It is calculated from the latest finish time of the head event. For its
calculation, we require the total project time.
4] Latest Start Time (LST): The difference between the latest finish time of head event
and the duration time of the activity is the LST of the activity. It is generally written
inside a triangle just below corresponding node.
5] Total Project Time: It is the shortest possible time in which the project can be
completed.
SAQ 03-03-06:Calculate various time estimates for the following network :
Solution: EST and LFT for the events is given by:
Event 1 2 3 4 5 6 7 8 9 10
EST 0 4 1 5 7 11 15 17 18 25

LFT 0 12 1 13 7 16 15 17 18 25
The various time estimates for the activities are:
Activity Duration EST LST EFT LFT

1-2 4 0 8 4 12
1-3 1 0 0 1 1
2-4 1 4 12 5 13
3-4 1 1 12 2 13
3-5 6 1 1 7 7
4-9 5 5 13 10 18
5-6 4 7 12 11 16
5-7 8 7 7 15 15
6-8 1 11 16 12 17
7-8 2 15 15 17 17
8-9 1 17 17 18 18
8-10 8 17 17 25 25
9-10 7 18 18 25 25
03-04: SLACK AND FLOATS IN NETWORK ANALYSIS

There are many activities where the maximum time available to finish the
activity is more than the time required to complete that activity i.e. duration of that
activity. Total float available is nothing but difference between these two.
SLACK:
Slack means freedom for scheduling or to start the event. It is calculated by

using following formula:
For any event (E),
𝑆𝑙𝑎𝑐𝑘 (𝐸) = 𝐿𝐹𝑇 (𝐸) – 𝐸𝑆𝑇 (𝐸)
Remark:
An event for which the slack time is zero is known as ‘Critical event’. The
critical path can be located by all those events or activities for which slack time is
either zero or it has the least float value. Presence of float time in the project signifies
underutilisation of the resources and also the flexibility inherent in the process for
scheduling.
Types of Floats:
There are three types of floats viz. Total float, Free float and Independent float.
Float Values:
I] Total Float:

This can be calculated for any activity by using following rules:
iii) Subtract EST of tail event from LFT of head event.
iv) Then, subtract the duration time of the activity from value obtained in step
i), to get required total float of the activity.
Thus,
𝑻𝒐𝒕𝒂𝒍 𝑭𝒍𝒐𝒂𝒕
= [(𝑳𝑭𝑻 𝒐𝒇 𝒉𝒆𝒂𝒅 𝒆𝒗𝒆𝒏𝒕 − 𝑬𝑺𝑻 𝒐𝒇 𝒕𝒂𝒊𝒍 𝒆𝒗𝒆𝒏𝒕)
− 𝑫𝒖𝒓𝒂𝒕𝒊𝒐𝒏 𝒐𝒇 𝒂𝒄𝒕𝒊𝒗𝒊𝒕𝒚]
Total float is the extra surplus time which can be allocated to an activity or this
is the period of time up to which an activity can be delayed beyond its earliest finish
time without extending the overall project time.
The value of total float for any activity can help in making following
conclusions:
Total Float Value
Negative Zero Positive
Resources are just

Resources are not sufficient to complete Resources are extra or
adequate & the activity the activity or say the surplus
may not finish in time activity cannot be
delayed
induct extra resources one has the freedom to

or say critical path rellocate the resources
needs crashing in order or the activity can be
to reduce the negative delayed by so much
float period
2] Free Float:
It is that fraction from the total float of a single activity which can be used for
rescheduling that activity without affecting the succeeding activity i.e. all the
succeeding activities will start at their EST. It can be calculated as:
𝐹𝑟𝑒𝑒 𝐹𝑙𝑜𝑎𝑡 = 𝑇𝑜𝑡𝑎𝑙 𝐹𝑙𝑜𝑎𝑡 – 𝑆𝑙𝑎𝑐𝑘 𝑡𝑖𝑚𝑒 𝑜𝑓 𝑡ℎ𝑒 ℎ𝑒𝑎𝑑 𝑒𝑣𝑒𝑛𝑡
3] Independent Float:
The time by which an activity can be rescheduled without affecting the

preceding or the succeeding activities is known as Independent float. It is calculated as:
𝑰𝒏𝒅𝒆𝒑𝒆𝒏𝒅𝒆𝒏𝒕 𝑭𝒍𝒐𝒂𝒕 = 𝑬𝑺𝑻 𝒐𝒇 𝒔𝒖𝒄𝒄𝒆𝒆𝒅𝒊𝒏𝒈 𝒆𝒗𝒆𝒏𝒕 – 𝑬𝑭𝑻 𝒐𝒇 𝒕𝒉𝒆 𝒂𝒄𝒕𝒊𝒗𝒊𝒕𝒚

OR
𝑰𝒏𝒅𝒆𝒑𝒆𝒏𝒅𝒆𝒏𝒕 𝑭𝒍𝒐𝒂𝒕 = 𝑭𝒓𝒆𝒆 𝒇𝒍𝒐𝒂𝒕 – 𝑻𝒂𝒊𝒍 𝒆𝒗𝒆𝒏𝒕’𝒔 𝒔𝒍𝒂𝒄𝒌 𝒕𝒊𝒎𝒆
Remark:
The basic difference between slack and float times is that slack is used for
events only whereas float is applied for activities.
Use of float information in decision making:
I] Independent float can be used in allocating resources elsewhere and increasing the
time of some non-critical activity.
II] Free float can be used without affecting subsequent activities.
III] Total float can affect both previous and subsequent activities.
IV] Negative float signifies reduction (crash) in target time to finish the work in time.
SOLVED PROBLEMS 04
Problem 03-03-07:Calculate the various float values for the activi ties described in the
following network diagram represented along with EST and LFT of each activity.
Solution: The various float time estimates can be calculated in the tabular form with
the help of network diagram given above as follows:
For activity 1-2:
Total Float = [LFT of (2) – EST (1)] – Duration = [15 - 0] – 15 = 15 – 15 = 0
Free Float = Total Float – Slack time of head event i.e. (2)
= Total Float – [LFT (2) – EST (2)]
= 0 – [15 – 15] = 0
Independent Float = Free Float – Slack time of tail event i.e. (1)

= 0 – [LFT (1) – EST (1)]
= 0 – [0 -0] = 0.
On the similar lines, we can calculate for all activities and form the following
table:
Activity Duration Start Finish Float Times

Times Times
EST LST EFT LFT Total Free Independent
1-2 15 0 0 15 15 0 0 0
1-3 1 0 3 15 18 3 3 3
2-3 3 15 15 18 18 0 0 0
2-5 5 15 32 20 37 17 7 7
3-4 8 18 18 26 26 0 0 0
3-6 12 18 28 30 40 10 10 10
4-5 1 26 36 27 37 10 0 0
4-6 14 26 26 40 40 0 0 0
5-6 3 27 37 0 40 10 10 0
6-7 14 40 40 30 54 0 0 0
Problem 03-03-08:Construct a network diagram and find the critical path, EST and
LST for the following data:
Job Description Successor Time in

Activity minutes
S Start A 0
A Processing letter B 4
B Dictating letter D 3
C Order execution E 7
D Supervision E 6
E Dispatching F 2
F Finish None 0
Solution: The network for the given problem can represented as follows:
(We have taken activities on nodes with duration in minutes)

There can be two paths in the problem viz.
Path I: A→B→D→E→F
Path II: A→C→E→F
Along Path I: Total time taken is 4+3+6+2 = 15 mins .
Along Path II: Total time taken is 4+7+2 = 13 mins .
Now, EST (A) = 0
EST (B) = EST (A) + Duration (A) = 0+4 = 4
EST (D) = EST (B) + Duration (B) = 4+3 = 7
EST (E) = EST (D) + Duration (D) = 7+6 = 13
EST (F) = EST (E) + Duration (E) = 13+2 = 15
EST (C) = EST (A) + Duration (A) = 0+4 = 4
Now, we will calculate LFT and hence LST for each activity.
Along Path I: Project time = 15 mins.
Take LFT (F) = 15
⟹ LST (F) = LFT (F) – Duration (F) = 15 – 0 = 15
LFT (E) = LFT (F) – Duration (F) = 15 – 0 = 15
⟹ LST (E) = LFT (E) – Duration (E) = 15 – 2 = 13
LFT (D) = LFT (E) – Duration (E) = 15 – 2 = 13
⟹ LST (D) = LFT (D) – Duration (D) = 13 – 6 = 7
LFT (B) = LFT (D) – Duration (D) = 13 – 6 = 7
⟹ LST (B) = LFT (B) – Duration (B) = 7 – 3 = 4
LFT (A) = LFT (B) – Duration (B) = 7 – 3 = 4
⟹ LST (A) = LFT (A) – Duration (A) = 4 – 4 = 0.
Thus, along Path I we have;
Activity A B D E F
EST 0 4 7 13 15
LST 0 4 7 13 15
LFT 4 7 7 15 15

Slack 4 3 0 2 0
Along Path II:
At C, EST = 4, LST = 6
At E, EST = 13, LST = 13.
Thus, we have the following:
Activity A B C D E F
EST 0 4 4 7 13 15
LFT 4 7 13 7 15 15
LST 0 4 6 7 13 15
Slack 4 3 9 0 2 0
The critical activities are those with either 0 or least slack value.
Thus, A→B→D→E→F is critical path with maximum target time of 15 mins.
Problem 03-03-09: Draw the network diagram and find the critical path for the
following problem:
Activity Immediate Predecessor Duration (in Days)

A - 5
B A 4
C A 3
D A 2
E B 8
F C 7
G D 6
H E, F, G 4
Solution: The network diagram is as given below:
There can be 3 paths in a problem viz.
Path I: A→B→E→H with time = 5+4+8+4 = 21.
Path II: A→C→F→H with time = 5+3+7+4 = 19.
Path III: A→D→G→H with time = 5+2+6+4 = 17.
Now, let us calculate EST and LFT for each of event and then we can have the
table for activities containing required values.

Event 1 2 3 4 5 6 7
EST 0 5 9 8 7 17 21
LFT 0 5 9 10 11 17 21
Activity Duration EST LST EFT LFT Float

=LST-EST
1-2 5 0 0 5 5 0
2-3 4 5 5 9 9 0
2-4 3 5 7 8 10 2
2-5 2 5 9 7 11 4
3-6 8 9 9 17 17 0
4-6 7 8 10 15 17 2
5-6 6 7 11 13 17 4
6-7 4 17 17 21 21 0
As, float for activities 1-2, 2-3, 3-6 and 6-7 is zero; these are critical activities.
Therefore, the critical path is A→B→E→H with project time = 21 Days.
SELF-TEST 04
MCQ 03-08: The _______ total float value represents the sufficient availability of
resources to complete the act.
a) Negative
b) Positive
c) Zero
d) Non-negative
MCQ 03-09: The time by which an activity can be rescheduled without affecting the
preceding or the succeeding activities is known as
a) Independent Float
b) Free Float
c) Total Float
d) Slack Time

SAQ 03-03-07: Explain the terms: Total Float, Free Float, Independent Float for
activities.

Solution: Total Float: Total float is the extra surplus time which can be allocated to an
activity or this is the period of time up to which an activity can be delayed beyond its
earliest finish time without extending the overall project time.
The formula for total float is given by;
𝑇𝑜𝑡𝑎𝑙 𝐹𝑙𝑜𝑎𝑡 = [(𝐿𝐹𝑇 𝑜𝑓 ℎ𝑒𝑎𝑑 𝑒𝑣𝑒𝑛𝑡 − 𝐸𝑆𝑇 𝑜𝑓 𝑡𝑎𝑖𝑙 𝑒𝑣𝑒𝑛𝑡) − 𝐷𝑢𝑟𝑎𝑡𝑖𝑜𝑛 𝑜𝑓 𝑎𝑐𝑡𝑖𝑣𝑖𝑡𝑦]
Free Float: It is that fraction from the total float of a single activity which can be used
for rescheduling that activity without affecting the succeeding activity i.e. all the
succeeding activities will start at their EST. It can be calculated as:
𝐹𝑟𝑒𝑒 𝐹𝑙𝑜𝑎𝑡 = 𝑇𝑜𝑡𝑎𝑙 𝐹𝑙𝑜𝑎𝑡 – 𝑆𝑙𝑎𝑐𝑘 𝑡𝑖𝑚𝑒 𝑜𝑓 𝑡ℎ𝑒 ℎ𝑒𝑎𝑑 𝑒𝑣𝑒𝑛𝑡
Independent Float: The time by which an activity can be rescheduled without affecting
the preceding or the succeeding activities is known as Independent float. It is calculated
as:
𝑰𝒏𝒅𝒆𝒑𝒆𝒏𝒅𝒆𝒏𝒕 𝑭𝒍𝒐𝒂𝒕 = 𝑬𝑺𝑻 𝒐𝒇 𝒔𝒖𝒄𝒄𝒆𝒆𝒅𝒊𝒏𝒈 𝒆𝒗𝒆𝒏𝒕 – 𝑬𝑭𝑻 𝒐𝒇 𝒕𝒉𝒆 𝒂𝒄𝒕𝒊𝒗𝒊𝒕𝒚
OR
𝑰𝒏𝒅𝒆𝒑𝒆𝒏𝒅𝒆𝒏𝒕 𝑭𝒍𝒐𝒂𝒕 = 𝑭𝒓𝒆𝒆 𝒇𝒍𝒐𝒂𝒕 – 𝑻𝒂𝒊𝒍 𝒆𝒗𝒆𝒏𝒕’𝒔 𝒔𝒍𝒂𝒄𝒌 𝒕𝒊𝒎𝒆
SAQ 03-03-08: Write a short note on use of float information in decision making.
Solution: Float information can be used in decision making as follows:
I] Independent float can be used in allocating resources elsewhere and increasing the
time of some non-critical activity.
II] Free float can be used without affecting subsequent activities.
III] Total float can affect both previous and subsequent activities.
IV] Negative float signifies reduction (crash) in target time to finish the work in time.
SUMMARY
In this unit, we have studied CPM in detail along with its main objectives. We
have also learned the methods to determine the critical path in a network and total
project completion time.

1. Find the critical path for the following problem:
Task Predecessors Duration

A - 3

B A 5
C A 4
D B 6
E B, C 7
F D, E 4
2. Find EST, EFT, LST, LFT and total float for each of activity of following
network diagram:
3. For the following data provided
a) Draw the network diagram
b) Identify critical path and find total project duration
c) Determine total, free and independent floats
Activity Duration (in

Days)
0-1 2
1-2 8
1-3 10
2-4 6
2-5 3
3-4 3
3-6 7
4-7 5
5-7 2
6-7 8
4. Find critical path from A to G by labeling method:

5. Given the following network, find EST, LST, EFT and LFT. Also, find the
minimum time required to complete the project as a whole and calculate the
various types of float values.
6. A project has following time schedule:
Activity Time in Activity Time in

months months
1-2 2 4-6 3
1-3 2 5-8 1
1-4 1 6-9 5
2-5 4 7-8 4
3-6 8 8-9 3
3-7 5
Construct network and compute:
(i) Total float for each activity
(ii) Critical path and its duration
7. Calculate EST and LFT for each activity in following network:
8. The following table lists the activities of a maintenance project:
Activity Duration in
months
1-2 2
1-3 2
1-4 1
2-5 4
3-6 5
3-7 8
4-7 3

5-8 1
6-8 4
7-9 5
8-9 3
(i) Draw the project network
(ii) Find critical path and duration of the project.
KEY WORDS
CPM, Critical Path, EST, EFT, LST, LFT, Project time, Float, Slack, Independent float,
Free float, Total Float.
REFERENCES
BOOKS
& COMPANY LTD

PRAGATI EDITION
MOOCS
YOUTUBE VIDEOS
1) https://youtu.be/vUMGvpsb8dc
2) https://youtu.be/qIkL1DmMgt0
3) https://youtu.be/_lMbpyp-NvM
4) https://youtu.be/GwWEWLjcOJI
5) https://youtu.be/KowpQzsoN8Y
6) https://youtu.be/vBFTTFFli4U
WIKIPEDIA
OER

UNIT 03-04: PROJECT EVALUATION AND REVIEW
TECHNIQUE (PERT)
LEARNING OBJECTIVES
 Understand the applications of PERT to various projects having non-

deterministic activities
 Understand the differences between CPM and PERT
 Understand the requirements and practical limitations of PERT
 Understand the algorithm for PERT
INTRODUCTION
During the last few years, various techniques for planning and control of large
projects have been developed specially in defense and industries. Perhaps the best
known of such techniques is PERT. PERT is short form of ‘Project Evaluation and
Review Technique’. The PERT technique deals with projects whose activities are non-
deterministic in nature whereas CPM deals with the projects whose activities are
deterministic in nature.
04-01: PERT
The basic difference between PERT and CPM lies in the emphasis of view point.
CPM gives more emphasis on the operations of project whereas PERT gives more
importance to the event, the beginning or ending of one or more activities.
Time is most essential and basic variable in PERT. It is assumed that there is
always some factor of uncertainty in estimating times (or some other measure of
performance) of any operation which had not been done before. It is generally observed
that the time required to complete any job/activity is non -deterministic due to following
reasons:
a) Human skill required in doing various jobs like welding, painting etc. are
subject to performance variability due to fluctuations in human behavior
and natural fatigue.
b) Uncertainty due to variation in the nature of work. e.g. time required in

doing a particular task depends on the nat ure inputs which may vary
from time to time.

c) Availability of resources varies from time to time e.g. procurement time
of raw material, delivery of machines, manpower availability may vary
from period to period.
d) Variation/changes in environment factors like rain, temperature etc. and

other local conditions may affect the duration of some activity from time
to time.
e) Activities based on research and development process are usually

uncertain in nature.
Due to presence of uncertainty in duration of various activ ities, the deterministic

approach of CPM in many situations becomes unrealistic. This uncertainty aspect of
time factor can be accounted for by using probabilistic approach. Here, the number of
activities may not have a single critical path but have some p robability of being critical.
The PERT technique deals with projects whose activities are non -deterministic
in nature. In PERT, we try to find best estimate of time using some appropriate
statistical method. The expected duration of each activity leads to determination of
expected duration of the project.
Time Estimates in PERT:
In PERT, each activity will have 3-time estimates (values):
1) Optimistic Time (𝒕𝟎 ):
It is the shortest possible time to complete the activity if all goes well.
2) Most Likely Time (𝒕𝒏):
It is the modal value of the activity time distribution or the time which most often is
required if the activity is repeated a number of times.
3) Pessimistic Time (𝒕𝒑 ):
It is the longest time for an activity under adverse conditions. This time is most
difficult to estimate.
Expected duration and variance of activities:
The best estimator of time (or some other measure of performance) connected
with any operation/activity is known as ‘Expected duration of activity’. It is also known
as ‘Mean duration’ for activity.
The expected duration (𝑡 ) for each task can be obtained by using formula:

(𝒕𝟎 + 𝟒𝒕𝒏 + 𝒕𝒑 )
𝒕𝒆 =
𝟔
where,𝑡 = 𝐿𝑒𝑎𝑠𝑡 𝑇𝑖𝑚𝑒 (𝑜𝑝𝑡𝑖𝑚𝑖𝑠𝑡𝑖𝑐)
𝑡 = 𝑀𝑜𝑠𝑡 𝐿𝑖𝑘𝑒𝑙𝑦 𝑇𝑖𝑚𝑒
𝑡 = 𝐺𝑟𝑒𝑎𝑡𝑒𝑠𝑡 𝑇𝑖𝑚𝑒 (𝑃𝑒𝑠𝑠𝑖𝑚𝑖𝑠𝑡𝑖𝑐)
The variance of 𝑡 is given by;
( 𝒕𝒑 − 𝒕𝟎 )
𝑽(𝒕𝒆 ) = 𝝈𝒆 = [ ]
𝟔
where, 𝜎 = 𝑆𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝑑𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛 𝑜𝑓 𝑡 .
Here, it is assumed that time estimate follows ‘Beta Distribution’.
SOLVED PROBLEMS 01
Problem 03-04-01: A mother note that, when her teenage son uses the telephone, he
takes no less than 10 minutes for a call and sometimes as much as one hour. Twenty
minutes calls are more frequent than calls of any other duration. If son’s phone call
were an activity in a PERT project:
(i) What would be the phone call’s expected duration?
(ii) What would be its variance?
(iii) In scheduling the project, how much time would be allocated for
the phone call?
Solution: Here,
Optimistic time 𝑡 = 10 minutes
Most likely time 𝑡 = 20 minutes
Pessimistic time 𝑡 = 60 minutes
( )
(i) Expected duration (𝑡 ) =
( )
⟹ 𝑡 = = 25 minutes.
( )
(ii) 𝑉(𝑡 ) = 𝜎 = [ ]
60 − 10
⟹ 𝑉(𝑡 ) = [ ] = 69.44
6

(iii) In scheduling the project, the time to be allocated to the phone call is 25
minutes.
Problem 03-04-02: Consider the following table summarizing the details of a project:
Activity Predecessor(s) Duration (in weeks)
o m p
A - 5 6 7
B - 1 3 5
C - 1 4 7
D A 1 2 3
E B 1 2 9
F C 1 5 9
G C 2 2 8
H E, F 4 4 10
I D 2 5 8
J H, G 2 2 8
(i) Construct the project network
(ii) Find the expected duration and variance of each activity
(iii) Find the critical path and expected project completion time
Solution:(i) The project network can be drawn as follows:
(ii) Formulae:
Expected duration (time) = Mean duration (𝑡 ) =
Variance (𝜎 ) = [ ]

Activity o m p 𝑡 variance
(𝑚𝑒𝑎𝑛 𝑑𝑢𝑟𝑎𝑡𝑖𝑜𝑛)
A 5 6 7 6 0.11
B 1 3 5 3 0.44
C 1 4 7 4 1.00
D 1 2 3 2 0.11
E 1 2 9 3 1.78
F 1 5 9 5 1.78
G 2 2 8 3 1.00
H 4 4 10 5 1.00
I 2 5 8 5 1.00
J 2 2 8 3 1.00
(iii) To find critical path, we need time duration for each activity. We will take
mean duration as duration for activity. We also need to determine EST, LFT.
For each activity i-j, the formula to obtain EST is:
𝐸𝑆 = max(𝐸𝑆 + 𝐷 )
𝑤ℎ𝑒𝑟𝑒, 𝐸𝑆 = 𝐸𝑆𝑇 𝑜𝑓 𝑗𝑡ℎ 𝑒𝑣𝑒𝑛𝑡 𝑎𝑛𝑑 𝐷 = 𝑑𝑢𝑟𝑎𝑡𝑖𝑜𝑛 𝑓𝑟𝑜𝑚 𝑖𝑡ℎ 𝑡𝑜 𝑗𝑡ℎ 𝑒𝑣𝑒𝑛𝑡)
𝑇𝑎𝑘𝑒 𝐸𝑆𝑇 𝑓𝑜𝑟 𝑖𝑛𝑖𝑡𝑖𝑎𝑙 𝑒𝑣𝑒𝑛𝑡 = 0
For each activity i-j, the formula to obtain LFT of i th node is:
𝑇𝑎𝑘𝑒 𝐿𝐹𝑇 𝑜𝑓 𝑙𝑎𝑠𝑡 𝑛𝑜𝑑𝑒 𝑎𝑠 𝑐𝑜𝑚𝑝𝑙𝑒𝑡𝑖𝑜𝑛 𝑡𝑖𝑚𝑒.
𝐿𝐹𝑇 𝑜𝑓 𝑖𝑡ℎ 𝑛𝑜𝑑𝑒 = min(𝐿𝐹𝑇 𝑜𝑓 𝑗𝑡ℎ 𝑛𝑜𝑑𝑒 − 𝐷 )
Critical path is longest path in network for which following conditions should satisfy;
1) 𝐸𝑆 = 𝐿𝐹𝑇 (Starting node)
2) 𝐸𝑆 = 𝐿𝐹𝑇 (Ending node)
3) 𝐸𝑆 − 𝐸𝑆 = 𝐿𝐹𝑇 − 𝐿𝐹𝑇 = 𝐷
For node (1), (4), (6), (7) and (8), EST = LFT and 3 rd condition is also satisfied for each
activity1-4, 4-6, 6-7 and 7-8.

Thus, critical path is 1→4→6→7→8.
The expected completion time = sum of durations of activities on critical path
= 4+5+5+3 = 17 weeks
Therefore, the expected project completion time is 17 weeks.
SELF-TEST 01
MCQ 04-01: _______ is the longest time for an activity under adverse conditions.
a) Optimistic time
b) Most likely time
c) Pessimistic time
d) Expected time
MCQ 04-02: _______ is the shortest possible time to complete the activity if all goes
well.
a) Optimistic time
b) Most likely time
c) Pessimistic time
d) Expected time
MCQ 04-03: In PERT, each activity will have ______ time estimates.
a) One
b) Two
c) Three
d) Four

SAQ 03-04-01: Write down any four reasons due to which the time required to
complete any job/activity is non-deterministic.
Solution: The time required to complete any job/activity is non-deterministic due to

following reasons:
i) Human skill required in doing various jobs like welding, painting etc. are
subject to performance variability due to fluctuations in human behavior and
natural fatigue.

ii) Uncertainty due to variation in the nature of work. e.g. time required in
doing a particular task depends on the nature inputs which may vary from
time to time.
iii) Availability of resources varies from time to time e.g. procurement time of
raw material, delivery of machines, manpower availability may vary from
period to period.
iv) Variation/changes in environment factors like rain, temperature etc. and

other local conditions may affect the duration of some activity from time to
time.
SAQ 03-04-02: What are the time estimates in PERT? Explain each of them.
Solution: The following are the time estimates in PERT:
1) Optimistic Time (𝒕𝟎 ):
It is the shortest possible time to complete the activity if all goes well.
2) Most Likely Time (𝒕𝒏):
It is the modal value of the activity time distribution or the time which most often is
required if the activity is repeated a number of times.
3) Pessimistic Time (𝒕𝒑 ):
It is the longest time for an activity under adverse conditions. This time is most
difficult to estimate.
04-02: PERT ANALYSIS

In PERT, for each activity we replace the three time estimates by the expected
duration of the activity. Using these time estimates we do the forward and backward
pass calculations to find the critical path.
In PERT analysis,
1. The duration of the critical path based on the expected duration of each activity
gives the expected duration of the project.
2. The variances of the activities lying on the critical path are summed to find the
variance of the project duration.
Algorithm for PERT:
Step 1: Prepare suitable network for the given problem.
Step 2: Find the various time estimates and label them on the network.

Step 3: Determine the critical path for the network obtained in step 2.
Step 4: Find the total expected project duration ( 𝑇̅ ) by adding the expected time
estimates obtained by using the formula
(𝑡 + 4𝑡 + 𝑡 )
𝑡 =
6
For various activities lying on the critical path.
Step 5: Find the variance of the estimates ( 𝜎 ) by using the formula
(𝑡 −𝑡 )
𝑉(𝑡 ) = 𝜎 = [ ]
6
And add the variances of the activities lying on the critical path.
Step 6: Using the results of steps 4 and 5, calculate the standard normal variate
𝐷 − 𝑇̅
𝑧=
√𝜎
𝑤ℎ𝑒𝑟𝑒, 𝐷 = 𝑃𝑟𝑜𝑗𝑒𝑐𝑡 𝐷𝑢𝑟𝑎𝑡𝑖𝑜𝑛
Step 7: Find the probability of finishing the project on some fixed target by using
the tables of normal distribution and the value of z calculated in step 6.
SOLVED PROBLEMS 02
Problem 03-04-03: The following table contains the data for the activities on critical
path. Find the probability of completing the project on or before 22 weeks.
Activity Expected Variance

Duration
C 4 1.00
F 5 1.78
H 5 1.00
J 3 1.00
Solution: We require, P(𝑥 ≤ 22).
𝜇 = 𝑚𝑒𝑎𝑛 𝑑𝑢𝑟𝑎𝑡𝑖𝑜𝑛 𝑓𝑜𝑟 𝑐𝑟𝑖𝑡𝑖𝑐𝑎𝑙 𝑝𝑎𝑡ℎ = 4 + 5 + 5 + 3 = 17
𝜎 = 𝑚𝑒𝑎𝑛 𝑣𝑎𝑟𝑖𝑎𝑛𝑐𝑒 = 1.00 + 1.78 + 1.00 + 1.00 = 4.78
Thus, 𝜎 = 𝑚𝑒𝑎𝑛 𝑠𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝑑𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛 = √4.78 = 2.19 weeks
Thus, P (𝑥 ≤ 22) = 𝑃 [ ≤ .
] = 𝑃 [𝑧 ≤ 2.28] = 0.9887
(This probability can be obtained from standard normal distribution table)

Therefore, the probability of completing the project on or before 22 weeks is
0.9887 i.e. 98.87%.
Problem 03-04-04: For the given activities determine:
(1) Critical path using PERT
(2) Calculate variance and standard deviation
(3) Calculate probability of completing project in 26 days
Activity 𝑡 𝑡 𝑡
1-2 6 9 12
1-3 3 4 11
2-4 2 5 14
3-4 4 6 8
3-5 1 1.5 5
2-6 5 6 7
4-6 7 8 15
5-6 1 2 5
Solution:
Activity 𝑡 Variance
(𝑡 + 4𝑡 + 𝑡 ) ( )
= =[ ]
6
1-2 9 1
1-3 5 1.778
2-4 6 4
3-4 6 0.444
3-5 2 0.444
2-6 6 0.111
4-6 9 1.778
5-6 2 0.111
The network diagram containing LFT and EST along with the critical path
represented with the help of double lines is as follows:
Here, 𝑡 = 26 𝑑𝑎𝑦𝑠 𝑎𝑛𝑑𝑡 = 24 𝑑𝑎𝑦𝑠 (𝑎𝑙𝑜𝑛𝑔 𝑐𝑟𝑖𝑡𝑖𝑐𝑎𝑙 𝑝𝑎𝑡ℎ)

𝑡− 𝑡 26 − 24
𝑧= = = 0.7682
𝜎 2.6034
Therefore, the probability of completing project in 26 days is 0.7682.
SELF-TEST 02
MCQ 04-04: If for an activity, 𝑡 = 6, 𝑡 = 2, 𝑡 = 4 then, the expected duration 𝑡 is
a) 3
b) 6
c) 2
d) 0
MCQ 04-05: If for activities on critical path in PERT network, the variance s are 1, 4, 4,
then standard deviation is equal to
a) 6
b) 3
c) 9
d) 5

SAQ 03-04-03: Write an algorithm for PERT.
Solution: Step 1: Prepare suitable network for the given problem.
Step 2: Find the various time estimates and label them on the network.
Step 3: Determine the critical path for the network obtained in step 2.
Step 4: Find the total expected project duration ( 𝑇̅ ) by adding the expected time
estimates obtained by using the formula
(𝑡 + 4𝑡 + 𝑡 )
𝑡 =
6
For various activities lying on the critical path.
Step 5: Find the variance of the estimates ( 𝜎 ) by using the formula
(𝑡 −𝑡 )
𝑉(𝑡 ) = 𝜎 = [ ]
6
And add the variances of the activities lying on the critical path.
Step 6: Using the results of steps 4 and 5, calculate the standard normal variate

𝐷 − 𝑇̅
𝑧=
√𝜎
𝑤ℎ𝑒𝑟𝑒, 𝐷 = 𝑃𝑟𝑜𝑗𝑒𝑐𝑡 𝐷𝑢𝑟𝑎𝑡𝑖𝑜𝑛
Step 7: Find the probability of finishing the project on some fixed target by using the
tables of normal distribution and the value of z calculated in step 6.
SAQ 03-04-04: Write a short note on ‘PERT Analysis’.
Solution: In PERT, for each activity we replace the three time estimates viz. optimistic
time, pessimistic time and most likely time by the expected duration of the activity.
Using these time estimates we do the forward and backward pass calculations to find
the critical path.
In PERT analysis,
1. The duration of the critical path based on the expected duration of each activity
gives the expected duration of the project.
2. The variances of the activities lying on the critical path are summed to find the
variance of the project duration.
04-03: REQUIREMENTS AND PRACTICAL LIMITATIONS OF PERT

Requirements for the application of PERT technique:
1. All the individual tasks necessary to complete a given project must be put down
in a network. Events are shown by circles and activities are designed by arrows
leading from event to its successor events.
2. Events and activities must be sequenced on the network under a highly logical
set of rules.
3. Time estimates are also made for each activity of th e network on three-way
basis, namely optimistic, most likely and pessimistic times.
4. Critical path and slack time are computed. A critical sequence is one which
gives greater expected time to accomplish the work.
Practical limitations in using PERT:
1. The assumption of beta distribution may not be true always.
2. The expected time and the corresponding variance are only estimated values and
give the approximate values.

3. Time estimates have subjectiveness and to that extent the techniques could be
weak.
4. In computation of standard deviation of the critical path, independence of

activities is assumed. Limitations of resources may invalidate the independence.
5. It may not be always possible to sort out completely identifiable activities and
to pin point their starting and finishing points.
6. Cost-time trade-offs for deriving cost curve slopes are subjective and call for a
great deal of expertise of the technology as well as genuine effort to estimate.
SOLVED PROBLEMS 03
Problem 03-04-05: For the project
Task A B C D E F G H I J K
Least Time 4 5 8 2 4 6 8 5 3 5 6
Greatest time 8 10 12 7 10 15 16 9 7 11 13
Most likely 5 7 11 3 7 9 12 6 5 8 3
time
Find the earliest and latest expected times to each event and also the critical path in
the network.
Solution: The expected time (𝑡 ) for each task can be obtained by using formula
(𝑡 + 4𝑡 + 𝑡 )
𝑡 =
6
Where, 𝑡 = 𝑙𝑒𝑎𝑠𝑡 𝑡𝑖𝑚𝑒
𝑡 = 𝐺𝑟𝑒𝑎𝑡𝑒𝑠𝑡 𝑡𝑖𝑚𝑒
𝑡 = 𝑀𝑜𝑠𝑡 𝑙𝑖𝑘𝑒𝑙𝑦 𝑡𝑖𝑚𝑒
We will use 𝑡 as duration for each task and then we can solve the problem.
Task 𝑡 𝑡 𝑡 𝑡
A 4 8 5 5.3
B 5 10 7 7.2
C 8 12 11 10.8

D 2 7 3 3.3
E 4 10 7 7.0
F 6 15 9 9.6
G 8 16 12 12.05
H 5 9 6 6.3
I 3 7 5 5.0
J 5 11 8 8.0
K 6 13 3 5.2
This expected time can be used to find EST and LFT for each event.
Activity Task Expected EST LFT

duration
1-2 A 5.3 0 5.3
1-3 B 7.2 0 16.1
2-3 C 10.8 5.3 16.1
1-4 D 3.3 0 13.5
3-5 E 7.0 16.1 23.1
4-5 F 9.6 3.3 23.1
4-6 G 12.0 3.3 28.1
5-7 H 6.3 23.1 30.9
5-6 I 5.0 23.1 28.1
6-8 J 8.0 28.1 36.1
7-8 K 5.2 29.4 36.1
The critical path is 1→2→3→5→6→8 (for each node on critical path
EST = LFT) and the project completion time is 36.1 units.
Event 1 2 3 4 5 6 7 8
EST 0 5.3 16.1 3.3 23.1 28.1 29.4 36.1
LFT 0 5.3 16.1 13.5 23.1 28.1 30.9 36.1
Problem 03-04-06: For the following data;
(i) Draw the network diagram.

(ii) Calculate the expected duration and variance of duration for each activity.
Duration in weeks
1-2 1 5 3
2-3 1 7 4
2-4 1 5 3
2-5 5 11 8
3-6 2 6 4
4-6 5 7 6
5-7 4 6 5
6-7 1 5 3
Solution:
(i) Network Diagram:
(ii) Table containing expected duration and variance for each activity is as
follows:
We will use the following formulae:

( )
Expected duration: 𝑡 =
( )
Variance: 𝜎 = [ ]
Duration in 𝑡 𝜎
weeks
1-2 1 5 3 3 0.4444
2-3 1 7 4 4 1
2-4 1 5 3 3 0.4444
2-5 5 11 8 8 1
3-6 2 6 4 4 0.4444
4-6 5 7 6 6 0.1111
5-7 4 6 5 5 0.1111
6-7 1 5 3 3 0.4444
SELF-TEST 03
MCQ 04-06: While using PERT, time estimates are made for each activity of the
network on basis of

a) Optimistic time
b) Most likely time
c) Pessimistic time
d) All of these
MCQ 04-07: Which of the following is limitation in using PERT?
a) Time estimates have subjectiveness
b) Events and activities are sequenced under highly logical set of rules
c) Critical path and slack times are computed
d) None of these

SAQ 03-04-05: Write down requirements for the application of PERT.
Solution: The following are the requirements of application of PERT:
1. All the individual tasks necessary to complete a given project must be put down
in a network. Events are shown by circles and activities are designed by arrows
leading from event to its successor events.
2. Events and activities must be sequenced on the network under a highly logical
set of rules.
3. Time estimates are also made for each activity of the network on three -way
basis, namely optimistic, most likely and pessimistic times.
4. Critical path and slack time are computed. A critical sequence is one which
gives greater expected time to accomplish the work.
SAQ 03-04-06: Write a note on practical limitations in using PERT.
Solution: The following are practical limitations in using PERT:
1. The assumption of beta distribution may not be true always.
2. The expected time and the corresponding variance are only estimated values and
give the approximate values.
3. Time estimates have subjectiveness and to that extent the techniques could be
weak.
4. In computation of standard deviation of the critical path, independence of

activities is assumed. Limitations of resources may invalidate the independence.

5. It may not be always possible to sort out completely identifiable activities and
to pin point their starting and finishing points.
6. Cost-time trade-offs for deriving cost curve slopes are subjective and call for a
great deal of expertise of the technology as well as genuine effort to estimate.
04-04: COMPARISON OF CPM AND PERT

The following are some of the distinguishing poi nts in CPM and PERT:
1. CPM is activity oriented whereas PERT is event oriented. Here, an activity

represents part of the work to be completed over a period of time. Event is a
point of time showing starting and finishing of the work.
2. CPM stresses more on cost considerations whereas PERT is concerned

essentially with time factor.
3. There are three time estimates in PERT for each of the activities whereas in
CPM there is a single time value.
4. CPM is good for repetitive projects whereas PERT is appropriate for n on-
repetitive projects.
5. PERT can be analyzed statistically whereas in the case of CPM statistical

analysis is not possible.
Difficulties in using Network Methods:
1. Difficulties are faced while securing the realistic time estimates. When we
consider new and non-repetitive kind of projects, the times estimates produced
are often by guess.
2. Developing a clear logical network is also a tedious task. It totally depends upon
the input data and thus plan can be no better than the personnel who provides
the data.
3. The planning and implementation of networks requires a trained person in the

field of network methodology.
4. Determination of level of network detail is another cumbersome area. The level

of detail varies from planner to planner and depends upon the judgement and
experience.
5. The natural tendency to oppose the changes results in the difficulty of

persuading the management to accept these techniques.
Applications of Network Techniques:

PERT and CPM have applications in various fields. Following are some of the
typical areas in which these techniques are widely used:
1. Construction Industry: It is one of the largest areas in which the network

techniques of project management have found application.
2. Manufacturing: The design, development and testing of new machines, installing

machines and plant layouts are some of the examples of how it can be applied to
the manufacturing function of a firm.
3. Maintenance Planning: Shutdown and maintenance of power plants, chemical

plants, steel furnaces and overhauling of large machin es have been carried out
by using PERT.
4. Research and Development: R and D has been the most extensive area where
PERT has been used for development of new products, processes and systems.
5. Administration: Networks have been used by the administration for st reamlining

paperwork system, for making major administrative system revisions, for long
range planning and developing staffing plans, etc.
SOLVED PROBLEMS 04
Problem 03-04-07: Write down any three difficulties in using network methods.
Solution: Following are some of the difficulties faced while using the network
methods:
(1) Developing a clear logical network is also a tedious task. It totally

depends upon the input data and thus plan can be no better than the
personnel who provides the data.
(2) The planning and implementation of networks requires a trained person

in the field of network methodology.
(3) Determination of level of network detail is another cumbersome area.

The level of detail varies from planner to planner and depends upon the
judgement and experience.
Problem 03-04-08: How one can apply network techniques in the field of marketing
and inventory planning?
Solution: CPM and PERT are the two network techniques having wide variety of
applications. Marketing and Inventory planning also involves the use of network
techniques.

Marketing: Networks have been used for advertising programmes, for developing and
launching of new projects ad for planning their distribution.
Inventory planning: Installation of production and inventory control, acquisition of

spare parts, etc. have been greatly helped by network techniques.
SELF-TEST 04
MCQ 04-08: ______ is the most essential and basic variable in PERT.
a) Event
b) Time
c) Activity
d) Path
MCQ 04-09: Which of the following s not true for PERT?
a) Can be analyzed statistically
b) Appropriate for non-repetitive projects
c) Focused on cost considerations
d) Concerned essentially with time factor

SAQ 03-04-07: Write a short note on “comparison of CPM and PERT”.
Solution: The following are some of the distinguishing points in CPM and PERT:
1. CPM is activity oriented whereas PERT is event oriented. Here, an activity

represents part of the work to be completed over a period of time. Event is a
point of time showing starting and finishing of the work.
2. CPM stresses more on cost considerations whereas PERT is concerned

essentially with time factor.
3. There are three time estimates in PERT for e ach of the activities whereas in
CPM there is a single time value.
4. CPM is good for repetitive projects whereas PERT is appropriate for non -
repetitive projects.
5. PERT can be analyzed statistically whereas in the case of CPM statistical

analysis is not possible.

SAQ 03-04-08: Write down any four applications of network techniques.
Solution: PERT and CPM have applications in various fields. Following are some of
the typical areas in which these techniques are widely used:
1. Construction Industry: It is one of the largest areas in which the network

techniques of project management have found application.
2. Manufacturing: The design, development and testing of new machines, installing

machines and plant layouts are some of the examples of how it can be applied to
the manufacturing function of a firm.
3. Maintenance Planning: Shutdown and maintenance of power plants, chemical

plants, steel furnaces and overhauling of large machines have been carried out
by using PERT.
4. Research and Development: R and D has been the most extensive area where
PERT has been used for development of new products, processes and systems .
SUMMARY
In this unit, we have studied PERT technique to deal with projects whose
activities are non-deterministic in nature. We have studied the comparison of the CPM
and PERT along with difficulties while using these network techniques. We have seen
the areas where these network techniques can be applied. We have also learned the
requirements and limitations while applying PERT.

1. A project is represented by the network shown below and has the following data:
Task A B C D E F G H I
Least Time 5 18 26 16 15 6 7 7 3
Greatest Time 10 22 40 20 25 12 12 9 5
Most likely time 8 20 33 18 20 9 10 8 4
Determine the following:
(a) Expected task time and their variance.

(b) EST and LFT to each node.
(c) Critical path and project completion time.
2. Write a short note on time estimates in probabilistic PERT networks.
3. A small project consisting of eight activities has the following time estimates in
weeks:
Time
estimates in
Activity Predecessor weeks
𝑡 𝑡 𝑡
A - 2 4 12
B - 10 12 26
C A 8 9 10
D A 10 15 20
E A 7 7.5 11
F B, C 9 9 9
G D 3 3.5 7
H E, F, G 5 5 5
(a) Draw PERT network diagram.
(b) Determine expected time and variance for each of the activity.
4. For the data given below, find the following:
(i) The expected task times and their variances
(ii) EST and LFT for each event
(iii) The critical path
Task A B C D E F G H I J K L
𝑡 3 1 2 6 8 0 5 6 1 3 8 2
𝑡 5 2 4 8 12 0 7 9 2 6 15 4
𝑡 6 3 6 12 17 0 9 12 3 8 20 6
Precedence relationship: -
C, D > A; E > B, C; F, H > E; G > D, F; J > G; I, K > H; L > J, I.
5. A small project is composed of 7 activities whose time estimates are listed in the
table below:

Activity 1-2 1-3 1-4 2-5 3-5 4-6 5-6
Optimistic time 1 1 2 1 2 2 3
Most likely time 1 4 2 1 5 5 6
Pessimistic time 7 7 8 1 14 8 15
(a) Draw the project network.
(b) Find the expected duration and variance for each activity.
6. Draw the network diagram for the following data:
Optimistic Most likely Pessimistic

Activity
time time time
1-2 1 5 3
2-3 1 7 4
2-4 1 5 3
2-5 5 11 8
3-6 2 6 4
4-6 5 7 6
5-7 4 6 5
6-7 1 5 3
7. Explain the PERT method to solve the networks with the help of an example.
8. Write a short note on requirements and practical limitations while applying

PERT.
KEY WORDS
PERT, Optimistic time, Most likely time, Pessimistic time, PERT analysis, expected
duration, variance
REFERENCES
BOOKS
& COMPANY LTD

PRAGATI EDITION
MOOCS
YOUTUBE VIDEOS
1) https://youtu.be/WrAf6zdteXI

2) https://youtu.be/J1WwNKDdDC0
3) https://youtu.be/N_3wIn1mMKI
WIKIPEDIA
https://www.engineeringenotes.com/networking/network-analysis/top-4-problems-
on-pert-network-analysis-networking/15725
OER

C REDIT 04

UNIT 04-01: SIMULATIONS AND IT’S THEORY
LEARNING OBJECTIVES
 Understand what simulation is and how it can be used.
 Understand the advantages of simulation technique.
 Understand the limitations of simulation.
 Learn the applications of simulation.
INTRODUCTION
Simulation is one of the widely used techniques by corporate managers as an aid
for decision-making. Designers and analysts have long used the techniques of
simulation by physical sciences. Simulation is the representative model of real
situation. For example, in a city, a children's park with various signals and crossing is a
simulated model of city traffic. There are many real-world problems which cannot be
represented by a mathematical model due to stochastic nature of the problem, the
complexity in problem formulation and many values of the variables are not known in
advance and there is no easy way to find these values. Hence, simulation promises to
become an important tool for tackling this problem. Simulation determines the effect of
a number of alternate policies without disturbing the real system.
John Von Newmann and Stainslaw Ulam made first important application of
simulation for determining the complicated behavior of neutrons in a nuclear shielding
problem, being too complicated for mathematical analysis. After the remarkable success
of this technique on neutron problem, it has become more popular and has many
applications in business and industry. The development of digital computers has further
increased the rapid progress in the simulation technique.
01-01: INTRODUCTION TO SIMULATION

A) When to use simulation?
Operation research has wide scope for solving various types of ma nagerial
decision-making problems. Some of the O.R. technique like linear programming,
network models we discussed in foregoing chapter. There are also other O .R. techniques
like dynamic programming, queuing theory, etc. are not sufficient to solve O.R.
problems. Each technique has its own limitations. We discuss some of the m.
Limitation of L.P. model: It has following limitations.

i. It is one time decision process and assumes average value for the decision
variable. If the planning horizon is long, like 15 years, L.P. problem may deal
with the yearly average data, but it will not consider the variation over the
months to months and weeks to weeks.
ii. L.P. model assumes the data to be known with certainty. In many real situations
(L.P. model) there are uncertainties which can be ignored.
Limitation of dynamic model: It tackles with the two main important limitations of
L.P. model like uncertainty and can analyze multi-period planning problem. But it has
following limitations
i) Dynamic programming model can be used to tackle very simple situation

involving only few variables.
ii) If the number of few variables is bit larger, the computation task becomes quite
complex and involved.
Similarly, there are limitations to the network model, queuing model , etc. Hence, we
need one technique which covers all the limitations of models like L .P. model, dynamic
model, dynamic stochastic model, etc. There are many techniques to solve the O.R.
models but best so far is simulation.
B) What is simulation?
Simulation is an imitation of reality. Some of the simulation model examples as below :
 In laboratories, a series of experiments are carried out on simulated models to

determine the behavior of the real system in real environments.
 Children's bike park with various crossing and traffic signals is a simulated
model of a city traffic system.
 Testing an airplane model in a wind tunnel, on the basis of which we determine

the performance of a real aircraft in real operating conditions.
 The planetarium shows present a beautiful simulat ion of a planetary system.
 Other examples of simulation are the environment in a geological garden and a

natural history museum.
All of the above examples have tried to imitate the reality to see what might
happen under real working conditions. This imitat ion of reality, which can take the
form of physical or mathematical equations, can be called simulations.

Simulation is an operations research techni que used to represent real-life problems
through easy-to-manipulate numbers and mathematical symbols. For e xample, chess
simulation battle games, backgammon simulation racing games, and other games that
simulate hunting and diplomacy have been invented. Decision makers use simulation to
gain the ability to conduct situational experiments under controlled condit ions.
Few definitions of simulation are stated below:
 Simulation is the process of designing a model of a real system and

conducting experiments with this model for the purpose of understanding
the behavior (within the limits imposed by a criterion or set of criteria) for
the operation of the system. — Shannon
 Simulation is the use of a system model that has the designed

characteristics of reality in order to produce the essence of actual
operation. — Churchman
 A simulation of a system or an organism is the operation of a model or

simulator which is a representation of the system or organism. The model
is amenable to manipulation which would be impossible, too expensive or
unpractical to perform on the entity it portrays. The operation of the model
can be studied and for it, properties concerning the behavior of the actual
system can be inferred. — Shubik
 Simulation is a numerical technique for conducting experiments on a

digital computer, which involves certain types of mathematical and logic al
relationships necessary to describe the behavior and structure of a
complex real-world system over extended periods of time.
— Naylor et al.
 ‘X simulated Y’ is true if and only if
(i) X and Y are formal systems,
(ii) Y is taken to be the real system,
(iii) X is taken to be an approximation to the real system and
(iv) The rules of validity in X are non -error-free, otherwise X will become
the real system.
These definitions point out that simulation can be equally applied to military
war games, business games, economic models, etc. Simulation also involves logical and

mathematical modeling, which involves the use of computers to test the behavior of the
system in real-life situations using iteration or continuous experimentation to ada pt.
The highlights what simulation is and what it is not in the below table :
It is It is not
1) A technique which uses computers. 1) An analytical technique which

provides exact solution.
2) An approach for reproducing the
processes by which events of chance 2) A programming language (but it
and change are created in a computer. can be program aimed into a set
of commands that can form
3) A procedure for testing and
alanguage to facilitate the
experimenting on models to answer
programming ofsimulation).
what if . . ., then so and so . . . types of
questions.
In general, the simulation technique is a dependable tool in situation where

mathematical analysis is either too complex to too costly.
SOLVED PROBLEMS 01
Problem 04-01-01: Give some examples to illustrate the scope of application of
simulation.
Solution: The following few examples illustrate scope of applications of simulation.
1) Aircraft pilots or Astronauts are trained in a simulator to expose them with various
problems that they are likely to face in the sky while flying real aircraft.
2) Computer designers simulate with a computer system configuration (its speed, size,
computing qualities, memory and so on) in terms of costs of such configurations
and resulting computational service that it provides to users.
3) Queuing system decision-makers simulate the effect of probabilistic nature of

arrival rate of customers and the service rate of the server to serve the customer on
the cost of waiting against the cost of idle time of service facilities in the queuing
system.
4) Hospital management may simulate alternative scheduling rules of the ambulances,

their locations, the response time to an emergency call and, of course, the overall
service quality and the costs incurred if the ambulances were to be configured in a
certain way (in types, number, location, scheduling and staffing).
Problem 04-01-02: When simulation should be used?

Solution: Simulation technique will be used if
 Actual observation of a system may be too expensive.
 The problem is too big or intricate to handle with linear, dynamic and standard
probabilistic models.
 The standard sensitivity analysis is too clumsy and computationally burdensome for
observing the actual environment.
 It is not possible to develop a mathematical model. Even though a mathematical

model can be formulated, a straight forward analytical solution may not be available.
 It is not possible to perform validating experiments on mathematical models

describing the system.
 There may not be sufficient time to allow the system to operate extensively.
SELF-TEST 01
MCQ 01-01: The purpose of using simulation technique is to
a) imitate a real-world situation,
b) understand properties and operating characteristics of complex real-life

problems,
c) reduce the cost of experiment on a model of real situation,
MCQ 01-02: Simulation is defined as
a) a technique that uses computers.
b) an approach for reproducing the processes by which events by chance and

changes are created in a computer.
c) a procedure for testing and experimenting on models to answer what if . . .,

then so and so . . . types of questions.

SAQ 04-01-01: Define simulations and highlights when simulation is appropriate tool?
Solution: It is the imitation of the operation of a real-world process or system over

time. It is acting out or mimicking an actual or probable real-life condition, event or
situation to find a cause of a past occurrence or to forecast future effects of assumed
circumstances or factors.

It is used to describe and analyze the behavior of a system.
For example flight simulator,
When simulation is an appropriate tool:
1. It enables one to study internal interactions of a complex system or of a

subsystem within a complex system.
2. The effect of information, organization and environmental changes on the

model's behavior can be simulated and observed.
3. The performance of the system under investigation can be improved by

knowledge gained from simulations.
4. Simulation results can help to find which variables are the most im portant ones
and how variables interact.
5. It can be used as a learning tool.
6. It can be used to verify analytical solutions.
7. By simulating diff. capabilities for m/c, re. can be determined.
8. Simulation models designed for training allow learning without any c ast.
9. Simulation in the form of animation can show the system in action, so that the
plan can be visualized.
10. The interactions in modern complex system like factory, water fabrication, etc.
can be treated only through simulation.
SAQ 04-01-02: Define simulation. When simulation is not appropriate tool?
Solution: Simulation: A simulation of a system or an organism is the operation of a

model or simulator which is a representation of the system or organism. The model is
amenable to manipulation which would be impossible, too expensive or unpractical to
perform on the entity it portrays. The operation of the model can be studied and for it,
properties concerning the behavior of the actual system can be inferred.
Simulation is not appropriate tool if
1. Problem is solvable by common sense.
2. Problem can be solved mathematically.
3. Direct experiments are easier.
4. Simulation costs exceed the savings.

5. Simulation requires time which is not available.
6. No i/p data is available and simulation needs data.
7. Simulation cannot be verified and validated.
8. System behavior is too complex or unknown.
01-02: ADVANTAGES OF THE SIMULATION TECHNIQUE

There are so many advantages of simulation if we compare the simulation
technique with some other mathematical programming and standar d probability
analysis. Some of these are as follows:
i) Simulation technique is easier to use than the mathematical models and is

considered quite superior to the mathematical analysis.
ii) Simulation models are comparatively flexible and can be modified to accu mulate
the changing environments of the real situations.
iii) Computer simulation can compress the performance of a system over several years
and involving large calculation into a few minutes of computer running time.
iv) It is a valuable and convenient method of breaking down a complicated system into
subsystem and then studying each of these subsystems individually or jointly with
other.
v) Simulation experiments are done on the model, not on the system itself.
Experimentation takes into consideration additional information during analysis that
most quantitative models do not permit. In other words, simulation can be used to
‘experiment’ on a model of a real situation, without incurring the costs of operating
on the system.
vi) Simulation can be used as a pre-service test to try out new policies and decision
rules for operating a system before running the risk of experimentation in the real
system.
vii) Simulation has advantageously been used for training the operating and managerial
staff in the operation of complex plane.
01-03: LIMITATIONS OF THE SIMULATION TECHNIQUE

In spite of all the advantages, the simulation technique suffers from the following
limitations:

(i) Simulation models are expensive and take a long time to develop. For
example, a corporate planning model may take a long time to develop and
may also prove to be expensive.
(ii) Quantification of the variable is another difficulty. In the number of

situations, it is not possible to quantity all the variable that affects the
behavior of the system.
(iii) It is good to rely on simulation technique every time. Some simple

problem, which can more appropriately be handle by other technique of
mathematical programming.
(iv) It is the trial-and-error approach that produces different solutions in repeated

runs. This means it does not generate optimal solutions to problems.
(v) The simulation model does not produce answers by itself. The user has to
provide all the constraints for the solutions that he wants to examine.
01-04: APPLICATIONS OF SIMULATION

Simulation is quite versatile and commonly applied technique for solving
decision problem. It has been applied successfully to wide range of problems of science
and technology as given below:
i) In the field of basic science, it has been used to evaluate the area under the
curve, to estimate the value of 𝜋, in matrix inversion and study of practical
diffusion.
ii) In the industrial problem including shop floor management, design of compute r
systems, design of queuing system, inventory control, communication network,
chemical process, nuclear reactor and scheduling of the production process.
iii) In business and economic problems, including customer behavior, price

determination, economic forecasting, portfolio selection and capital budgeting.
iv) In social problems, including population growth, effect environment on health

and group behavior.
v) In biomedical systems, including fluid balance, distribution of electrolyte in

human body and brain activities.
vi) In the design of weapon system, war strategies and tactics.
vii) In the study of projects involving risky investment.

SOLVED PROBLEMS 04
Problem 04-01-03: Discuss ‘Simulation is a bypass for complex mathematical
analysis.’
Solution: Many important managerial decision problems are too intricate to be solved
by mathematical programming and experimentation with the actual system, even if
possible is too costly and risky. Simulation offers the solution by allowing
experimentation with model of the system without interfering with the real system.
Simulation is, thus, often a bypass for complex mathematical analysis.
Problem 04-01-04: What are the major steps in the simulation process?
Solution: To use simulation, a manager should (1) define a problem, (2) introduce the
variables associated with the problem, (3) construct a mathematical model, (4) set up
possible courses of action for testing, (5) run the experiment, (6) consider the results
(possibly deciding to modify the model or change da ta inputs) and (7) decide what
course of action to take.
SELF-TEST 04
MCQ 01-03: Few causes of simulation analysis failure are
a) inadequate level of user participation,
b) inappropriate levels of detail,
c) incomplete mix of essential skills,
MCQ 01-04: As simulation is not an analytical model, therefore, the result of

simulation must be viewed as
a) Unrealistic,
b) Approximation,
c) Exact,
d) Simplified.

SAQ 04-01-03: What are the advantages and limitations of simulation models? Any
three each.
Solutions: Advantage of simulation:

i) Simulation can be used as a pre-service test to try out new policies and
decision rules for operating a system before running the risk of
experimentation in the real system.
ii) Computer simulation can compress the performance of a system over several
years and involving large calculation into a few minutes of computer running
time.
iii) Simulation has advantageously been used for training the operating and
managerial staff in the operation of complex plane.
Limitations of simulation:
i) Simulation models are expensive and take a long time to develop. For
example, a corporate planning model may take a long time to develop and
may also prove to be expensive.
ii) Quantification of the variable is another difficulty. In the number of

situations, it is not possible to quantity all the variable that affects the
behavior of the system.
iii) It is the trial-and-error approach that produces different solutions in repeated

runs. This means it does not generate optimal solutions to problems.
SAQ 04-01-04: Define simulation? Give application of simulation in four different

fields?
Solution: Definition: Simulation is the process of designing a model of a real system

and conducting experiments with this model for the purpose of understanding the
behavior (within the limits imposed by a criterion or set of criteria) for the operation of
the system.
Applications of simulation:
i) In the industrial problem including shop floor management, design of computer

systems, design of queuing system, inventory control, communication network,
chemical process, nuclear reactor and scheduling of the production process.
ii) In business and economic problems, including customer behavior, price

determination, economic forecasting, portfolio selection and capital budgeting.
iii) In social problems, including population growth, effect environment on health

and group behavior.
iv) In biomedical systems, including fluid balance, distribution of electrolyte in

human body and brain activities.

SUMMARY
Simulation is a powerful and intuitive technique and uses computer to simulate
the operation of an entire system or process. Simulation has so many applications like
in study projects involving risky investment, in binomial system, in social problem, in
industrial problem, etc.

1. What is simulation? Describe its advantage in solving the problems. Give its
main limitations with suitable example.
2. ‘When it becomes difficult to use an optimization technique for solving a

problem, one has to resort to simulation technique. ‘Discuss.
3. Define simulation. Why is simulation used? Give one application area when this
technique is used in practice.
4. Why is a computer necessary in conducting a real-life simulation?
5. Do you think the application of simulation will strongly increase in the next ten
years? Give reasons for your answer.
6. Distinguish between solutions derived from simulation models and solutions

derived from analytical models?
KEY WORDS
Simulation, imitation.
REFERENCES
BOOKS
& COMPANY LTD

PRAGATI EDITION
MOOCS
YOUTUBE VIDEOS
1) https://youtu.be/oZkNjGgaNVY

WIKIPEDIA
https://en.wikipedia.org/wiki/Simulation
http://www.universalteacherpublications.com/univ/ebooks/or/Ch16/simintro.htm
https://www.coursehero.com/file/p38956db/10-3-What-types-of-management-
problems-can-be-solved-more-easily-by-using/
https://www.ques10.com/p/42701/define-simulation-explain-when-simulation-is-an-
-1/
OER

UNIT 04-02: MONTE CARLO SIMULATION
LEARNING OBJECTIVES
 Apply Monte Carlo simulations technique in solving inventory problem,

queuing problem, investment problem, maintenance problem, job sequencing
problem.
 Develop random number intervals and use them to generate outcomes.
INTRODUCTION
During the World War II, Jhon Von Numann and Stainslaw Ulam study how far
neutrons would travel through different materials, for this they use simulation technique
first time. This technique is approximate but gives workable solution to the pr oblem. It
is become very popular and important technique in business and industry after
remarkable success in neutrons problem. ‘Monte Carlo’ is the code number given to the
technique by above two scientists. Means Monte Carlo method of simulation was
developed by these two scientists.
Monte Carlo methods comprised that bunch of mathematics which is concern

with experiments on random numbers and it provides solutions to complicated
operation research problem by combining it with analytical or iterative p rocedures. If
the model involves random sampling from a probability distribution, procedure is called
Monte Carlo simulation.
Steps involved in Monte Carlo Method are as follows:
Step 1: Select the measure of effectiveness (objective function) of the problem.
Step 2: Identify the variables, which have the greatest effect on the objective of the
problem.
Step 3: Determine the cumulative probability distribution of each variable selected in

the step 2.
Step 4: Generating random numbers and then assigning an appropriate set of random
numbers to represent value or range (interval) of values for each random variable.
Step 5: Conducting the simulation experiment using random sampling.
Step 6: Repeating Step 4 until the required number of simulation runs has been
generated.
Step 7: Designing and implementing a course of action and maintaining control.

02-01: SIMULATION OF INVENTORY PROBLEMS
In this section, we have to study to apply Monte Carlo method for solving
inventory problem.
The word inventory means a physical stock of material or goods or commodities

or other economic resources that are stored or reserved or kept in stock or in hand for
smooth and efficient running of future affairs of an organization at the minimum cost of
funds or capital blocked in the form of materials or goods (Inventories).
The function of directing the movement of goods through the entire

manufacturing cycle from the requisitioning of raw materials to the inventory of
finished goods in an orderly manner to meet the objectives of maximum customer
service with minimum investment and efficient (low cost) plant operation is termed as
inventory control.
SOLVED PROBLEMS 01
Problem 04-02-01: A bakery keeps stock of a popular brand of cake. Previous
experience shows the daily demand pattern for the item with associated probabilities, as
given below
Daily demand (number) 0 15 25 35 45 50
Probability 0.01 0.15 0.20 0.50 0.12 0.02
Use the following sequence of random numbers 48,78,09,51,56,77,15,14,68,09.
i) Using the sequence, simulate the demand for the next 10 days.
ii) Also, find the stock situation if the owner of the bakery decides to make 35 cakes
every day. Also, estimate the daily average demand for the cakes on the basis of the
simulated data.
Solution: Using the daily demand distribution, we first obtain a probability distribution
as shown in table
Daily Cumulative Random number Random Number fitted

Probability
demand probability interval
0 0.01 0.01 0
09(3), 15(7), 14(8),
15 0.15 0.16 01-15
09(10)
25 0.20 0.36 16-35
48(1), 78(2), 51(4),
35 0.50 0.86 36-85
56(5), 77(6), 68(9)
45 0.12 0.98 86-97
50 0.02 1.00 98-99

In order to simulate the demand, the number 00 is assigned to zero demand, number 01 -
15 are assigned to demand of 15 cakes, and 16-35 are assigned to demand of 25 cakes
and so on. We have given the 10 random numbers which is fitted in the last column of
above table. These random numbers give the demand for cakes for next 10 days. Serial
numbers of these random numbers are shown in the parenthesis for instance 09(3)
represent the 09 which is at the third position.
The simulation calculations for a period of 10 days are given in table, here bakery
owner decide to make 35 cakes every day.
Days No. of daily cake made Random number Simulated demand Stock
1 35 48 35 0
2 35 78 35 0
3 35 09 15 20
4 35 51 35 0
5 35 56 35 0
6 35 77 35 0
7 35 15 15 20
8 35 14 15 20
9 35 68 35 0
10 35 09 15 20
Total 270 80
Therefore, the number of cakes demanded in the next 10 days is:
35, 35, 15, 35, 35, 35, 15, 15, 35, 15.
And the average daily demand for the cakes = = 27 cakes.
Then, total stock of the left 80 cakes at the end of 10 days.
Problem 04-02-02: A company trading in motor vehicle spares wishes to determine the
level of stock it should carry for the item in its range. Demand is not certain and there
is a lead time for stock replenishment. For one item X, if the following information is
obtained:
Demand (units/day) 3 4 5 6 7
Probability 0.1 0.2 0.3 0.3 0.1
Carrying cost per unit per day = 20 paise,
ordering cost per order = Rs. 5,
lead time for replenishment = 3 days.
Stock in hands at the beginning of the simulation exercise was 20 units. You are
required to carry out a simulation run over a period of 10 days with the objective of
evaluating the following.

Order 15 units when present inventory plus any outst anding order falls below 15 units.
The sequence of random nos. used is 0, 9, 1, 1, 5, 1, 8, 6, 3, 5, 7, 1, 2 ,9 using the first
number for one day. Your calculation should include the total cost of operating this
inventory rule for 10 days.
Solution: First, demand in each of the 10 days is determined. The probability

distribution of demand is shown in following table :
Probability distribution of demand

Demand Cumulative Range/ Random umber
Probability
probability intervals fitted
3 0.1 0.1 0 0(1)
4 0.2 0.3 01-02 1(3), 1(4), 1(6)
5 0.3 0.6 03-05 5(5), 3(9), 5(10)
6 0.3 0.9 06-08 8(7),6(8)
7 0.1 1.0 09 09(2)
Thus, using random number sequence and last column in above table, the demand for X
on the ten days is 3, 7, 4, 4, 5, 4, 6, 6, 5 and 5 units respectively. Here , 14 random
number are given out of these we use the first 10 random number.
The inventory carrying cost and ordering cost are computed in the following table :
Simulations of demand, delivery and costs

Units Load Inventory Ordering
Demand Units
Day time Inventory carrying cost
(Units) received
ordered (days) cost (Rs.) (Rs.)
0 - 20 4.00
1 3 17 3.40
2 7 15 3 10 2.00 5.00
3 4 6 1.20
4 4 2 0.40
5 5 15 3 15 12 2.40 5.00
6 4 8 1.60
7 6 2 0.40
8 6 15 3 15 12 2.40 5.00
9 5 7 1.40
10 5 2 0.40
Total 19.00 15.00
The cost of operating inventory for 10 days = Rs. (19+15) = Rs.34.
SELF-TEST 01
MCQ 02-01: Simulation technique in inventory control can use for
a) To find the demand for next finite days.
b) Find the stock after finite day.

c) To find the cost of inventory for finite days.
MCQ 02-02: If in the inventory problem we have to simulate the demand for next 20
days and for this they are given 25 random numbers then
a) we have to use any 20 random numbers among 25.
b) we have to use first 20 random numbers among 25.
c) we have to use 5 to 25 random numbers among 25.
d) we have to use distinct random numbers only.

SAQ 04-02-01: What are the different steps involved in Monte Carlo method?
Solution: Steps involved in Monte Carlo method are as follows.
Steps involved in Monte Carlo Method are as follows:
Step 1: Select the measure of effectiveness (objective function) of the problem.
Step 2: Identify the variables, which have the greatest effect on the objective of the
problem.
Step 3: Determine the cumulative probability distribution of each variable selected in

the step 2.
Step 4: Generating random numbers and then assigning an appropriate set of random
numbers to represent value or range (interval) of values for each random variable.
Step 5: Conducting the simulation experiment using random sampling.
Step 6: Repeating Step 4 until the required number of simulation runs has been
generated.
Step 7: Designing and implementing a course of action and maintaining control.
SAQ 04-02-02: A company manufactures around 200 mopeds. Depending upon the
availability of raw materials and other conditions, the daily production has been
varying from 196 mopeds to 204 mopeds, whose probability distribution is as given
below:
Production/day: 196 197 198 199 200 201 202 203 204
Probability :0.05 0.09 0.12 0.14 0.20 0.15 0.11 0.08 0.06

The finished mopeds are transported in a specially designed three -storied lorry that can
accommodate only 200 mopeds. Using the following 15 random numbers: 82, 89, 78,
24, 53, 61, 18, 45, 23, 50, 77, 27, 54 and 10, simulate the mopeds waiting in the
factory? What will be the average number of mopeds waiting in the factory? What will
be the number of empty spaces in the lorry?
Solution: Using production per day distribution, the daily production distribution is
shown in table:
Production/ day Probability Cumulative probability Random number interval

196 0.05 0.05 00-04
197 0.09 0.14 05-13
198 0.12 0.26 14-25
199 0.14 0.40 26-39
200 0.20 0.60 40-59
201 0.15 0.75 60-74
202 0.11 0.86 75-85
203 0.08 0.94 86-93
204 0.06 1.0 94-100
Based on the given 15 random numbers, simulation experiment of the productio n per
day is show in table:
No. of mopeds Empty space in

Days Random number Production /day
waiting lorry
1 82 202 2
2 89 203 3
3 78 202 2
4 24 198 2
5 53 200 0
6 61 201 1
7 18 198 2
8 45 200 0
9 4 196 4
10 23 198 2
11 50 200 0
12 77 202 2
13 27 199 1
14 54 200 0
15 10 197 3
Average number of mopeds waiting in the factory = [2 + 3 + 2 + 1 + 2 + 1] =
1 𝑚𝑜𝑝𝑒𝑑 (𝑎𝑝𝑝𝑟𝑜𝑥. . ).
Average number of empty spaces in the lorry = = 0.86.
02-02: SIMULATION OF QUEUING PROBLEM

In this section, we have to study to apply Monte Carlo method for solving
queuing problem.

SOLVED PROBLEMS 02
Problem 04-02-03: A dentist schedules all his patients for 30-minute appointments.
Some of the patients take more 30 minutes some less, depending on the type of dental
work to be done. The following summary shows the various categories of work, their
probabilities and time actually needed to complete the work:
Category Time required (Minutes) Probability of category

Filling 45 0.40
Crown 60 0.15
Cleaning 15 0.15
Extraction 45 0.10
Check-up 15 0.20
Simulate the dentist’s clinic for four hours and determine the average waiting time for
the patient as well as the idleness of the doctor. Assume that all the patients show up
at the clinic at exactly their schedule arrival time, starting at 8 a.m. Use the following
random numbers for handling the above problem:
40, 82, 11, 34, 25, 66, 17 and 79.
Solution: The time taken by the dentist to treat the eight patients arriving in four hours
at the clinic is calculated in the table below:
Random Random
Time Cumulative
Category Probability number number
(minutes) probability
interval fitted
11(3), 34(4)
Filling 45 0.40 0.40 0-39
25(5), 17(7)
Crown 60 0.15 0.55 40-54 40(1)
Cleaning 15 0.15 0.70 55-69 66(6)
Extraction 45 0.10 0.80 70-79 79(8)
Check-up 15 0.20 1.0 80-99 82(2)
Thus, the times taken by the dentist to treat the eight patients are
60, 15, 45, 45, 45, 15, 45 and 45, respectively.
Let us simulate the dentist clinic (for eight patients) starting at 8.00 a.m.
Dentist
Patients Arrival Waiting time on the part Idle time for
treatment
no. time of the patient the dentist
Starts End
1 8.00 8.00 9.00 0 -
2 8.30 9.00 9.15 30 -
3 9.00 9.15 10.00 15 -
4 9.30 10.00 10.45 30 -
5 10.00 10.45 11.30 45 -
6 10.30 11.30 11.45 60 -
7 11.00 11.45 12.30 45 -
8 11.30 12.30 13.15 60 -
Total 285 -

Therefore, average waiting time = = 35.628 minutes.
Average idleness of the dentist = Nil.
Problem 04-02-04: The management of ABC company is considering the question of

marketing a new product. The fixed cost required in the project is Rs. 4,000. Three
factors are uncertain, viz., the selling price, variable cost and the annual sales volume.
The product has a life of only one year. The management has the data on these three
factors as under:
Selling Probability Variable Probability Sales volume Probability

price (Rs.) cost (Rs.) (Rs.)
3 0.2 1 0.3 2,000 0.3
4 0.5 2 0.6 3,000 0.3
5 0.3 3 0.1 5,000 0.4
Considering the following sequence of thirty random numbers: 81, 32, 60, 04, 46,
31, 67, 25, 24, 10, 40, 02, 39, 68, 08, 59, 66, 90, 12, 64, 79, 31, 86, 68, 82, 89, 25, 11,
98, 16.
Using the sequence (First 3 random numbers for the first trial, etc.) simulate t he
average profit for the above project on the basis of 10 trails.
Solution: The cumulative probability distribution and ran dom number interval for
selling price, variable cost sales volume is shown below:
Cumulative Random number

Selling price Probability
probability interval
3 0.2 0.2 00-19
4 0.5 0.7 20-69
5 0.3 1.0 70-99
Variable cost (Rs.)
1 0.3 0.3 00-29
2 0.6 0.9 30-89
3 0.1 1.0 90-99
Sales volume
(Units)
2,000 0.3 0.3 0.3
3,000 0.3 0.6 0.9
5,000 0.4 1.0 1.0
The simulation experiment sheet for finding average profit is shown in table
Selling Sales
No. of Random Random Variable Random
price volume
trials number number cost (Rs.) number
(Rs.) (‘000 units)
1 81 5 32 2 60 5
2 04 3 46 2 31 3
3 67 4 25 1 24 2
4 10 3 40 2 02 2
5 39 4 68 2 08 2

6 59 4 66 2 90 5
7 12 3 64 2 79 5
8 31 31 86 2 68 5
9 28 82 89 2 25 2
10 11 11 98 3 16 2
Trial number 𝑃𝑟𝑜𝑓𝑖𝑡 = (𝑆𝑒𝑙𝑙𝑖𝑛𝑔 𝑝𝑟𝑖𝑐𝑒 − 𝑐𝑜𝑠𝑡 𝑝𝑟𝑖𝑐𝑒) × 𝑠𝑎𝑙𝑒𝑠 𝑣𝑜𝑙𝑢𝑚𝑒 − 𝑓𝑖𝑥𝑒𝑑 𝑐𝑜𝑠𝑡
1 (5 – 2) × 5,000 – 4,000 = 11,000
2 (3 – 2) × 3,000 – 4,000 = (– 1000)
3 (4 – 1) × 2,000 – 4,000 = 2,000
4 (3 – 2) × 2,000 – 4,000 = (– 2,000)
5 (4 – 2) × 2,000 – 4,000 = 0
6 (4 – 2) × 5,000 – 4,000 = 6,000
7 (3 – 2) × 5,000 – 4,000 = 1,000
8 (4 – 2) × 5,000 – 4,000 = 6,000
9 (5 – 2) × 2,000 – 4,000 = 2,000
10 (3 – 3) × 2,000 – 4,000 = (– 4,000)
Total 2100
Average profit per trial = = 𝑅𝑠. 2100 .
SELF-TEST 02
MCQ 02-03: Monte Carlo method is useful for solving
a) Queuing problem.
b) Inventory problem.
c) Investment and budgeting problem.
MCQ 02-04: While assigning random numbers in Monte Carlo simulation, it is
a) not necessary to assign the exact range of random number interval as the
probability.
b) necessary to develop a cumulative probability distribution .
c) necessary to assign the particular appropriate random number.

SAQ 04-02-03: A firm has a single channel service station with the following arrival
and service time probability distributions:
Interarrival time (minutes) Probability Service time(minutes) Probability

10 0.10 5 0.08
15 0.25 10 0.14

20 0.30 15 0.18
25 0.25 20 0.24
30 0.30 25 0.22
30 0.14
The customer’s arrival at the service station is a random phenomenon and the time
between the arrivals varies from 10 to 30 minutes. The service time varies from 5
minutes to 30 minutes. The queuing process begins at 10 a.m. and proceeds for nearly 8
hours. An arrival immediately, goes to the service facility if it is free. Otherwise, it
waits in a queue. The queue discipline is first-come first-served.
If the attendant’s wages are Rs. 10 per hour and the customer’s waiting time costs Rs .
15 per hour, then would it be an economical proposition to engage a second attendant?
Answer using Monte Carlo simulation technique. Random numbers are 93, 14, 72, 10,
21, 81, 87, 90, 38, 71, 63, 14, 53, 64, 42, 07, 54, 66.
Solution: The cumulative probability distributions and random number inter val, both
for interarrival time and service time, are shown in tables below:
Interarrival time in Cumulative Random No.

Probability
minutes probability interval
10 0.10 0.10 0-9
15 0.25 0.35 10-34
20 0.30 0.65 35-64
25 0.25 0.90 65-89
30 0.10 1.0 90-100
Service time in minutes
5 0.08 0.08 00-07
10 0.14 0.22 08-21
15 0.18 0.40 22-39
20 0.24 0.64 40-63
25 0.22 0.86 64-85
30 0.14 1.0 86-100
The simulation worksheet developed to the given problem is shown in table:
Interarri
Arriv Rando Service Exi
vel Arrival Wetti Ran Servi
al m time t Time in
time(mi ng d. ce
numb numb starts(mi tim system
time n.) time No. time
er er n.) e
(Min.)
( 10)=(6)+
(1) (2) (3) (4) (5) (6) (7) (8) (9) (9 )
1 20 15 10.15 10.15 0 26 15 30 15
2 73 25 10.40 10.40 0 43 20 60 20
3 30 15 10.55 11.00 5 98 30 90 35
4 99 30 11.25 11.30 5 87 30 120 35
5 66 25 11.45 12.00 15 58 20 140 35
6 83 25 12.10 12.20 10 90 30 170 40
7 32 15 12.25 1.05 35 84 25 175 60
8 75 25 12.50 1.30 40 60 20 124 60

9 4 10 1.00 1.50 50 8 10 225 60
10 15 15 1.15 2.00 45 50 20 245 65
11 29 15 1.30 2.20 50 37 15 260 65
12 62 20 1.50 2.35 45 42 20 280 65
13 37 20 2.10 2.55 45 28 15 295 60
14 68 25 2.35 3.10 35 84 25 320 60
15 94 30 3.05 3.35 30 65 25 345 55
From the 15 samples of waiting time, 225 minutes, and the time spent, 545 minutes, by
the customer in the system, we compute all average waiting time in the system and
average service time as follows:
𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑤𝑎𝑖𝑡𝑖𝑛𝑔 𝑡𝑖𝑚𝑒 = 380/15 = 25.3 𝑚𝑖𝑛𝑢𝑡𝑒𝑠.
545
𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑠𝑒𝑟𝑣𝑖𝑐𝑒 𝑡𝑖𝑚𝑒 = = 36.33 𝑚𝑖𝑛𝑢𝑡𝑒𝑠.
15
Thus, the average cost of waiting and service is given by
25.30
𝐶𝑜𝑠𝑡 𝑜𝑓 𝑤𝑎𝑖𝑡𝑖𝑛𝑔 = 15 × ( ) = 𝑅𝑠 6.32 𝑝𝑒𝑟 ℎ𝑜𝑢𝑟.
60
36.33
𝐶𝑜𝑠𝑡 𝑜𝑓 𝑠𝑒𝑟𝑣𝑖𝑐𝑒 = 10 × ( ) = 𝑅𝑠 6.05 𝑝𝑒𝑟 ℎ𝑜𝑢𝑟.
60
Since the average cost of service, per hour, is less than the aver age cost of waiting per
hour, thereforesecond attendant may be hired.
SAQ 04-02-04: Give any three limitations of Monte Carlo simulations?
Solution: Limitations of Monte Carlo simulation:
1. It only provides us with statistical estimates of results, not exact figures.
2. It is fairly complex and can only be carried out using specially designed
software that may be expensive.
3. The complexity of the process may cause errors leading to wrong results that
can be potentially misleading.
02-03: SIMULATION OF INVESTMENT AND BUDGETING, JOB SEQUENCING

PROBLEM
In this Section, we have to study the Monte Carlo method for simulating
investment and budgeting problem and job sequencing problem.
SOLVED PROBLEMS 03
Problem 04-02-05: The investment corporation wants to study the investment projects
based on three factors, namely, market demand in unit’s price per unit minus cost per
unit, and investment required. These factors are believed to be independent of each

other. In analyzing a new consumer product, the Corporation estimates the following
probability distributions:
Annual demand Price minus cost per unit Investment required

Units Probability Rs. Probability Rs. Probability
20000 0.05 3.00 0.10 27,50,000 0.25
25000 0.10 5.00 0.20 30,00,000 0.50
30000 0.20 7.00 0.40 35,00,00 0.25
35000 0.30 9.00 0.20
40000 0.20 10.00 0.10
45000 0.10
50000 0.05
Using the simulation process, repeat the trial 10 times, compute the return on
investment for each trial taking these three factors into account. What is the most likely
return? Use the following random number for annual demand, (piece -cost) and
investment required.
28, 57, 60, 17, 64, 20, 27, 58, 61, 30; 19, 07, 90 ,02, 57, 28, 29, 83, 58, 41;, 18,
67, 71, 43, 68, 47, 24 ,19, 97.
Solution: First of all, we have to find cumulative probability distribution and the
random number interval for these three factors.
Cumulative Random number

Annual demand (000’s) Probability
probability interval
25 0.05 0.05 00-04
30 0.10 0.15 05-14
35 0.20 0.35 15-34
40 0.30 0.65 35-64
45 0.20 0.85 65-84
50 0.10 0.95 85-94
55 0.05 1.0 95-99
(Price cost) per unit
3.00 0.10 0.10 00-09
5.00 0.20 0.30 10-29
7.00 0.40 0.70 30-69
9.00 0.20 0.90 70-89
10.00 0.10 1.0 90-99
Investment required(000’s)
(Rs.)
2750 0.25 0.25 00-24
3000 0.50 0.75 25-74
3500 0.25 1.00 75-99
The yearly return can be determined by the formula
(𝑝𝑟𝑖𝑐𝑒 − 𝑐𝑜𝑠𝑡) × 𝑁𝑢𝑚𝑏𝑒𝑟𝑜𝑓 𝑢𝑛𝑖𝑡𝑠 𝑑𝑒𝑚𝑎𝑛𝑑𝑒𝑑

𝑅𝑒𝑡𝑢𝑟𝑛(𝑅) = × 100
𝑖𝑛𝑣𝑒𝑠𝑚𝑒𝑛𝑡

𝑟𝑜𝑓 ×𝑑 𝑑
or 𝑅𝑒𝑡𝑢𝑟𝑛(𝑅) = × 100
𝑣
The result of simulation is in the table
Random
Random number Simulated
Simulated Random
number for Simulated investment Simulated
Trials demand number for
for (price - price (Rs.) (000’s) return (%)
(000’s) investment
demand profit) (Rs.)
per cost
1 28 35 19 5.00 18 2750 6.36
2 57 40 07 3.00 67 3000 4.00
3 60 40 90 10.00 16 2750 14.55
4 17 35 02 3.00 71 3000 3.50
5 64 40 57 7.00 43 3000 9.33
6 20 35 28 5.00 68 3000 5.83
7 27 35 29 5.00 47 3000 5.83
8 58 40 83 9.00 24 2750 13.10
9 61 40 85 7.00 19 2750 10.18
10 30 35 41 7.00 97 3500 7.00
From the above table, we observe that the highest return is 14.55%, which is
corresponding to the annual demand 40000 units resulting a profit of Rs. 10 per unit
and the required investment will be Rs. 27,50,000.
Problem 04-02-06: Find the value of 𝜋 experimentally by simulation.
Solution: Draw a coordinate axes OX and OY. With center O draw an arc PR of unit
radius as shown in Fig. and complete the square OPQR. Equation of the circle is
𝑥 + 𝑦 = 1. From random number table C-1 (From D. H. Hira Book last pages) , select
any two random numbers, say 0.2068 and 0.7295 (first two four-digit number with
decimal from the second column) and let 𝑥 = 0.2068 and 𝑦 = 7295. Plot the point P
(0.2068, 0.7295). Obviously, if 𝑥 + 𝑦 ≤ 1, 𝑃 will lie inside or on arc of circle but if
𝑥 + 𝑦 > 1,the point 𝑃 will lie outside the arc but within the square.

In this manner, hundreds or thousands of pair of random numbers are selected
and it is ascertained whether the points representing them lie in/on the arc or beyond
the arc but inside the square. Suppose N is the number if points considere d, out of n lie
in/on the arc. Then,
𝑟 𝑜 𝑑 𝑟 ( ) 𝜋
= = = , or 𝜋 = .
𝑁 𝑟 𝑜𝑓 𝑞𝑢 𝑟 . 𝑁
The above equation gives the experimental value of 𝜋. Obviously, the larger the
sample N. closer will be obtained value to the true value of 𝜋.
SELF-TEST 03
MCQ 02-05: The formula for finding simulated return is
𝑜 × 𝑣
a) 𝑅𝑒𝑡𝑢𝑟𝑛(𝑅) = × 100.
𝑑 𝑑
𝑜 ×𝑑 𝑑
b) 𝑅𝑒𝑡𝑢𝑟𝑛(𝑅) = × 100.
𝑣
𝑟𝑜𝑓 ×𝑑 𝑑
c) 𝑅𝑒𝑡𝑢𝑟𝑛(𝑅) = × 100.
𝑣
𝑟𝑜𝑓 × 𝑣
d) 𝑅𝑒𝑡𝑢𝑟𝑛(𝑅) = × 100.
𝑑 𝑑
MCQ 02-06: To make simulation more popular, we need to avoid
a) large cost over runs.
b) prolonged delays.
c) user dissatisfaction with simulation results.

SAQ 04-03-05: Philips India is engaged in manufacturing different type’s equipment
for various consumers. The company has two assembly lines to produce its product. The
processing time for each of the assembly lines is regarded as random variable and is
described by the following distribution:
Processing time Assembly X Assembly Y

40 0.10 0.20
2 0.15 0.40
44 0.40 0.20
46 0.10 0.15
48 0.25 0.05
Using the following random numbers, generate data on processing times for 10 units of
the product and compute the expected processing time for the product.

4236, 7573, 4943, 1283, 2014, 3604, 9344, 5316, 7606, 0089.
For the purpose, read the number horizontally, taking the first two digits for processing
time on assembly X and the last two digits for processing time on assembly Y.
Solution: First we find the cumulative probability and random number intervals.
Assembly X Assembly Y
Processing
Cum. R.N. Cum. R. N.
time Prob. Prob.
Prob. Interval Prob. Interval
40 0.10 0.10 00-09 0.20 0.20 00-19
42 0.15 0.25 10-24 0.40 0.60 20-59
44 0.40 0.65 25-64 0.20 0.80 60-79
46 0.10 0.75 65-74 0.15 0.95 80-94
48 0.25 1.00 75-99 0.05 1.00 95-99
We have given that, for the purpose, read the number horizontally, taking the first two
digits for processing time on assembly X and the last two digits for processing time on
assembly Y. For instants 4236, we have to take 42 and 36 for assembly X and Y
respectively and so on. Then, we have to find corresponding time. For this , we made the
following table:
Assembly X Assembly Y Total time

Sr. No.
R. No. Time(min.) R. No. Time(min.) (min.)
1 45 44 36 42 86
2 75 48 73 44 92
3 49 44 43 42 86
4 12 42 83 46 88
5 20 42 14 40 82
6 36 44 04 40 84
7 93 48 44 42 90
8 53 44 16 40 48
9 76 48 06 40 88
10 00 40 89 46 86
Total time 866
Thus, total time for 10 units=866 minutes.
And expected time for the product = = 86.6 𝑚𝑖𝑛𝑢𝑡𝑒𝑠 .
SAQ 04-02-06: List applications of Monte Carlo simulations?
Solution: Applications of Monte Carlo simulation is
1. It is used to value projects that require significant amounts of funds and may
have future financial implications on a company.
2. It can be used to simulate profits or losses in the online trading of stocks.
3. Simulation of the values of assets and liabilities of a pension benefit scheme.

4. It can also be used to value complex securities such as American/European
options.
02-04: SIMULATION OF MAINTENANCE PROBLEM

In this Section, we have simulate the maintenance problem using Monte Carlo Method.
SOLVED PROBLEMS 04
Problem 04-02-07: A plant has a large number of similar machines. The machine
breakdown or failure is random and independent. The shift in charge of the plant
collected the data about the various machine’s breakdown times, the repair time
required on hourly basis, and the record for the past 100 observations. This is shown
below was:
Time recorded machine Repair time

Probability Probability
breakdown (hours) required
0.5 0.5 1 0.28
1 0.6 2 0.52
1.5 0.16 3 0.20
2 0.33
2.5 0.21
3 0.19
For each hour that one machine is down due to being, or waiting to be, repaired, the
plant loses Rs.70 by way of lost production. A repairman is paid at Rs.20 per hour.
Simulate this maintenance system for 15 breakdowns. How many repairmen should the
plant hire for repair work?
Solution: The random numbers coding for the hourly breakdowns and the repair times
are shown in table
Time between Cumulative Random number

Probability
breakdowns (hours) probability range
0.5 0.05 0.05 00 − 04
1 0.06 0.11 05 − 10
1.5 0.16 0.27 11 − 26
2 0.33 0.60 27 − 59
2.5 0.21 0.81 60 − 80
3 0.19 1.00 81 − 99
Repair time required (hours)
1 0.28 0.28 00-27
2 0.52 0.80 28-79
3 0.20 1.00 80-99
The simulation worksheet is shown in table below. It is assumed that the first day
begins at midnight (00.00 hours) and also that the repairman begins work at 00.00

hours. The first breakdown occurred at 2.30 a.m. and the second occurred after 3 hours,
at clock time of 5.30 a.m.
R.
Repai R. N. Repai Tota
No Repair
Breakdow Time Time of r For r l Waitin
. time
n number betwee Breakdow work Repai work idle g time
for require
(B. N.) n n begins r ends time (Hr.)
B. d
at work at (Hr.)
N
1 61 2.5 02.30 2.30 87 3 5.30 3.00 -
2 85 3 5.30 5.30 39 2 7.30 2.00
3 16 1.5 7.00 7.30 28 2 9.30 2.30 0.30
4 46 2 9.00 9.30 97 3 12.30 3.30 0.30
5 88 3 12.00 12.30 69 2 14.30 2.30 0.30
6 8 1 13.00 14.30 87 3 17.30 4.30 1.30
7 82 3 16.00 17.30 52 2 19.30 3.30 1.30
8 56 2 18.00 19.30 52 2 21.30 3.30 1.30
9 22 1.5 19.30 21.30 15 1 22.30 3.00 2.00
10 49 2 21.30 22.30 85 3 1.30 4.00 1.00
11 44 2 23.30 1.30 41 2 3.30 4.00 2.00
12 33 2 1.30 3.30 82 3 6.30 5.00 2.00
13 77 2.5 4.00 6.30 98 3 9.30 5.30 2.30
14 87 3 7.00 9.30 99 3 12.30 5.30 2.30
15 54 2 9.00 12.30 23 2 14.30 5.30 3.30
38.30 36 57.3 21.30
𝑇𝑜𝑡𝑎𝑙 𝑐𝑢𝑟𝑟𝑒𝑛𝑡 𝑚𝑎𝑖𝑛𝑡𝑒𝑛𝑎𝑛𝑐𝑒 𝑐𝑜𝑠𝑡 = 𝐼𝑑𝑙𝑒 𝑡𝑖𝑚𝑒 𝑐𝑜𝑠𝑡 + 𝑅𝑒𝑝𝑎𝑖𝑟𝑚𝑎𝑛’𝑠 𝑤𝑎𝑔𝑒
+ (𝑅𝑒𝑝𝑎𝑖𝑟 𝑡𝑖𝑚𝑒 + 𝑊𝑎𝑖𝑡𝑖𝑛𝑔 𝑡𝑖𝑚𝑒) × 𝐻𝑜𝑢𝑟𝑙𝑦 𝑟𝑎𝑡𝑒 + 𝑇𝑜𝑡𝑎𝑙 ℎ𝑜𝑢𝑟𝑠 × 𝐻𝑜𝑢𝑟𝑙𝑦 𝑤𝑎𝑔𝑒𝑠
= 57.30 × 70 + 38.30 × 20 = 𝑅𝑠 4,777.
Maintenance cost with additional repairmen.
If the plant hires two more repairmen, then no machine will wait to be repaired. Thus,
the total idle time would only be the repairing time of 36.00 hours.
Therefore,𝑇𝑜𝑡𝑎𝑙 𝑐𝑜𝑠𝑡 = 36 × 70 + (38.30 × 2) × 20 = 𝑅𝑠 4,052.
This shows that hiring more than two repairmen would only increase the tota l
maintenance cost. Hence, the plant should ideally hire one additional repairman.
SELF-TEST 04
MCQ 02-07: Which of the following statement is FALSE?
a) The results of simulation experiment should be viewed as exact.
b) Simulation is defined as a technique that uses computer.
c) One of the causes of simulation analysis failure is incomplete mix of essential

skills.

d) The purpose of using simulation technique is to reduce the cost of experiment
on a real-life model.
MCQ 02-08: The Monte Carlo method is useful for solving
a) Inventory problem.
b) Maintenance problem.
c) Job sequencing problem.

SAQ04-04-01: Popa Ltd. trade in a perishable commodity. Each day Popa Ltd. receives
supplies of the goods from a wholesaler but the quantity supplied is a random variable,
as is the subsequent retail customer demand for the commodity. Both supply and
demand are expressed in batches of 50 units and over the past working year (300 days),
Popa Ltd. has kept records of supplies and demands. The results are given below:
Wholesaler Number of days Customer’s Number of days

supplies occurring demand occurring
50 60 50 60
100 90 100 60
150 90 150 150
200 60 200 30
Popa Ltd. buys the commodity at Rs. 6 per unit and sells at Rs. 10 per unit.
At present unsold units at the end of the day are worthless and there are no storage
facilities. Popa Ltd. estimates that each unit of unsatisfied demand on any day costs
them Rs. 2. Using the random numbers 8, 4, 8, 0, 3, 3, 4, 7 , 9, 6,1 and 5. Then simulate
six days trading and estimate the annual profit. Also , repeat the exercise to estimate the
value of storage facilities.
Solution: Wholesaler’s supplies and customers demand are simulated for 6 days as
shown in table:
No. of Total no. No. of days R. N.

Supply R. N. fitted
days of days out of 10 range
50 60 60 2 00-01 0(4),1(11)
100 90 150 5 02-04 4(2),3(5),3(5), 4(6)
150 90 240 8 05-07 7(7),6(10),5(12)
200 60 300 10 08-09 8(1),8(3), 9(8)
Costumer’s
demand
50 60 60 2 00-01 0(4),1(11)
100 60 120 4 02-03 3(5),3(6)
8(1),4(2),8(3), 7(8)
150 150 270 9 04-08
6(10),5(12), 4(7)

200 30 300 10 09 9(9)
For total days divide column by 30.
Random number: 8 4 8 0 3 3 4 7 9 6 1 5
Supplies: 200 100 200 50 100 100 100 150 200 150 50 150
Demand: 150 150 150 50 100 100 150 150 200 150 50 150
Thus, wholesaler’s supplies in the next 6 days are: 200, 100, 200, 50, 100 and 100
units, respectively.
Customer’s demand in the next 6 days is: 150, 150, 200, 150, 50 and 150 units .
The shortage and net profit or loss in the 6 days is now calculated as shown in table:
Supply Demand Shortage

Days Cost Cost Cost Net profit or loss (Rs.)
Unit Unit Unit
(Rs) (Rs.) (Rs.)
1 200 1200 150 1500 - - 1500 – 1200 = 30
2 100 600 150 1500 50 300 1000 – 600 – 100 = 300
3 200 1200 200 2000 - - 2000 – 1200 = 800
4 50 300 150 1500 100 200 500 – 300 – 200 = 0
5 100 600 50 500 - - 500 – 600 = – 100
6 100 600 150 1500 50 100 1000 – 600 – 100 = 300
Net profit from table for 6 days = Rs 1,600
1600 × 300
𝐴𝑛𝑛𝑢𝑎𝑙 𝑝𝑟𝑜𝑓𝑖𝑡 = = 𝑅𝑠. 80,000.
6
The value of storage facilities has been shown in table
Day Supply Demand Storage

1 200 150 50
2 100 150 Nil
3 200 200 Nil
4 50 150 Nil
5 100 50 50
6 100 150 Nil
Shortage cost of Rs. 100 for 50 units could be avoided.
If the storage facilities, then Rs. 200 of shortage cost could be avoided by storing
50 + 50 = 100 𝑢𝑛𝑖𝑡𝑠.
These 100 units could be then sold, yielding a profit of 𝑅𝑠. 4 × 100 = 𝑅𝑠 400.
The value of storage facilities = 200 + 400 = 𝑅𝑠. 600

SUMMARY
In this Section, we discussed Monte Carlo method. It has applications to a wide
variety of areas such as queuing system, inventory system, replacement, P.E.R.T.
projects, financial risk analysis, health care system, distribution system, etc.

1. Explain with suitable example the Monte Carlo method in inventory control and
capital budgeting.
2. A company trading in motor vehicle spare parts wishes to determine the levels of
stock it should carry for the items in its range. The demand is not certain and there
is a lead time for stock replenishment. For an item A, the following information is
obtained:
Demand (units/day): 3 4 5 6 7
Probability: 0.10 0.20 0.30 0.30 0.10
Carrying cost (per unit/day): Rs. 2,
Ordering cost (per order): Rs. 50,
Lead time for replenishment: 3 days.
Stock on hand at the beginning of the simulation exercise was 20 units.
Carry out a simulation run over a period of 10 days with the objective of evaluating the
inventory rule: Order 15 units when present inventory plus any outstanding order falls
below 15 units.
You may use random numbers in the sequence of: 0, 9, 1, 1, 5, 1, 8, 6, 3, 5, 7, 1, 2, 9,

using the first number for day one. Your calculation should include the total cost of
operating this inventory rule for 10 days.
3. In a workshop, fitters report at the tool crib after 8.00 a.m. to get the tools where the tool
room attends are present to issue them. The arrival and service time distribution are shown
below.
Arrival distribution Service distribution

Interarrival time (min) Frequency Service time (min) Frequency
3.5-4.5 0.05 3.5-4.5 0.10
4.5-5.5 0.20 4.5-5.5 0.20
5.5-6.5 0.35 5.5-6.5 0.40
6.5-7.5 0.25 6.5-7.5 0.20
7.5-8.5 0.10 7.5-8.5 0.10
8.5-9.5 0.05

Determine the average waiting time of fitter in the queue. What is the average time he
spends in the system? Simulate the queuing phenomenon for a period of 2 hours. Also ,
find the average waiting time in non-empty queue and per cent time the cool room
attendant remains busy in this period.
4. Consider an inventory control problem in which demand during lead time as well as
lead time distributions are given in table. The reorder point is 6 units reorder
quantity is 12 units. If the ordering cost is Rs.100/order, inventory carrying cost is
Rs.4/unit/week and the shortage cost is Rs. 60/unit/week, find the total inventory
cost for 15 weeks. Assume an initial inventory of 10 units.
Demand Probability Lead time (weeks) Probability

0 0.10 2 0.20
1 0.45 3 0.65
2 0.30 4 0.15
3 0.15
Assume the following random numbers for demand: 49, 67, 06, 30, 95, 01, 10, 70,
80, 66, 69, 76, 86, 56 and 84. Also, assume the following random numbers for lead
time: 84, 79, 15, 03.
5. A book store wishes to carry a particular book in stock. The demand of the book is
not certain and there is a lead time of 2 days for stock replenishment. The
probabilities of demand are given below:
Demand (units/day): 0 1 2 3 4
Probability: 0.05 0.10 0.30 0.45 0.10
Each time an order is placed, the store incurs an ordering cost of Rs . 10 per order. The
store also incurs a carrying cost of Rs. 0.5 per book per day. The inventory carrying
cost is calculated on the basis of stock at the end of each day. The manager of the book
store wishes to compare two options for his inventory decision .
A: Order 5 books when the present inventory plus any outstanding order falls
below 8 books.
B: Order 8 books when the present inventory plus any outstanding order falls
below 8 books.
Currently (beginning of 1 day) the store has a stock of 8 books plus 6 books ordered
two days ago and are expected to arrive the next day. Carryout simulation runs for 10
days to recommend an appropriate option. You may use random numbers in the
sequences, using the first number for day one 89, 34, 78, 63, 61, 81, 39, 1 6, 13, 73.

KEY WORDS
Monte Carlo method, simulation, random number, frequency, inventory.
REFERENCES
BOOKS
& COMPANY LTD

PRAGATI EDITION
MOOCS
YOUTUBE VIDEOS
1) https://youtu.be/6E7q79tpQu0
2) https://youtu.be/uYfgzaLDD8I
3) https://youtu.be/8REqfDLmsVU
WIKIPEDIA
https://en.wikipedia.org/wiki/Monte_Carlo_method
https://analystprep.com/cfa-level-1-exam/quantitative-methods/monte-carlo-
simulation-applications/
OER

UNIT 04-03: GENERATION OF RANDOM NUMBERS
LEARNING OBJECTIVES
 Understand the concept of random number s.
 Learn the 𝑀𝑖𝑑 − 𝑠𝑞𝑢𝑎𝑟𝑒 𝑚𝑒𝑡ℎ𝑜𝑑 of generating pseudo-random number.
 Find pseudo-random number using 𝑀𝑖𝑥𝑒𝑑 𝐶𝑜𝑛𝑔𝑟𝑢𝑒𝑛𝑐𝑒 𝑀𝑒𝑡ℎ𝑜𝑑.
 Find pseudo-random number using 𝑀𝑢𝑙𝑡𝑖𝑝𝑙𝑖𝑐𝑎𝑡𝑖𝑣𝑒 𝐶𝑜𝑛𝑔𝑟𝑢𝑒𝑛𝑡𝑖𝑎𝑙 𝑀𝑒𝑡ℎ𝑜𝑑.
 Find pseudo-random number using 𝐴𝑑𝑑𝑖𝑡𝑖𝑣𝑒 𝐶𝑜𝑛𝑔𝑟𝑢𝑒𝑛𝑡𝑖𝑎𝑙 𝑀𝑒𝑡ℎ𝑜𝑑.
INTRODUCTION
In the previous section we discuss the Monte–Carlo method for solving
inventory problem, queuing problem, maintenance problem, investment and budgeting
problem etc. While solving these we have given the random numbers in the problem. In
this section, we have to study concept of random number and methods of finding
random numbers.
Definition: Random number is a number in a sequence of numbers whose probability

of occurrence is the same as that of any other number in that sequence.
Definition: Random numbers which are generated by some deterministic process but
which satisfy statistical test for randomness are called Pseudo-random numbers.
Random number can be obtaining from the table of random number which is
stored in the memory of the computer. In this case , the process occupies the large
portion of the computer memory. In this process , while solving problem only one
random number table is used and this is objectionable. Hence, we have to use some
different formulae or method for finding random numbers. The necessary formulae
occupy relatively little space. We discuss some method of generating random numbers.
03-01: MID-SQUARE METHOD

Mid-Square method of generating pseudo-random numbers we have to discuss
here. Steps of Mid-square method
Step 1: A four-digit number is taken.
Step 2: Squaring a four-digit number which is taken in step 1 a high digit figure is
obtained.
Step 3: From the answer of step 2 middle four digits are picked up.

Step 4: Step3 yields second random number.
Step 5: Repeat steps and obtain a sequence of pseudo-random numbers.
SOLVED PROBLEMS 01
Problem 04-03-01: Find the sequence of 9 random numbers using Mid-square method.
Take𝑥 = 2421.
Solution: We have given that 𝑥 = 2421.
Sr. No. Random numbers Square of random number

1 2421 05861241
2 8612 74166544
3 1665 02772225
4 7722 59629284
5 6292 39589264
6 5892 34715664
7 7156 51208336
8 2083 04338889
9 3388 -
Thus, the sequence of random number is 2421, 8612, 1665, 7722, 6292, 5892, 7156,
2083 and 3388.
Problem 04-03-02: Discuss the situations may occurring in using Mid-square method.
Solution: In this method, one may come across the following situations :
1) A series may vanish because random number obtained is 0000.
2) A random number reproduces itself. 𝑒. 𝑔. 𝑥 = 4600, 𝑥 = 7600, 𝑥 = 7600.
3) A loop occurs, 𝑒. 𝑔. 𝑥 = 6100, 𝑥 = 2100, 𝑥 = 4100, 𝑥 = 8100 and the process

continues in this situation.
SELF-TEST 01
MCQ 03-01: Which of the following situation may occur in Mid -square method of
generating random numbers?
a) Series may vanish.
b) A random number reproduces.
c) A loop can occur.
MCQ 03-02: While processing the Mid-square method we have to consider
a) Four-digit number.

b) Three-digit number.
c) Two-digit number.
d) One digit number.

SAQ 04-03-01: Find the sequence of 10 random numbers using Mid-square method.
Take𝑥 = 5623.
Solution: We have given that 𝑥 = 5623.
Sr. No. Random numbers Square of random number

1 5623 31618129
2 6181 38204761
3 2047 04190206
4 1902 03617604
5 6176 38142976
6 1429 02042041
7 0420 176400
8 7640 58369600
9 3696 -
Thus, the sequence of random number is 5623, 6181, 2047, 1905, 6176, 1429, 0420,
7640, and 3696.
SAQ 04-03-02: What are the steps of Mid-square method?
Solution: Steps of Mid-square method are as follows:
Step 1: A four-digit number is taken.
Step 2: Squaring a four-digit number which is taken in step 1 a high digit figure is
obtained.
Step 3: From the answer of step 2 middle four digits are picked up.
Step 4: Step3 yields second random number.
Step 5: Repeat steps and obtain a sequence of pseudo-random numbers.
03-02: MIXED CONGRUENCE METHOD:

Pseudo-random number may also be generated by some arithmetic operations.
There are so many methods are given in the literature but most commonly used method
is congruence method or the residue method. Thi s method is described by the
expression 𝑟 = (𝑎𝑟 + 𝑏)(𝑚𝑜𝑑𝑢𝑙𝑜 𝑚), …………………(*)

where 𝑎, 𝑏 𝑎𝑛𝑑 𝑚 are constant and 𝑟 , 𝑟 are the 𝑖 and (𝑖 + 1) random numbers.
Then, 𝑟 is the reminder or residue. For finding sequenc e of pseudo random numbers
we must know the values of 𝑎 , 𝑏 𝑎𝑛𝑑 𝑚 and also 𝑟 . The 𝑟 is any random number and
is called 𝑠𝑒𝑒𝑑. Finding pseudo-random number sequence using expression (*) is called
Mixed Congruence method.
SOLVED PROBLEMS 02
Problem 04-03-03: Use the mixed congruential method to generate a sequence of 10
pseudo random number such that 𝑟 = (16𝑟 + 18)(𝑚𝑜𝑑𝑢𝑙𝑜 23), where 𝑟 = 1.
Solution: We have to find string of random number using mixed congruential method as
follows:
𝑖 𝑟 𝑟 = (16𝑟 + 18)(𝑚𝑜𝑑𝑢𝑙𝑜 𝑚) Random number

0 16 × 1 + 18 32
1 = = 1 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 11 11
23 23
1 16 × 11 + 18 194
11 = = 8 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 10 10
23 23
2 16 × 8 + 18 178
10 = = 7 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 117 17
23 23
3 16 × 17 + 18 290
17 = = 12 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 14 14
23 23
4 16 × 14 + 18 242
14 = = 10 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 112 12
23 23
5 16 × 12 + 18 210
12 = = 9 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 3 3
23 23
6 16 × 3 + 18 66
3 = = 2 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 20 20
23 23
7 16 × 20 + 18 338
20 = = 14 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 16 16
23 23
8 16 × 16 + 18 274
16 = = 11 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 21 21
23 23
9 16 × 21 + 18 354
21 = = 15 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 9 9
23 23
The string of random number obtained is 1, 11, 10, 17, 14, 12, 3, 20, 16, 21, 9 after
which it is repeating.
Problem 04-03-04: Use the mixed congruential method to generate a sequence of 5

pseudo random number such that 𝑟 = (3𝑟 + 6)(𝑚𝑜𝑑𝑢𝑙𝑜 10) where 𝑟 = 1.
follows:
𝑖 𝑟 𝑟 = (3𝑟 + 6)(𝑚𝑜𝑑𝑢𝑙𝑜 10) Random number

3×1+6 9
0 1 = = 0 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 9 9
10 10
3 × 9 + 6 33
1 9 = = 3 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 3 3
10 10

3 × 3 + 6 15
2 3 = = 1 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 5 5
10 10
3 × 5 + 6 21
3 5 = = 2 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 1 1
10 10
3×1+6 9
4 1 = = 0 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 9 9
10 10
The string of random number obtained is 1, 9, 3, 5, 1 and 9 after which it is repeating.
SELF-TEST 02
MCQ 03-03: A sequence of 5 pseudo random number such that 𝑟 = (3𝑟 +
6)(𝑚𝑜𝑑𝑢𝑙𝑜 10), where 𝑟 = 1.
a) 1, 9, 3, 5, 1 and 9
b) 1, 9, 3, 5, 1 and 5
c) 1, 9, 3, 4, 1 and 9
d) 1, 1, 3, 9, 1 and 9
MCQ 03-04: While finding pseudo random number sequence the random number 𝑟 is
called
a) Seed.
b) Non-seed.
c) Origin.

SAQ 04-03-03: Use the mixed congruential method to generate a sequence of five two-
digit random number such that 𝑟 = (21𝑟 + 53)(𝑚𝑜𝑑𝑢𝑙𝑜 100), where𝑟 = 46.
follows:
Random
𝑖 𝑟 𝑟 = (21𝑟 + 53)(𝑚𝑜𝑑𝑢𝑙𝑜 100)
number
21 × 46 + 53 1019
0 46 = = 10 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 19 19
100 100
21 × 19 + 53 452
1 19 = = 4 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 52 52
100 100
21 × 52 + 53 1145
2 52 = = 11 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 45 45
100 100

21 × 45 + 53 998
3 45 = = 9 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 98 98
100 100
21 × 98 + 53 2111
4 98 = = 21 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 11 11
100 100
21 × 11 + 53 284
5 11 = = 2 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 84 84
100 100
The string of random number obtained is 46, 19, 52, 45, 98, 11, 84.
SAQ 04-03-04: Explain the Mixed Congruence method?
Solution: Pseudo-random number may also be generated by some arithmetic operations.

is congruence method or the residue method. This method is described by the
Then 𝑟 is the reminder or residue. For finding sequence of pseudo random numbers,
is called 𝑠𝑒𝑒𝑑. Finding pseudo-random number sequence using expression (*) is called
Mixed Congruence method.
03-03: ADDITIVE CONGRUENTIAL METHOD

If 𝑎 = 1 , expression (*) in section 03.03 is reduces to additive type
𝑟 = (𝑟 + 𝑏)(𝑚𝑜𝑑𝑢𝑙𝑜 𝑚) , Where b and m are constant and 𝑟 and 𝑟 are 𝑖 and

(𝑖 + 𝑖) random number.
SOLVED PROBLEMS 03
Problem 04-03-05: Use the additive congruential method to generate a sequence of 10
pseudo random number such that 𝑟 = (𝑟 + 18)(𝑚𝑜𝑑𝑢𝑙𝑜 23), where𝑟 = 1.
follows:
𝑖 𝑟 𝑟 = (𝑟 + 18)(𝑚𝑜𝑑𝑢𝑙𝑜 23) Random number

0 1 + 18 19
1 = = 0 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 19 19
23 23
1 19 + 18 37
19 = = 1 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 14 14
23 23
2 14 + 18 32
14 = = 1 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 9 9
23 23
3 9 + 18 27
9 = = 1 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 4 4
23 23

4 4 + 18 22
4 = = 0 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 22 22
23 23
5 22 + 18 40
22 = = 1 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 17 17
23 23
6 17 + 18 35
17 = = 1 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 12 12
23 23
7 12 + 18 30
12 = = 1 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 7 7
23 23
8 7 + 18 25
7 = = 1 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 2 2
23 23
9 2 + 18 20
2 = = 0 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 20 20
23 23
The string of random number obtained is 1, 19, 14, 9, 4, 22, 17, 12, 7, 2, 20, 15, 10,
5, 0, 18, 13, 8, 3, 21, 16, 11, 6 and 1 after which it is repeating.
Problem 04-03-06: Use the additive congruential method to generate a sequence of 5

pseudo random number such that 𝑟 = (𝑟 + 6)(𝑚𝑜𝑑𝑢𝑙𝑜 10), where 𝑟 = 1.
Solution: We have to find string of random number using additive congruential method
as follows:

1+6 7
0 1 = = 0 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 7 7
10 10
7 + 6 13
1 7 = = 1 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 3 3
10 10
3+6 9
2 3 = = 0 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 9 9
10 10
9 + 6 15
3 9 = = 1 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 5 5
10 10
5 + 6 11
4 5 = = 1 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 1 1
10 10
The string of random number obtained is 7, 3, 9, 5 and 1 after which it is repeating.
SELF-TEST 03
MCQ 03-05: A sequence of 5 pseudo random number such that 𝑟 = (𝑟 +
6)(𝑚𝑜𝑑𝑢𝑙𝑜 10), where 𝑟 = 1.
a) 7, 3, 1, 5 and 9
b) 7, 3, 9, 5 and 1
c) 7, 3, 5, 9 and 1
d) 7, 4, 9, 3 and 1
MCQ 03-06: The expression of additive congruential method is obtained from mixed
congruential method by putting
a) 𝑎 = 1

b) 𝑎 = 0
c) 𝑏 = 0
d) 𝑏 = 1

SAQ 04-03-05: Use the additive congruential method to generate a sequence of five
two-digit random number such that 𝑟 = (𝑟 + 91)(𝑚𝑜𝑑𝑢𝑙𝑜 100), where 𝑟 = 46.
Solution: We have to find string of random number using additive congruential method
as follows:
0 46 + 91 137
46 = = 1 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 37 37
100 100
1 37 + 91 128
37 = = 1 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 28 28
100 100
2 28 + 91 119
28 = = 1 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 19 19
100 100
3 19 + 91 110
19 = = 1 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 10 10
100 100
4 10 + 91 101
10 = = 1 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 1 1
100 100
The string of random number obtained is 46, 37, 28, 19, 10 and 1 after which it is
repeating.
SAQ 04-03-04: Explain the additive Congruence method?

Then 𝑟 is the reminder or residue. For finding sequence of pseudo random num bers,
is called 𝑠𝑒𝑒𝑑. If 𝑎 = 1 , expression (*) is reduces to additive type 𝑟 = (𝑟 +
𝑏)(𝑚𝑜𝑑𝑢𝑙𝑜 𝑚) , Where b and m are constant and 𝑟 and 𝑟 are 𝑖 and (𝑖 + 𝑖) random
number.

03-04: MULTIPLICATIVE CONGRUENTIAL METHOD
If 𝑏 = 0 , expression (*) in section 03.03 is reduces to Multiplicative type
𝑟 = 𝑎𝑟 (𝑚𝑜𝑑𝑢𝑙𝑜 𝑚) , Where b and m are constant and 𝑟 and 𝑟 are 𝑖 and (𝑖 + 𝑖)

random number.
SOLVED PROBLEMS 04
Problem 04-03-07: Use the multiplicative congruential method to generate a sequence
of 10 pseudo random number such that 𝑟 = 16𝑟 (𝑚𝑜𝑑𝑢𝑙𝑜 23), where 𝑟 = 1.
Solution: We have to find string of random number using multiplicative congruential

method as follows:
𝑖 𝑟 𝑟 = 16𝑟 (𝑚𝑜𝑑𝑢𝑙𝑜 23) Random number

16 × 1 16
0 1 = = 0 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 16 16
23 23
16 × 16 256
1 16 = = 11 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 3 3
23 23
16 × 3 48
2 3 = = 2 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 2 2
23 23
16 × 2 32
3 2 = = 0 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 9 9
23 23
16 × 9 144
4 9 = = 6 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 6 6
23 23
16 × 6 96
5 6 = = 3 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 4 4
23 23
16 × 4 64
6 4 = = 2 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 18 18
23 23
16 × 18 288
7 18 = = 12 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 12 12
23 23
16 × 12 192
8 12 = = 8 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟8 8
23 23
16 × 8 128
9 8 = = 5 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 13 13
23 23
The string of random number obtained is 1, 16, 3, 2, 9, 6, 4, 18, 12, 8, 13 and 1 after
which it is repeating.
Problem 04-03-08: Use the multiplicative congruential method to generate a sequence

of 5 pseudo random number such that 𝑟 = 3𝑟 (𝑚𝑜𝑑𝑢𝑙𝑜 10), where 𝑟 = 1.

method as follows:

0 3×1 3
1 = = 0 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 3 3
10 10
1 3×3 9
3 = = 0 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 9 9
10 10

2 3 × 9 27
9 = = 2 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 7 7
10 10
3 3 × 7 21
7 = = 2 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 1 1
10 10
4 3×1 3
1 = = 0 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 3 3
10 10
The string of random number obtained is 1, 3, 9, 7, 1 and 3 after which it is repeating.
SELF-TEST 04
MCQ 03-07: A string of pseudo random number may not be
a) A loop.
b) Reproductive.
c) Sequence of distinct random number.
MCQ 03-08: A sequence of 5 pseudo random number such that 𝑟 = (3𝑟 )(𝑚𝑜𝑑𝑢𝑙𝑜 10),
where𝑟 = 1.
e) 1, 3, 7, 9, 1 and 3
f) 1, 3, 3, 7, 1 and 3
g) 1, 3, 9, 7, 1 and 3
h) 1, 9, 7, 3, 1 and 3

SAQ 04-03-07: Generate a sequence of 5 three digits random numbers such that𝑟 =
301𝑟 (𝑀𝑜𝑑𝑢𝑙𝑜 1000)𝑎𝑛𝑑 𝑟 = 501.

method as follows:
301 × 501 150801

0 501 = = 150 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 801 801
1000 1000
301 × 801 241101
1 801 = = 241 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 101 101
1000 1000
301 × 101 30401
2 101 = = 30 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 401 401
1000 1000
301 × 401 120701
3 401 = = 120 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 701 701
1000 1000

301 × 701 211001
4 701 = = 211 + 𝑟𝑒𝑚𝑎𝑖𝑛𝑑𝑒𝑟 001 1
1000 1000
The string of random number obtained is 501, 801, 101, 401, 701 and 1.
SAQ 04-03-08: Explain the multiplicative Congruence method?

There are so many methods are given in the literature but most commonly used met hod
expression 𝑟 = (𝑎𝑟 + 𝑏)(𝑚𝑜𝑑𝑢𝑙𝑜 𝑚), ………………… (*)
Then 𝑟 is the reminder or residue. For finding sequence of pseudo random numbers ,
is called 𝑠𝑒𝑒𝑑. If 𝑏 = 0 , expression (*) is reduces to Multiplicative type 𝑟 =
𝑎𝑟 (𝑚𝑜𝑑𝑢𝑙𝑜 𝑚) , Where b and m are constant and 𝑟 and 𝑟 are 𝑖 and (𝑖 + 𝑖) random
number.
SUMMARY
In this section, we discuss method of finding random numbers which is used for
solving simulation problem using Monte Carlo’s method. For finding random numbers
we use Mid-Square Method, Additive Congruence method, Multiplicative Congruence
method and Mixed Congruence method.

1. With the help of example explain the Mid-square method of generating random
number.
2. With the help of example explain the mixed congruential method of generating
random number.
3. With the help of example explain the additive congruential method of generating
random number.
4. With the help of example explain the Multiplicative congruential method of

generating random number.
5. Explain the residue method of generating random number.
6. In the mixed congruence method of generation of random number, a random

number 𝑟 + 1 is given by 𝑟 = (𝑎𝑟 + 𝑏)(𝑚𝑜𝑑𝑢𝑙𝑜 𝑚) , where a, b and c are

constant. Generate 20 random numbers taking 𝑟 = 24, 𝑎 = 51, 𝑏 = 59 𝑎𝑛𝑑 𝑚 =
93.
7. In the additive congruence method of generation of random number, a random

number 𝑟 + 1 is given by 𝑟 = (𝑟 + 𝑏)(𝑚𝑜𝑑𝑢𝑙𝑜 𝑚) , where a, b and c are
constant. Generate 20 random numbers taking 𝑟 = 24, 𝑎 = 51, 𝑏 = 59 𝑎𝑛𝑑 𝑚 =
93.
8. In Multiplicative congruence method of generation of random number, a random

number 𝑟 + 1 is given by𝑟 = 𝑎𝑟 (𝑚𝑜𝑑𝑢𝑙𝑜 𝑚), where a, b and c are constant.
Generate 20 random numbers taking 𝑟 = 24, 𝑎 = 51 𝑎𝑛𝑑 𝑚 = 93.
KEY WORDS
Random number, congruence, residue, modulo
REFERENCES
BOOKS
& COMPANY LTD

PRAGATI EDITION
MOOCS
YOUTUBE VIDEOS
1) https://youtu.be/aSlkVy3mbR0
2) https://youtu.be/d7C6MJ49lMM
3) https://youtu.be/GtOt7EBNEwQ
WIKIPEDIA
https://en.wikipedia.org/wiki/Random_number_generation
OER

UNIT 04-04: SIMULATION LANGUAGES
LEARNING OBJECTIVES
 Learn the different simulation languages for designing simulation programme.
INTRODUCTION
A computer simulation language is used to describe the operation of a
simulation on a computer. There are two major types of simulation: continuous and
discrete event though more modern languages can handle more complex combinations.
Most languages also have a graphical interface and at least a simple statistic gathering
capability for the analysis of the results. We discuss some of the simulation languages
in this section.
04-01: SIMULATION LANGUAGES

The performance of programming and execution of a simulation assignment
relies upon the programming language used. Further to the overall motive languages
including FORTRAN and PL 1, a huge variety of specialized laptop languages are used
to put in writing simulation programmes.
FORTRAN:
 Standing for FORmula TRANslation.
 It is introduced by IBN in 1957.
 The FORTRAN, being incredibly preferred in the nature, may be used for any
simulation assignment.
 Being well-known and usually available on computer systems, F ORTRAN is

quiet frequently writing the simulation programmes.
 FORTRAN is generally considered to be extra efficient in computer time and

storage requirement.
 However, programming in FORTRAN is greater tough and time consuming, as

compared to important simulation languages.
 While the complexities of the simulation assignment boom, bookkeeping of the

difficult info of the simulation turns into tough and this makes the programming
in the FORTRAN more difficult.

Consequently, for sensible situations, simulation programs ought to be
written in specialized simulation language, which might be designed to fulfill
the following objectives:
To with no trouble describe the factors which commonly appear in simulation,

which includes the era of the random variates for most of the statistical
distributions.
Flexibility of changing the layout configuration of the machine which will

recollect change configuration.
Internal timing and manage mechanism for bookkeeping of the important data
all through the simulation run.
To achieve effortlessly the data and information about the conduct of the device.
To provide easy operational technique, along with altering the initial nation of
the device and form of output records to be generated, etc.
GASP and SIMSCRIPT are widely used widespread simulation language,

which can easily do the activity of FORTRAN or PL 1. These are FORTRAN
base language and consequently the expertise of FORTRAN is prerequisite for
gaining knowledge of GASP and SIMSCRIPT.
GASP (General purposed simulation language):
 GPPS is the most commonly used simulation language which was developed by
IBM.
 It is easy to learn and incorporates all the features which are unique to
simulation.
 It is problem-oriented language, but has a wide range of applications.
 It employs the next event incrementing time flow mechanism and uses integral
time units.
 The calculation in integer arithmetic helps to keep the round off error to
minimum.
 In this language, the system to be simulated is flow charted in the form of

bocks diagram, and the blocks are then written in GPSS statement.
SIMSCRIPT:
 It was developed in early 1960’s by RAND corporation of USA.

 This language depends neither on any predefined coding forms nor any
intermediate language.
 The advantages of SIMSCRIPT are that it reduces the time needed to program
simulations of even moderate complexity and provides increased flexibility in
modifying such models in accordance with the findings of preliminary analysis
and other circumstances.
Simulation programming language are very economical in respect of users,

programming time, though they take slightly larger CPU tie in execution of the of the
programme.
Some of the may simulation languages are, DYNAMO, SIMPAC, SIMULATE,

SIMULA, CSMP, GPS, ESP and CSL.
SOLVED PROBLEMS 01
Problem 04-04-01: Discuss the simulation language FORTRAN.
Solution: FORTRAN:
 Standing for FORmula TRANslation.
 It is introduced by IBN in 1957.
 The FORTRAN, being incredibly preferred in the nature, may be used for any
simulation assignment.
 Being well-known and usually available on computer systems, FORTRAN is quiet

frequently writing the simulation programmes.
 FORTRAN is generally considered to be extra efficient in computer time and

storage requirement.
 However, programming in FORTRAN is greater tough and time consuming, as

compared to important simulation languages.
 While the complexities of the simulation assignment boom, bookkeeping of the

difficult info of the simulation turns into tough and this makes the programming in
the FORTRAN more difficult.
Problem 04-04-02: State the different languages used for simulation?
Solution: There are so many simulation languages like FORTRAN, PL 1, SIMSCRIPT,

GASP, DYNAMO, SIMPAC, SIMULATE, SIMULA, CSMP, GPS, ESP and CSL, etc.

SELF-TEST 01
MCQ 04-01: Which of the following is not the special purpose simulation language?
a) BASIC.
b) SIMPAC.
c) GASP.
d) SIMPSCRIPT.
MCQ 04-02: Special simulation languages are useful because they
a) reduce programme preparation time and cost .
b) have the capability to generate random variables .
c) require no prior programming knowledge.

SAQ 04-04-01: Discuss some GASP language of simulation.
Solution: GASP (General purposed simulation language):
 GPPS is the most commonly used simulation language which was developed by
IBM.
 It is easy to learn and incorporates all the features which are unique to
simulation.
 It is problem-oriented language, but has a wide range of applications.
 It employs the next event incrementing time flow mechanism and uses integral
time units.
 The calculation in integer arithmetic helps to keep the round off error to
minimum.
In this language, the system to be simulated is flow charted in the form of bocks
diagram, and the blocks are then written in GPSS statement.
SAQ 04-04-02: Explain the simulation language SIMSCRIPT.
Solution: SIMSCRIPT:
 It was developed in early 1960’s by RAND corporation of USA.
 This language depends neither on any predefined coding forms nor any
intermediate language.

 The advantages of SIMSCRIPT are that it reduces the time needed to program
simulations of even moderate complexity and provides increased flexibility in
modifying such models in accordance with the findings of preliminary analysis
and other circumstances.
SUMMARY
In this section, we discuss the simulation languages which are required to write
simulation program. Some of the simulation languages are FORTRAN, PL1,
SIMSCRIPT, GASP, DYNAMO, SIMPAC, SIMULATE, SIMULA, CSMP, GPS, ESP
and CSL, etc.

1. Write note on simulation languages.
2. Explain three simulation languages of simulation.
KEY WORDS
FORTRAN, PL1, SIMSCRIPT, GASP, DYNAMO, SIMPAC, SIMULATE, SIMULA,
CSMP, GPS, ESP and CSL.
REFERENCES
BOOKS
& COMPANY LTD
MOOCS
YOUTUBE VIDEOS
1) https://youtu.be/RetcEFsAJ0g
2) https://youtu.be/TprGoTrXz4Y
WIKIPEDIA
http://www.simscript.com/
https://en.wikipedia.org/wiki/Simulation_language
OER

A NSWERS TO S ELF -T ESTS AND E ND OF U NIT E XERCISES
UNIT 01-01: BASICS OF OPERATIONS RESEARCH

SELF-TEST
MCQ 01-01: (c) ; MCQ 01-02: (a) ; MCQ 01-03 : (c); MCQ 01-04 (d);
MCQ 01-05: (c); MCQ 01-06:(d); MCQ 01-07: (d); MCQ 01-08:(c);

1 to 10: Theory.
UNIT 01-02: LINEAR PROGRAMMING PROBLEMS (LPP)

SELF-TEST
MCQ 02-01: (d) ; MCQ 02-02: (d) ; MCQ 02-03: (d); MCQ 02-04: (d);
MCQ 02-05: (a); MCQ 02-06:(c); MCQ 02-07: (b) ; MCQ 02-08:(a);

1: Multiple solutions.
2:𝑥 = 6 , 𝑦 = 12 𝑎𝑛𝑑 𝑀𝑖𝑛 𝑍 = 240
3: 𝑥 = 0 , 𝑥 = 5 𝑎𝑛𝑑 𝑍 = 5.
4: Multiple optimal solution.
5: Unbounded solution.
6: First formulate the given problem then solve it by graphical method.
𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑒 𝑍 = 20𝑥 + 10𝑥 subject to the constraints 3𝑥 + 2 𝑥 ≤ 34 , 2𝑥 + 𝑥 ≥ 18
and 𝑥 , 𝑥 𝑒𝑎𝑐ℎ ≥ 0 and 𝑥 = 10, 𝑥 = 2 𝑎𝑛𝑑 𝑀𝑎𝑥 𝑍 = 560.
7:𝑀𝑖𝑛𝑖𝑚𝑖𝑧𝑒 𝑍 = 3𝑥 + 2𝑦 subject to constraint
5𝑥 + 𝑦 ≥ 10, 2𝑥 + 2𝑦 ≥ 12, 𝑥 + 4𝑦 ≥ 12 𝑎𝑛𝑑 𝑥 ≥ 0, 𝑦 ≥ 0.
8: First formulate the given problem then solve it by graphical method.
𝑀𝑖𝑛 𝑍 = 𝑥 + 𝑥 subject to the constraints 5𝑥 + 10 𝑥 ≤ 50 , 𝑥 + 𝑥 ≥ 1𝑥 ≤ 4 and

𝑥 , 𝑥 𝑒𝑎𝑐ℎ ≥ 0 and it has multiple solution 𝑀𝑖𝑛 𝑍 = 1.
UNIT 01-03: THE SIMPLEX METHOD:

SELF-TEST
MCQ 03-01: (b); MCQ 03-02: (c) ; MCQ 03-03 : (d); MCQ 03-04 (a);

MCQ 03-05: (b); MCQ 03-06:(d); MCQ 03-07: (a) ; MCQ 03-08:(d) .
1: 𝑥 = 0 , 𝑥 = 450 𝑎𝑛𝑑 𝑀𝑎𝑥 𝑍 = 1800
2: 𝑥 = 7, 𝑦 = 5 𝑎𝑛𝑑 𝐶 = 81.
3: 𝑥 = 200, 𝑥 = 600 𝑎𝑛𝑑 𝑀𝑎𝑥 𝑍 = 𝑅𝑠. 2600.
4: 𝑥 = 3, 𝑥 = 0 𝑎𝑛𝑑 𝑍 = 9.5
5: 𝑥 = , 𝑥 = , 𝑥 = 0 𝑎𝑛𝑑 𝑍 = 44.5.
6: 𝑥 = ,𝑥 = , 𝑎𝑛𝑑 𝑍 =
7: 𝑥 = , 𝑦 = 5, 𝑧 = 0, 𝑍 = .
8: 𝑥 = ,𝑥 = 𝑎𝑛𝑑 𝑀𝑖𝑛 𝑍 = 𝑅𝑠.
UNIT 01-04: DUALITY

SELF-TEST
MCQ 04-01: (c) ; MCQ 04-02: (d) ; MCQ 04-03 : (a); MCQ 04-04 (d);
MCQ 04-05: (d); MCQ 04-06:(c); MCQ 04-07: (c) ; MCQ 04-08:(d) .

1: Theory
2: Theory
3: Theory
4: Min = 3y + 4y + y + 6y subject to 5y – 2y + y – 3y ≥ 2
6y + y – 5y – 3y ≥ 5, – y + 4y + 3y + 7y ≥ 6, and y , y , y , y ≥ 0.
5: Max = 2y + 3y – 5y subject to 2y + 3y – y ≤ 2 ,3y + 3y – 4y ≤ 3,5y +

7y – 6y = 4 and y , y ≥ 0 and y unrestricted.
6:x = 5, x = 7 , = 114
7: primal 𝑥 = 0, 𝑥 = 40, 𝑀𝑎𝑥 𝑍 = 𝑅𝑠. 200
Dual 𝑦 = 0, 𝑦 = 5 𝑎𝑛𝑑 𝑀𝑖𝑛 𝑤 = 𝑅𝑠. 200

8: Theory.
UNIT 02-01: GAME MODELS AND RELATED THEORY

SELF-TEST
MCQ 01-01:(b); MCQ 01-02: (b) ; MCQ 01-03: (a); MCQ 01-04: (b);
MCQ 01-05:(b); MCQ 01-06: (c) ; MCQ 01-07: (b); MCQ 01-08 : (c).

1: Maximin = Minimax = 5, Value of game = 5, optimal strategies = (𝐴 , 𝐵 ).
2:−1 ≤ 𝑣𝑎𝑙𝑢𝑒𝑜𝑓𝑔𝑎𝑚𝑒 ≤ 1.
3: Hint: Pay off matrix for column player is negative of pay off matrix for row player.
4: Maximin = Minimax = -5, Value of game = -5.
5: 2 ≤ 𝑣𝑎𝑙𝑢𝑒𝑜𝑓𝑔𝑎𝑚𝑒 ≤ 4.
6:2 ≤ 𝑣𝑎𝑙𝑢𝑒𝑜𝑓𝑔𝑎𝑚𝑒 ≤ 3.
7: Maximin = Minimax = 1, Value of game = 1.
8: Theory.
UNIT 02-02: TWO PERSON ZERO SUM GAME

SELF-TEST
MCQ 02-01:(a); MCQ 02-02: (b) ; MCQ 02-03: (c); MCQ 02-04: (b);
MCQ 02-05:(b); MCQ 02-06: (d) ; MCQ 02-07: (c); MCQ 02-08: (b);
MCQ 02-09:(b); MCQ 02-10: (a).

1, 2, 3: Theory
4:Fair game, Value of game = 0, optimal strategies = (𝐼𝐼, 𝐼𝐼).
5: 𝑝 ≥ 5, 𝑞 ≤ 5.
6:Maximin = Minimax = 6, Saddle point exists, optimal strategies = (2, 3).
7: Maximin = Minimax = 1, Value of game = 1, optimal strategies = (1, 1).
8: Maximin = Minimax = 3, strictly determinable game.

UNIT 02-03: DOMINANCE IN GAMES
SELF-TEST
MCQ 03-01:(a); MCQ 03-02: (b) ; MCQ 03-03: (c); MCQ 03-04: (b);
MCQ 03-05:(a); MCQ 03-06: (b) ; MCQ 03-07: (b); MCQ 03-08: (c);
MCQ 03-09:(a).

1:Value of game = 8, A [0,0,1,0], B [1,0,0,0,0].
2:Value of game = 13.
3:Theory.
4:Value of game = 3.5
5: Value of game = 6
6: without saddle point, Value of game = 4.
7: without saddle point, Value of game = 4.
8: Value of game = 1 unit, both should play their second strategy.
UNIT 02-04: MIXED STRATEGIES (2×N AND M×2 GAMES)

SELF-TEST
MCQ 04-01:(c); MCQ 04-02: (b) ; MCQ 04-03: (a); MCQ 04-04: (b);
MCQ 04-05:(a); MCQ 04-06: (b) ; MCQ 04-07: (b); MCQ 04-08: (b).

1:No saddle point, value of game = 3.5, optimum strategies: A (1/2, 1/2), B (0, ½, ½, 0) or A (1/2,
1/2), B (0, 0,07/8, 1/8)
2:Value of game = 2.5, A (1/2, 1/2), B (0, ½, ½, 0)
3:Value of game = 1/3, A (0, 5/9, 4/9, 0), B (3/9, 6/9)
4:Theory.
5:Theory.
6:Value of game =245/16, A [0, 15/16, 1/16], B [0, 11/16, 0, 5/16].
7: Value of game = 2, A’s optimal strategy [0, 0, 1], B’s optimal strategy [2/5, 3/5, 0].
8:Value of game = 6, strategies: A2, B1.

UNIT 03-01: NETWORK ANALYSIS
SELF-TEST
MCQ 01-01:(b); MCQ 01-02: (a) ; MCQ 01-03: (c); MCQ 01-04: (a);
MCQ 01-05:(c); MCQ 01-06: (b) ; MCQ 01-07: (d); MCQ 01-08: (c);
MCQ 01-09: (d).

1:
2:
Activity A B C D E F G H I J
Successor 2 3 4 5 6 6 6 7 7 8
Predecessor 1 2 2 3 3 4 5 5 6 7
3:Theory.
4:
5: Activity dependence table:
Event Description of event Immediately preceding event

S Drives into garage -
A Tank filled with petrol S
B Oil checked up A
C Top up the oil B

D Adjusting tyre pressures S
E Pays for petrol and oil D, C
F Drives off E
Draw the network diagram using this activity dependence table.
6:Similar to example 2.
7:
8:
UNIT 03-02: FULKERSON ’S RULE

SELF-TEST
MCQ 02-01:(b); MCQ 02-02: (a) ; MCQ 02-03: (c); MCQ 02-04: (d);
MCQ 02-05:(a); MCQ 02-06: (a) ; MCQ 02-07: (b); MCQ 02-08: (a);
MCQ 02-09: (d).

1:
2:

3: a=1, b=2, c=3, d=4, e=5, f=6 g=7, h=8, i=9.
4:Theory.
5: Optimum cost = Rs. 1770, Optimum time = 11 Days.
6:36 days schedule cost is Rs. 12,200.
7: Optimum cost = Rs. 4640, Optimum duration = 16 Days.

8: Theory.
UNIT 03-03: CRITICAL PATH METHOD (CPM)

SELF-TEST
MCQ 03-01:(a); MCQ 03-02: (b) ; MCQ 03-03: (d); MCQ 03-04: (a);
MCQ 03-05:(a); MCQ 03-06: (c) ; MCQ 03-07: (d); MCQ 03-08: (c);
MCQ 03-09: (a).

1: Critical path: A→B→E→F, Duration of project = 19 Days.
2: Hint: Use formulae.
3:
Hint: Use formulae to find floats and hence find critical path.
4: Critical path is A→D→B→C→E→G with distance 29.
5: Critical path is 1→2→3→4→5→8→10→11 with distance 29-time units.

6: The critical path consists of all those activities for which total float is zero. Here, critical path
is 1→3→6→9, with duration of 15 months.
7: Hint: Use formulae.

8: Critical path is the path on which, EST = LFT for each node. Here, critical path is
1→3→7→9, with duration of 15 months.
UNIT 03-04: PROJECT EVALUATION AND REVIEW TECHNIQUE

SELF-TEST
MCQ 04-01:(c); MCQ 04-02: (a) ; MCQ 04-03: (c); MCQ 04-04: (a);
MCQ 04-05:(b); MCQ 04-06: (d) ; MCQ 04-07: (a); MCQ 04-08: (b);
MCQ 04-09: (c).

1: Hint: Use the formulae to find expected task time and variance. Use the expected task time as
duration of activity so as to calculate EST and LFT. Critical path: 1→4→7; Duration= 42.83
weeks.
2: Hint: Use formulae to calculate expected duration and variance.
Critical path: A→C→E→H→K.
7: Theory.
8: Theory.
UNIT 04-01: SIMULATIONS AND IT ’S THEORY

SELF-TEST
MCQ 01-01: (d ) ; MCQ 01-02: (d) ; MCQ 01-03: (d); MCQ 01-04 (b) .

1 to 6: Theory.
UNIT 04-02: MONTE CARLO SIMULATION

SELF-TEST
MCQ 02-01: (d); MCQ 02-02: (b) ; MCQ 02-03: (d); MCQ 02-04: (d);
MCQ 02-05: (c); MCQ 02-06:(d); MCQ 02-07: (a) ; MCQ 02-08:(d)

1: Theory
2: Total daily inventory cost = Daily ordering cost + Daily carrying cost = 150 + 15.60 = Rs.
165.60
3:14 minutes, 61/9 minutes, 89.26%
4: The inventory cost in the 15 weeks Rs. 612

5: Since option B (Rs. 65.5) has a lower total cost than option A (Rs. 67.5), therefore, the
manager should choose option B.
UNIT 04-03: GENERATION OF RANDOM NUMBERS:

SELF-TEST
MCQ 03-01: (d) ; MCQ 03-02: (a) ; MCQ 03-03 : (a); MCQ 03-04: (a);
MCQ 03-05: (b); MCQ 03-06:(a); MCQ 03-07: (c) ; MCQ 03-08:(c) .
1 to 5: Theory
6: 24, 122, 50, 48, 89 and so on.
7: 24, 86, 49, 15, 74 and so on
8: 24, 15, 21, 48, 30 and so on
UNIT 04-04: SIMULATION LANGUAGES

SELF-TEST
MCQ 04-01: (a ) ; MCQ 04-02: (b) .

1, 2: Theory.

Feedback Form
School of Sciences
(Formerly School of Architecture, Science & Technology)
Yashwantrao Chavan Maharashtra Open University,
Nashik – 422 222
FEEDBACK SHEET FOR THE STUDENT
Dear Student,
You have gone through this book, it is time for you to do some thinking for us.
Please answer the following questions sincerely. Your response will help us to
analyse our performance and make the future editions of this book more useful.
Your response will be completely confidential and will in no way affect your
examination results. Your suggestions will receive prompt attention from us.
Please submit your feedback online at this QR Code or at following link

https://forms.gle/rpDib9sy5b8JEisQ9
or email at: director.ast@ycmou.ac.in

or send this filled “Feedback Sheet” by post to above address.
(Please tick the appropriate box)
Pl. write your Program Code Course Code & Name
Style
01. Do you feel that this book enables you to learn the subject independently without
any help from others?
Yes No Not Sure

02. Do you feel the following sections in this book serve their purpose? Write the
appropriate code in the boxes.
Code 1 for “Serve the purpose fully”
Code 2 for “Serve the purpose partially”
Code 3 for “Do not serve any purpose”
Code 4 for “Purpose is not clear”
Warming up Check Point Answer to Check Points
To Begin With Summary References

Objectives Key Words
03. Do you feel the following sections or features, if included, will enhance self - learning
and reduce help from others?
Yes No Not Sure
Index
Glossary
List of “Important Terms Introduced”
Two Colour Printing
Content
04. How will you rate your understanding of the contents of this Book?
Very Bad Bad Average Good Excellent
05. How will you rate the language used in this Book?
Very Simple Simple Average Complicated Extremely Complicated
06. Whether the Syllabus and content of book complement to each other?
Yes No Not Sure
07. Which Topics you find most easy to understand in this book?
Sr.No. Topic Name Page No.
08. Which Topics you find most difficult to understand in this book?
Sr.No. Topic Name Page No.
09. List the difficult topics you encountered in this Book. Also try to suggest how
they can be improved.
Use the following codes:
Code 1 for “Simplify Text”
Code 2 for “Add Illustrative Figures”
Code 3 for “Provide Audio-Vision (Audio Cassettes with companion Book)”
Code 4 for “Special emphasis on this topic in counseling”
Sr.No. Topic Name Page No. Required Action Code
10. List the errors which you might have encountered in this book.
Sr.No. Page Line Errors Possible Corrections

No. No.
11. Based on your experience, how would you place the components of distance
learning for their effectiveness?
Use the following code.
Code 1 for “Most Effective” Code 3 for “Average” Code 5 fo r “Least Effective”
Code 2 for “Effective” Code 4 for “less Effective”
Printed Book Counseling Lab Journal
Audio Lectures Home Assignment YouTube Videos
Video Lectures Lab-Experiment Online Counseling
12. Give your overall rating to this book?
1. 2. 3. 4. 5.
13. Any additional suggestions:
Thank you for your co-operation!

MAT506 Operations Research

Uploaded by

Copyright:

Available Formats

MAT506 Operations Research

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

MAT506 Operations Research

Uploaded by

Copyright:

Available Formats

YASHWANTRAO CHAVAN MAHARASHTRA OPEN UNIVERSITY

V151: M.Sc. Mathematics

Maharashtra Open Operations Research

S25035: Operations Research Page 1

Yashwantrao Chavan Maharashtra Open University

Dear Students, Greetings!!!

- Prof. Dr. E. Vayunandan

S25035: Operations Research Page 3

- Dr. Sunanda More

S25035: Operations Research Page 4

S25035: Operations Research Page 5

 Understand the different definition of operations research.

 Learn characteristics of operation research.

 Understand the scope of operation research in industry, public utilities , etc.

01-01: OPERATIONS RESEARCH AND IT’S CHARACTERISTICS

Without a doubt we can say operation studies is the studies of operation. An

Definitions of Operations Research (O.R.):

S25035: Operations Research Page 6

– Committee on O.R .National Research Council, USA

Characteristics of Operation Research:

1) Interdisciplinary Team approach:

2) Operations Research increases the creative ability of the decision maker:

O.R. provides manager mathematical tools, techniques and various models to

3) Operations Research is a systems approach:

A business or a government organization or a defense organization may be

a) Judgment phase: This segment includes: (i) identity of the actual-existence

S25035: Operations Research Page 8

Problem 01-01-02: Explain the role of operation research in decision making.

b) Selling, Promotion, Marketing & Publicity: It emphasis on selection of the

c) Acquisition, Investigation, Manufacturing and Personnel Management : The

d) All of the above.

MCQ 01-02: Operation research is

a) Increase the creative capabilities of the decision maker.

S25035: Operations Research Page 9

c) Stable the creative capabilities of the decision maker.

d) Not effect on the creative capabilities of the decision maker.

SHORT ANSWER QUESTIONS 01

Solution: Characteristics of O.R. are as follows:

i) Interdisciplinary team approach.

ii) Scientific approach.

SAQ 01-01-02: Define O.R. and discuss O.R. as an interdisciplinary approach.

Solution: Write any one or two definitions of O.R.

Interdisciplinary Team approach: The highly optimized solution is found by a group

01-02: SCOPE OF OPERATIONS RESEARCH

S25035: Operations Research Page 10

3) In Planning for Economic Growth:

In India, we have long-term anticipating consistent financial development. Each

S25035: Operations Research Page 11

Problem 01-01-04: Explain the scope of O.R. in traffic control.

S25035: Operations Research Page 12

a) Operation research used for maximizes profit in the company.

b) Operation research used for minimize the profit in the company.

d) Operation research used for increase the productivity in the company.

MCQ 01-04: Which of the following is not application of operation research?

d) All of the above.

SHORT ANSWER QUESTIONS 02

Solution: Scope of O.R. in public utilities is as follows:

S25035: Operations Research Page 13

01-03: OPERATIONS RESEARCH MODELS

S25035: Operations Research Page 14