A GUIDE TO OPERATIONS

RESEARCH

Prof (Dr) B Dayal

B.Tech, M. Tech, MBA, Ph. D
 A GUIDE
TO
OPERATIONS
RESEARCH

BY:

Prof (Dr) Bhagwat Dayal

B. Tech, M. Tech, MBA, Ph. D

 CHAPTER - 1

INTRODUCTION TO OPERATIONS RESEARCH

1.1 Introduction

Since the advent of industrial revolution, the world has seen a remarkable growth in the
size and complexity of organizations. The artisan’s small shops of an earlier era have
evolved into the billion dollar corporations of today. An integral part of this
revolutionary change has been a tremendous increase in the division of labour and
segmentation of management responsibilities in these organizations. The results have
been spectacular. However, along with its blessings, this increasing specialization has
created new problems, problems that are still occurring in many organizations. One
problem is a tendency for the many components of an organization to grow into
relatively autonomous empires with their own goals and value systems, thereby losing
sight of how their activities and objectives mesh with those of the overall organization.
What is best for one component frequently is detrimental to another, so the components
may end up working at cross purposes. A related problem is that as the complexity and
specialization in an organization increase, it becomes more and more difficult to allocate
the available resources to the various activities in a way that is more effective for the
organization as a whole. These kind of problems and the need to find a better way to
solve them provided the environment for the emergence of operations research
(commonly referred to as OR).

The decision making is a key part of our daily life. The ultimate goal of all decisions is to
maximize benefits and minimize efforts and time. Operation Research (OR) gives
decision makers the power to make effective decisions and improve day-to-day
operations. Decision makers consider all available options, study the outcomes and
estimate the risks.

As the name implies, operations research involves “research on operations”. Thus,

operations research is applied to problems that concern how to conduct and coordinate
the operations (i.e., the activities) within an organization. The nature of an organization is
essentially immaterial, and in fact, OR has been applied extensively in such diverse areas
as manufacturing, transportation, construction, telecommunication, financial planning,
health care, the defence and public services, to name just a few. Therefore, the breath of
application is unusually wide.

1.2 The origins of operations research.

The roots of OR can be traced back many decades, when early attempts were made to use a
scientific approach in the management of organizations. However, the beginning of the
activity called operations research has generally been attributed to the military services
early in World War II. Because of the war effort, there was an urgent need to allocate scarce
resources to the various military operations and to the activities within each operation in an
effective manner. Therefore, the British and then the US military management called upon a
large number of scientists to apply a scientific approach to deal with this and other strategic
and tactical problems. In effect they were asked to do research on (military) operations.
These teams of scientists were the first OR teams. By developing effective methods of using
the new tool of Radar, these teams were instrumental in winning the air battle of Britain.
Through their research on how to better manage convoy and submarine operations, they also
played a major role in winning the battle of North Atlantic. Similar efforts assisted the Island
campaign in the pacific.

When the war ended, the success of OR in the war effort spurred interest in applying OR
outside the military as well. As the industrial boom following the war was running its course,
the problems caused by the increasing complexity and specializations in organizations were
again coming to the forefront. It was becoming apparent to a growing number of people,
including business consultants who had served on or with the OR teams during the war, that
these were basically the same problems that had been faced by the military in a different
context. By the early 1950s, these individuals had introduced the use of OR to a variety of
organizations in business, industries and government. The rapid spread of OR soon followed.

1.3 OR as a Tool in Decision Making.

Decision making is not the headache of only management; rather all of us make decisions.
The essential characteristics of all decisions are:

 Objective to be accomplished.
 Alternative available at your disposal.
 The influencing factors

OR focuses on the mathematical scoring of consequences of a decision aiming to

optimize the use of time, effort and resources, and avoid heavy mistakes and penalties
due to them. The act of obtaining the best results under any given circumstances is
known as optimizing. The key purpose of OR is to do preparatory calculations that aid
the decision making process.

1.4 The Objective

The objective of OR is to provide a scientific basis to the decision makers for solving the
problems involving the interaction of various components of organization, by employing a
team of scientists from different disciplines, all working together for finding a solution which
is the best in the interest of the organization as a whole. The solution thus obtained is known
as an optimal decision.

1.5 Definition of operations research

OR has been described as:

(a) A method
(b) An approach
 (c) A set of technique
(d) A team activity
(e) A combination of many disciplines
(f) An extension of particular discipline
(g) A new discipline
(h) A vocation
(i) Even a religion

Thus, OR is difficult to define as its boundaries are not marked. However, where there is a
problem, be it Simplex or complex, the operations research technique may be applied to find
the solution.

OR is also described as “science of use”; “quantitative common sense” and “scientific

approach to decision making problem”. But none of the descriptions are complete and cannot
be used to define Operations research. Some more improper definitions are as given below.

“Operations Research is the art of giving bad answers to problems, which otherwise have
worse answers” - TL Shatty

“Operations research is a scientific method of providing executive departments with a

quantitative basis for decisions under their control.”

- PM Morse & GE Kimball

“Operations research is defined as the application of scientific methods, techniques and tools
to the operation of a system with optimum solution to the problem.”

- Churchman, Ackoff and Aruoff

“Operations research is the use of scientific methods to provide criteria for decisions
regarding man, machine, and systems involving repetitive operations”

- More generalized definition

“Operations research is defined as the management of time and resources to obtain the
optimum results in an organization as a whole.

- My definition

1.6 Scope of Operations Research

Any problem, either simple or complicated, can use OR techniques to find the best
possible solution. The application of OR in various fields of everyday life is explained
below.

(a) In Defence Operation: In modern warfare, the defence operations are carried out by
three major independent components namely Air Force, Army and Navy. The
activities in each of these components can be further divided in four sub-components
namely: administration, intelligence, operations and training and supply. The
 applications of modern warfare techniques in each of the components of military
organizations require expertise knowledge in respective fields. Furthermore, each
component works to drive maximum gains from its operations and there is always a
possibility that the strategy beneficial to one component may be unfeasible for
another component. Thus, in defence operations, there is a requirement to coordinate
the activities of various components, which gives maximum benefit to the
organization as a whole, having maximum use of components. A team of scientists
from various disciplines come together to study the strategies of different
components. After appropriate analysis of the various courses of actions, the team
selects the best course of action, known as the ‘optimum strategy’.
(b) In Industry: The system of modern industries is so complex that the optimum point
of operation in its various components cannot be intuitively judged by an individual.
The business environment is always changing and any decision useful at one point
may not be so good some time later. There is always a need to check the validity of
decisions continuously against the situations. The industrial revolution with increased
division of labour and introduction of management responsibilities has made each
component an independent unit having their own goals. For example: production
department minimizes the cost of production but maximizes the output. Marketing
department maximizes the output but minimizes cost of unit sales. Finance
department tries to optimize the capital investment and personal department appoints
good people at minimum cost. Thus each department plans its own objectives and all
these objectives of various department or components come to conflict with one
another and may not agree to the overall objectives of the organization. The
application of OR techniques helps in overcoming the difficulty by integrating the
diversified activities of various components to serve the interest of the organization
as a whole efficiently. OR methods in the industry can be applied in the fields of
production, inventory controls, marketing, purchasing, transportation and competitive
strategies.
(c) Planning: In modern times, it has become necessary for every government to have
careful planning, for economic development of the country. OR techniques can be
fruitfully applied to maximize the per capita income, with minimum sacrifice and
time. A government can thus use OR for framing future economic and social policies.
(d) Agriculture: With increase in population, there is a need to increase agriculture
output. But this cannot be done arbitrarily. There are several restrictions. Hence the
need to determine the course of action serving the best under the given restrictions.
This problem can be solved by using the OR techniques.
(e) In Hospitals: OR methods can solve waiting problems in outpatient department of
big hospitals and administrative problems of the hospital organizations.
(f) In Transport: The OR methods can be applied to regulate the arrivals of trains and
processing times, minimize the passenger waiting time and reduce congestion,
formulate suitable transportation policy, thereby reducing the costs and time of trans-
shipment.
 (g) Research and Development: The OR methodologies can be applied for several
purposes like planning, controlling the research work, time plan, and product
introduction.

1.7 Salient Features of Operations Research

Some key features of OR are as follows:

1. It is system oriented. OR scrutinizes the problem from an organization’s perspective.

The results can be optimal for one part of the system, while the same can be
unfavourable for another part of the system.
2. It imbibes an Inter-disciplinary team approach. Since no single individual can
have thorough knowledge of all fast developing scientific know – how, personalities
from different scientific and managerial cadre form a team to solve the problem.
3. It makes use of scientific methods to solve the problems.
4. Increases the effectiveness of a management decision making ability.
5. It makes use of computer to solve large and complex problems.
6. It offers a quantitative solution.
7. It considers the human factor also.

1.8 Phases of Operations Research. The scientific method in OR study generally

involves three phases

1. Judgment Phase: This phase includes the following activities:

• Determination of the operation.

• Establishment of the objectives and values related to the operation.
• Determination of the suitable measures of effectiveness.
• Formulation of the problem related to the objectives.
2. Research Phase. This phase utilizes the following methodologies:
• Operation and data collection for a better understanding of the problems.
• Formulation of hypothesis and model.
• Observation and experimentation to test the hypothesis on the basis of additional data.
• Analysis of the available information and verification of the hypothesis using pre-
established measures of effectiveness.
• Prediction of various results and consideration of alternative methods.
3. Action Phase. It consists of making recommendations for the decision process by
those who first posed the problem for consideration or by any one in a position to make a
decision, influencing the operation in which the problem is occurred.
1.9 Operation Research Models
What is a model?
• A model is an ideal representation, or abstraction of a real life system.
• Objective is to identify significant factors that affect the real life system and their inter-
relationship.
• A model aids the decision making process as it provides a simplified description of
complexities and uncertainties of a problem in a logical structure.
• Major advantage is it does not interfere with real life system.
Classification of Operation Research models
• Physical models.
• Mathematical or symbolic models
• Models by nature of environment
• Models by the extent of generality.
(a) Physical Models
• Include all forms of diagrams, graphs and charts.
• Designed to deal with specific problems.
• Bring out significant factors and inter-relationship in pictorial form so as to facilitate
analysis.
• Iconic Models. Iconic model is an image of an object or system, represented on a
small scale. These models can simulate the actual performance of a product.
• Analog Models. These models are small physical systems that have similar
characteristics and work like an object it represents. E.g., toy
(b) Mathematical Models or Symbolic Models
• Mathematical models employ a set of mathematical symbols to represent the decision
variables of the system. The variables are related by mathematical system. E.g.:
allocation, sequencing, replacement models.
(c) Models by Nature of Environment
• Deterministic Models. Everything is defined and the results are certain. e.g.: EOQ
model
• Probabilistic Models. The input and output variables follow a probability
distribution. E.g.; game theory.
(d) Models by extent of Generality
• General Models. Can be applied in general and does not pertain to one problem
only. e.g.: linear programming
• Specific Models. Applicable under specific conditions only. E.g.; sales response
curve or equation as a function of advertising is applicable to marketing function
alone.
1.10 Methodology of Operation Research
The basic dominant characteristic feature of Operation research is that it employs
mathematical representations or models to analyze the problems. This distinct approach
represents an adaptation of scientific methodology used by physical sciences. The
scientific method translates a real given problem into a mathematical representation
which is solved and re-transformed into the original context.
Steps.

– Definition of the problem. Define the problem of interest and gather the
relevant data.

– Construction of the model. Formulate a mathematical model to represent the

problem.

– Solution of the model. Develop a computer based procedure for deriving

solutions to the problem from the model.

– Validation of the model. Test the model and refine it as needed.

– Implementation of the final result. Prepare for the ongoing application of

the model as prescribed by management.
1.10.1 Definition of the Problem
The root problem should be identified properly and understood. The major aspects are:-
• A description of the goal or objective of the study.
• Identification of the decision alternative to the system.
• Recognition of the limitation, restrictions and requirements of the system.
1.10.2 Construction of the Model
Decide on the most suitable model for representing the system.
• Model should specify quantitative expressions for the objective and the constraints of
the problem in term of decision variables.
• If the resulting model fits into one of the common mathematical models, a convenient
solution may be obtained by using mathematical techniques.
• If the mathematical relationships of the model are too complex to allow analytical
solutions, a simulation model may be more appropriate.
1.10.3 Solution of the Model
Develop a solution for the model and interpret the solution in the context of given
problem.
• Solution implies determination of a specific set of decision variables that would yield
an optimum solution.
• An optimum solution is one which maximizes or minimizes the performance of any
measure in a model subject to the conditions and constraints imposed on the model.

1.10.4 Validation of the Model

• A common method for testing the validity of a model is to compare its performance
with some past data available for the actual system.
• The model will be valid if under similar conditions of input, it can reproduce the past
performance of the system.
• The problems:

– There is no assurance that future performance will continue to duplicate past

behaviour.

– Data may not be available since non-existence of the system previously.

– Since the model is based on careful examination of the past data, it will always
reveal favourable results.
1.10.5 Implementation of the final result
The optimal solution obtained from a model should be applied to improve the
performance of the system and the validity of the solution should be verified under
changing conditions.
• It involves translation of these results into detailed operating instructions issued in an
understandable form to the individuals who will administer and operate the
recommended system.
• The interaction between the operations research team and the operating personnel will
reach its peak in this phase.
1.11 Tools and techniques of Operation Research
• Linear programming. Linear programming is used to find a solution for optimizing
a given objective. The objective may to optimize profit or minimize cost. Both the
objective and constraints must be expressed as linear expression of decision variables.
• Inventory control methods. The production, purchasing and material managers are
always confronted with questions, such as when to buy, how much to buy and how
much to keep in stock. Inventory models aims at optimizing inventory levels.
• Goal programming. In linear programming single objective function is taken into
consideration and other factors are taken as constraints. However, in real life there are
a number of important objective functions. Goal programming has several objective
functions, each having a target value and goal programme models are developed to
minimize deviation from these targets.
• Queuing model. The queuing theory is based on the concept of probability. It
indicates the capability of a given system and changes possible in the system when
you modify the system. In formulating a queuing model all constraints are needed to
be taken into account. There is no minimization or maximization of the objective
function. Therefore, the application of queuing theory cannot be viewed as an
optimization process. it can be used to estimate the required balance between
customer waiting time and the service capability of the system. several alternatives
are needed to be considered first, evaluate them through queuing model, study their
effect on the system and then make a choice. The criteria for evaluation will be
measures of efficiency of the system, such as the average length of the queue,
expected waiting time of the customer and the average time spent by the customer in
the system. In this approach, success primarily depends on the alternatives considered
and not so much on the queuing model developed.
• Transportation model. The transportation model is an important class of the linear
programs. The model studies the minimization of the cost of transporting a
commodity from a number of sources to several destinations. The supply at each
source and the demand at each destination are known. The objective of the model is to
develop an integral transportation schedule that meets all demands from the inventory
at a minimum total transport cost.
The transportation problem involves m sources, each of which has available a i (i = 1,
2, ….. m) units of homogeneous product and n destinations, each of which requires b j
(j = 1, 2, ….n) units of products. Here ai and bj are positive integers. The cost Cij of
transporting one unit of the product from the i th sources to the jth destination is given
for each I and j. it is assumed that the total supply and the total demand are equal.
Σi=1m ai = Σj=1n bj
The condition is guaranteed by creating either a fictitious destination with a demand
equal to the surplus if total demand is less than the total supply or a dummy source
with a supply equal to the shortage if total demand exceeds total supply. The cost of
transportation from the fictitious destination to all sources and from all destinations to
the fictitious sources are assumed to be zero so that total cost of transportation will
remain the same.
• The tools namely, queuing models, sequence models, transportation and assignment
model, network analysis are also tools of operational research.
1.12 Structure of mathematical model
Many industrial and business situations are concerned with planning activities. In each
case of planning, there are limited resources, such as men, machines, material and capital
at the disposal of the planner. One has to take decision regarding these resources to either
maximize production, or minimize the cost of production or maximize the profit. These
problems are referred as the problems of constrained optimization.
Linear programming is a technique for determining an optimal schedule of interdependent
activities, for the given resources. Therefore, one can say that programming refers to
planning and the process of decision making about a particular plan of action from a
given set of alternatives.
Any business activities or production activity, to be formulated as a mathematical model,
has three main constituents:
• Decision variables
• Objective function
• Constraints
1.12.1 Decision variables and parameters
Decision variables are the unknowns to be determined from the solution of the model.
The parameters represent the controlled variables of the system.
1.12.2 Objective Function.
The objective function defines the measures of effectiveness of the system as a
mathematical function of its variables. The optimal solution to the model is obtained
when the corresponding values of the decision variables yield the best value of the
objective function while satisfying all constraints. Therefore, the objective function acts
as an indicator for the achievement of the optimal solution.
While formulating a problem, the desire of decision maker is expressed as a function of
‘n’ decision variables. This function is the objective function of the linear programming
problem that is each of its items will have only one variable raised to the power one.
Some of the objective functions in practice are:
• Maximization of contribution or profit
• Minimization of costs
• Maximization of production rate or minimization of production time
• Minimization of labour turnover.
• Minimization of overtime
• Maximization of resource utilization
• Minimization of risk to environment or factory etc.
1.12.3 Constraints

Constraint can be defined as the physical limitation of the system, which limits the
decision variables to their feasible range or permissible values. These are expressed in the
form of constraining mathematical functions.

E.g.: to determine values of decision variables Xj, J=1, 2, 3, ----N, which will optimize Z
= f (X1, X2, ---- Xn)

Subject to the constraints:

• Gi (X1, X2, ------Xn) ~bi, i = 1, 2,----- m and XJ ≥ 0 j = 1, 2, 3,---- n where ~ is ≤, ≥,

or =.
Example to formulate mathematical model
Vitamin – A and vitamin - B are found in food -1 and food -2. One unit of food -1
contains 5 units of vitamin - A and 2 units of vitamin - B. One unit of food - 2 contains 6
units of vitamin - A and 3 units of vitamin - B. The minimum daily requirement of a
person is 60 units of vitamin - A and 80 units of vitamin - B. The cost per unit of food - 1
is USD 5 and one unit of food - 2 is USD 6. Assume that any excess units of vitamins are
not harmful. Find the minimum cost of the mixture (food - 1 and food - 2) which meets
the daily minimum requirements of vitamins.
Formulation of the mathematical model
Suppose X1 = the number of units of food - 1 in the mixture
X2 = the number of units of food - 2 in the mixture
Formulation of constraints:
For vitamin – A:
Since each unit of food-1 contains 5 units of vitamin A, we have that X 1 units of food-1
contains 5X1 units of vitamin A. (1)
Since each unit of food-2 contains 6 units of vitamin A, we have that X 2 units of food-2
contains 6X2 units of vitamin A. (2)
Therefore, the mixture contains 5X1 + 6X2 units of vitamin A (1 + 2)
Since the minimum requirement of vitamin A is 60 units, we can say that
5X1 + 6X2 ≥ 60 (3)
For vitamin B:
Since each unit of food-1 contains 2 units of vitamin B, we have that X 1 units of food-1
contains 2X1 units of vitamin B. (4)
Since each unit of food-2 contains 3 units of vitamin B, we have that X 2 units of food-2
contains 3X2 units of vitamin B. (5)
Therefore, the mixture contains 2X1 + 3X2 units of vitamin B (4 + 5)
Since the minimum requirement of vitamin B is 80 units, we can say that
2X1 + 3X2 ≥ 80 (6)
Formulation of cost function:
Given that the cost of one unit of food-1 is USD 5 and one unit of food-2 is USD 6,
therefore X1 units of food-1 costs USD 5X1 and X2 units of food-2 costs USD 6X2.
Therefore, the cost of mixture is given by cost = 5X1 + 6X2
If we write Z for the cost function, then
Z = 5X1 + 6X2
Since the number of units X1 and X2 are always non-negative, therefore we have
X1 ≥ 0, X2 ≥ 0
Therefore, the mathematical model becomes:
Minimize Z = 5X1 + 6X2
Subject to the constraints:
5X1 + 6X2 ≥ 60
2X1 + 3X2 ≥ 80
And X1 ≥ 0, X2 ≥ 0
Solved problem 1
The village butcher shop traditionally makes its meat loaf from a combination of lean
ground beef and ground pork. The ground beef contains 80 percent meat and 20 percent
fat and it costs to the shop rupees 80 per kilogram. The ground pork contains 68 percent
meat and 32 percent fat and costs rupees 60 per kilogram. How much each kind of meat
should the shop use in each kilogram of meat loaf if it wants to minimize its cost and to
keep the fat content of the meat loaf to not more than 25 percent?
Solution:
The objective is to minimize the cost (in rupees), Z of a kilogram of meat loaf, where
Z = 80 times the weight of ground beef used plus 60 times the weight of ground pork
used.
Defining
X1 = weight of ground beef used in each kilogram of meat loaf.
X2 = weight of ground pork used in each kilogram of meat loaf.
We express the objective as:
Minimize Z = 80 X1 + 60 X2 (1)
Constraints:
Each kilogram of meat loaf will contain 0.20 X1 kilogram of fat contributed from the beef
and 0.32 kilogram of fat contributed from the pork respectively. The total fat content of a
kilogram of meat loaf must be no greater than 0.25 kg.
Therefore, 0.20X1 + 0.32 X2 ≤ 0.25 (2)
The weight of the beef and pork used in each kg of meat loaf must sum to 1 kg.
Hence, X1 + X2 = 1 (3)
Finally the butcher shop may not use negative quantity of either meat.
Thus, X1, X2 ≥ 0 (4)
Combining these conditions with (1), (2), and (3), we obtain
Minimize Z = 80 X1 + 60 X2
Subject to the constraints
0.20X1 + 0.32 X2 ≤ 0.25
X1 + X2 = 1
And X1, X2 ≥ 0
Solved problem 2
A furniture maker has 6 m3 of wood and 28 hours of free time, in which he will make
decorative doors. Two models of doors have sold well in the past, so he will restrict
himself to those two. He estimates that model 1 requires 2 m 3 of wood and 7 hours of
time to assemble, while model 2 requires 1 m 3 of wood and 8 hours of time. The prices of
the models are rupees 135 and rupees 90 respectively. How many doors of each model
should the furniture maker assemble if he wishes to maximize his sale revenue?
Solution:
The objective is to maximize the revenue (in rupees), Z, where
Z = 135 times the number of model 1 doors plus 80 times the number of model 2 doors
produced.
Defining
X1 = number of model 1 doors to be produced.
X2 = number of model 2 doors to be produced.
We express the objective as:
Maximize Z = 135 X1 + 90 X2 (1)
Constraints:
The furniture maker is subjected to a wood constraint. As each model 1 door requires 2
m3 of wood, 2X1 m3 must be allocated to them. Likewise, 1X2 m3 of wood must be
allocated to the model 2 doors production. Hence the wood constraint is
Therefore, 2X1 + 1X2 ≤ 6 (2)
The furniture maker also has time constraint. The model 1 doors will consume 7X1 hours
and the model 2 doors 8X2 hours.
Hence, 7X1 + 8X2 ≤ 28 (3)
Finally it is obvious that negative quantities of either door models cannot be produced, so
two hidden constraints are X1 and X2 ≥ 0. Furthermore, since there is no revenue derived
from partially completed doors, another hidden constraint is that X 1 and X2 has to be
integers.
Thus, X1, X2 ≥ 0 and has to be integers (4)
Combining these hidden constraints with (1), (2), and (3), we obtain
Maximize Z = 135 X1 + 90 X2
Subject to the constraints
2X1 + 1X2 ≤ 6
7X1 + 8X2 ≤ 28
And X1, X2 ≥ 0 and has to be integers
Solved problem 3
Rakesh coal mines inc. operates three mines in Bihar. The coal from each mine is
separated into two grades before it is transported. The daily coal production of the mines
as well as their daily operating cost is as follows:

High grade tons / day Low grade tons / Operating cost

day Rupees / day
Mine I 4 4 20
Mine II 6 4 22
Mine III 1 6 18

Rakesh coal has committed itself to deliver 54 tons of high grade coal and 65 tons of low
grade coal by the end of the week. It also has labour contracts that guarantee employees
in each mine a full day pay for each day or fraction of a day the mine is open. Determine
the number of days each mine should be operated during the upcoming week, if Rakesh
mines has to fulfill its commitment at minimum total cost.
Solution:
Let X1, X2, and X3respectively denote the number of day that the mines I, II and III will
be operated during the upcoming week. Then the objective (measured in units of rupees
1000) is
Minimize Z = 20 X1 + 22 X2 + 18 X3 (1)
Constraints:
Rakesh is subjected to requirement constraints.
High grade coal requirement is
4 X1 + 6 X2 + X3 ≥ 54 (2)
And the low grade coal requirement is
4 X1 + 4 X2 + 6 X3 ≥ 65 (3)
As no mine can operate negative number of days, three hidden constraints are X1, X2 and
X3 ≥ 0. Furthermore, as no mine can operate more than 7 days in a week, another hidden
constraint is that X1, X2 and X3 ≤ 7 and has to be integers. Combining these hidden
constraints with (1), (2), and (3), we obtain
Minimize Z = 20 X1 + 22 X2 + 18 X3
Subject to the constraints:
4 X1 + 6 X2 + X3 ≥ 54
4 X1 + 4 X2 + 6 X3 ≥ 65
And 0 ≤ X1, X2, X3 ≤ 7 and has to be integers
Solved problem 4
A TV console manufacturer is beginning the last week of production of four different
models of steel television consoles, labeled I, II, III, and IV, each of which is to be
assembled and decorated. The models require 4, 5, 3 and 5 hours respectively for
assembling and 2, 1.5, 3 and 3 hours respectively for decoration. The profit on the models
are rupees 420, 420, 360 and 540 respectively. The manufacturer has 30,000 hours
available for assembling these products (750 assemblers working 40 hours a week) and
20,000 hours available for decoration (500 decorators working 40 hours a week). How
many of each model should the manufacturer produce during this last week to maximize
his vprofit? Assume that all units made can be sold.
Solution:
The objective is to maximize the profit (in rupees), Z, where
Z = 420 times the number of model I plus 420 times the number of model II plus 360
times the number of model III plus 540 times the number of model IV produced.
Defining
X1 = number of model I consoles to be produced.
X2 = number of model II consoles to be produced.
X3 = number of model III consoles to be produced.
X4 = number of model IV consoles to be produced.

We express the objective as:

Maximize Z = 420 X1 + 420 X2 + 360X3 + 540 X4 (1)
Constraints:
There are time constraints for the total time available for assembling and total time
available for decorations. These are respectively modeled by
4 X1 + 5 X2 + 3 X3 + 5 X4 ≤ 30,000 (2)
2 X1 + 1.5 X2 + 3 X3 + 3 X4 ≤ 20,000 (3)
Finally, it is obvious that negative quantities of any models cannot be produced, so four
hidden constraints are X1, X2, X3 and X4 ≥ 0. Furthermore, since there is no revenue
derived from partially completed models in the last week, another hidden constraint is
that X1, X2, X3 and X4 has to be integers.
Thus, X1, X2, X3 and X4 ≥ 0 and has to be integers (4)
Combining these hidden constraints with (1), (2), and (3), we obtain
Maximize Z = 420 X1 + 420 X2 + 360X3 + 540 X4
Subject to the constraints
4 X1 + 5 X2 + 3 X3 + 5 X4 ≤ 30,000
2 X1 + 1.5 X2 + 3 X3 + 3 X4 ≤ 20,000
And X1, X2, X3 and X4 ≥ 0 and has to be integers
Solved problem 5
The Oil and Natural Gas company (ONGC) produces two types of blended petroleum,
regular and premium, which it sells to its chain of service stations for rupees 720 and
rupees 840 per barrel, respectively. Both types of petroleum are blended from ONGC’s
inventory of refined domestic and refined foreign oil and must meet the following
specifications:

Maximum vapour Maximum Maximum Minimum

pressure octane rating demand bbl / deliveries bbl /
week week
Regular 23 88 100,000 50,000
Premium 23 93 20,000 5,000

The characteristics of the refined oil in the inventory are as follows:

Vapour Octane rating Inventory bbl Cost Rupees /

pressure bbl
Domestic 25 87 40,000 480
Foreign 15 98 60,000 900

What quantities of the two oils should ONGC blend into the two types of petroleum in
order to maximize weekly profit?
Solution:
Let
X1 = barrels of domestic blended into regular.
X2 = barrels of foreign blended into regular.
X3 = barrels of domestic blended into premium.
X4 = barrels of foreign blended into premium.
An amount (X1 + X2) of regular will be produced and generate a revenue of 720 (X 1 +
X2). An amount of (X3 + X4) of premium will be produced and generate a revenue of 840
(X3 + X4), An amount (X1 + X3) of domestic will be used at a cost of 480 (X 1 + X3) and
An amount (X2 + X4) of foreign will be used at a cost of 900 (X 2 + X4). The total profit Z
will be
Maximize Z = 720 (X1 + X2) + 840 (X3 + X4) - 480 (X1 + X3) - 900 (X2 + X4)
= 240 X1 – 180 X2 + 360 X3 – 60 X4 (1)
There are limitations imposed on production by demand, availability of supplies and
specifications on the blends. From the demands
X1 + X2 ≤ 100,000 (maximum demand for regular) (2)
X3 + X4 ≤ 20,000 (maximum demand of premium) (3)
X1 + X2 ≥ 50,000 (minimum regular required) (4)
X3 + X4 ≥ 5,000 (minimum premium required) (5)
From the availability
X1 + X3 ≤ 40,000 (availability of domestic) (6)
X2 + X4 ≤ 60,000 (availability of foreign) (7)
The constituents of a blend contribute to the overall octane rating according to their
percentage by weight. Likewise, there is constraint for their vapour pressure. Thus, octane
rating or regular is:
[87 X1 / (X1 + X2)] + [98 X2 / (X1 + X2)]
This should be at least 88.
Thus, [87 X1 / (X1 + X2)] + [98 X2 / (X1 + X2)] ≥ 88
Or X1 – 10X2 ≤ 0 (8)
Similarly [87 X3 / (X3 + X4)] + [98 X4 / (X3 + X4)] ≥ 93
Or 6 X3 – 5X4 ≤ 0 (premium octane rating constraint) (9)
For vapour pressure constraints:
[25 X1 / (X1 + X2)] + [15 X2 / (X1 + X2)] ≤ 23
Or 2X1 – 8X2 ≤ 0 (10)
[25 X3 / (X3 + X4)] + [15 X4 / (X3 + X4)] ≤ 23
Or 2 X3 – 8X4 ≤ 0 (11)
Further no petrol can be mixed with negative quantity, thus,
 X1, X2, X3 and X4 ≥ 0
Combining all these constraints
Maximize Z = 240 X1 – 180 X2 + 360 X3 – 60 X4
Subject to the constraints:
X1 + X2 ≤ 100,000
X3 + X4 ≤ 20,000
X1 + X2 ≥ 50,000
X3 + X4 ≥ 5,000
X1 + X3 ≤ 40,000
X2 + X4 ≤ 60,000
X1 – 10X2 ≤ 0
6 X3 – 5X4 ≤ 0
2X1 – 8X2 ≤ 0
2 X3 – 8X4 ≤ 0
X1, X2, X3 and X4 ≥ 0
Solved problem 6
A forest ranger plans to go on a camping trip. There are five items the ranger wishes to
take with him, but together they exceed the 60 kg weight limit she feels he can carry. To
assist himself in the selection process, he has assigned a value to each item in ascending
order of importance

Item 1 2 3 4 5
Weight, kg 52 23 35 15 7
Value 100 60 70 15 15

Which items should she take to maximize the total value without exceeding the weight
restriction?
Solution
Let Xi (I = 1, 2, 3, 4, 5) = amount of item i to be taken
We can formulate the objective as
Z = 100 X1 + 60 X2 + 70 X3 + 15 X4 + 15 X5
The weight limitation is
52 X1 + 23 X2 + 35 X3 + 15 X4 + 7 X5 ≤ 60 (1)
Since each item is either to be carried or not to be carried, Xi = 0 or 1
Therefore mathematical program
Maximize
Z = 100 X1 + 60 X2 + 70 X3 + 15 X4 + 15 X5
Subject to the constraints
52 X1 + 23 X2 + 35 X3 + 15 X4 + 7 X5 ≤ 60
And Xi = 0 or 1 (i = 1, 2, 3, 4, 5)
Solved problem 7
A 24 hour super market has the following minimal requirements for cashiers:

Period 1 2 3 4 5 6
Time of 3-7 7 - 11 11 - 15 15 - 19 19 - 23 23 - 3
day
Minimum 7 20 14 20 10 5
nos.

Period 1 follows immediately after period 6. A cashier works 8 consecutive hours,

starting at the beginning of one of the six periods. Determine a daily employee worksheet,
which satisfies the requirements with the least number of personnel.
Solution:
Let Xi (i = 1, 2, ….6) = number of cashiers beginning work at the start of the period
i.
We can model this problem by mathematical program
Minimize Z = X1 + X2 + X3 + X4 + X5 + X6
Subject to the constraints:
X1 + X6 ≥ 7
X1 + X2 ≥ 20
X2 + X3 ≥ 14
X3 + X4 ≥ 20
X4 + X5 ≥ 10
X5 + X6 ≥ 5
And Xi (i = 1, 2, 3, 4, 5, 6) ≥ 0 and has to be integral.
Solved problem 8:
A cheese shop has 20 kg of seasonal fruit mix and 60 kg of an expensive cheese with
which it will make two cheese spreads, deluxe and regular, that are popular during
Christmas week. Each kg of deluxe spread consists of 0.2 kg of fruit mix and 0.8 kg of
expensive cheese, while each kg of regular spread consists of 0.2 kg of fruit mix, 0.3 kg
of expensive cheese and 0.5 kg of filter cheese, which is cheap and plentiful in supply.
From past pricing policies, the shop has found out that the demand for each spread
depends upon its price as follows:
D1 = 190 – 25 P1 and D2 = 250 – 50 P2
Where D denotes demand in kg and P denotes price in rupees, and the subscripts 1 and 2
refer to the deluxe and regular cheese spreads, respectively. How many kg of each spread
should the cheese shop prepare, and what price should it establish, if it wishes to
maximize income and be left with no inventory of either spreads at the end of Christmas
week?
Solution:
Let X1 kg of deluxe spread and X 2 kg of regular spread be made. If all product can be
sold, the objective is to:
Maximize Z = P1X1 + P2X2 (1)
Now all product will indeed be sold (and none will be left over in inventory). If
production does not exceed demand, i.e., X1 ≤ D1 and X2 ≤ D2, this gives the constraints
X1 + 25 P1 ≤ 190 (2)
And X2 + 50 P2 ≤ 250 (3)
From the availability of fruit mix
0.2 X1 + 0.2 X2 ≤ 20 (4)
From the availability of expensive cheese
0.8 X1 + 0.3 X2 ≤ 60 (5)
There is no constraint on filler cheese, since the shop has as much as it needs. Finally,
neither production nor price can be negative, so four hidden constraints are X 1, X2, P1, P2
≥ 0.
Combining these hidden constraints with equations (1) to (5)
The mathematical program
Maximize Z = P1X1 + P2X2
Subject to the constraints;
X1 + 25 P1 ≤ 190
X2 + 50 P2 ≤ 250
0.2 X1 + 0.2 X2 ≤ 20
 0.8 X1 + 0.3 X2 ≤ 60
And X1, X2, P1, P2 ≥ 0.

Solved problem 9:
A plastic manufacturer has 1200 boxes of transparent wrap in stock at one factory and
1000 boxes at its second factory. The manufacturer has orders for this product from three
different retailers in quantities of 1000, 700 and 500 boxes, respectively. The unit
shipping costs (in cents per box) from the factories to the retailers are as follows:

Retailer 1 Retailer 2 Retailer 3

Factory 1 14 13 11
Factory 2 13 13 12

Determine a minimum cost shipping schedule for satisfying all demands from current
inventory.
Solution:
Let Xij (i = 1, 2 and j = 1, 2, 3) are the numbers of boxes to be shipped from factory i to
retailer j. then the objective will be:
Minimize Z = 14 X11 + 13 X12 + 11 X13 + 13 X21 + 13 X22 + 12 X23
Constraints:
Since amounts shipped from the factories cannot exceed supplies
X11 + X12 + X13 ≤ 1200(shipments from factory 1) (1)
X21 + X22 + X23 ≤ 1000(shipments from factory 2) (2)
Additionally total amount sent to retailers must meet their demands, hence
X11 + X21 ≥ 1000 (demand of retailer 1) (3)
X12 + X22 ≥ 700 (demand of retailer 2) (4)
X13 + X23 ≥ 500 (demand of retailer 3) (5)
Since the total quantity of supply (1000 +1200) is equal to total quantity of demands of
retailers (1000 + 700 + 500), each inequality constraints can be tightened to equality.
Further, no shipment can be of negative quantity and no no box can be split for shipment,
Xij ≥ 0 and has to be integers.
Combining objective function, all (1) to (5) constraints and hidden constraints, the
mathematical program becomes:
Minimize Z = 14 X11 + 13 X12 + 11 X13 + 13 X21 + 13 X22 + 12 X23
Subject to the constraints:
 X11 + X12 + X13 = 1200
X21 + X22 + X23 = 1000
X11 + X21 = 1000
X12 + X22 = 700
X13 + X23 = 500
And Xij ≥ 0 and has to be integers
Solved problem 10
A 400 meter medley relay involves four different swimmers, who successively swim 100
meters of back stroke, breast stroke, butter fly and free style. A coach has six very fast
swimmers whose expected times in seconds in the individual events are as given below:

Event 1 (back Event 2 Event 3 Event 4 (free

stroke) (breast stroke) (butter fly) style)
Swimmer 1 65 73 63 57
Swimmer 2 67 70 65 58
Swimmer 3 68 72 69 55
Swimmer 4 67 75 70 59
Swimmer 5 71 69 75 57
Swimmer 6 69 71 66 59

How should coach assign swimmers to the relay so as to minimize the sum of their time?
Solution:
The objective is to minimize total time, Z.
Let Xij (i = 1, 2, …, 6 and j = 1, 2, 3, 4) = number of times the swimmer i will be assigned
to event j. thus, the objective will be
Minimize Z = 65 X11 + 73 X12 + 63 X13 + 57 X14 + 67 X21 + 70 X22 + 65 X23 + 58 X24 + 68
X31 + 72 X32 + 69 X33 + 55 X34 + 67 X41 + 75 X42 + 70 X43 + 59 X44 + 71 X51 + 69 X52 +
75 X53 + 57 X54 + 69 X61 + 71 X62 + 66 X63 + 59 X64
Constraints
Since no swimmer can be assigned to more than one event
X11 + X12 + X13 + X14 ≤ 1 (1)
X21 + X22 + X23 + X24 ≤ 1 (2)
X31 + X32 + X33 + X34 ≤ 1 (3)
X41 + X42 + X43 + X44 ≤ 1 (4)
X51 + X52 + X53 + X54 ≤ 1 (5)
X61 + X62 + X63 + X64 ≤ 1 (6)
Since each event must have one swimmer assigned to it, we also have
X11 + X21 + X31 + X41 + X51 + X61 ≤ 1 (7)
X12 + X22 + X32 + X42 + X52 + X62 ≤ 1 (8)
X13 + X23 + X33 + X43 + X53 + X63 ≤ 1 (9)
X14 + X24 + X34 + X44 + X54 + X64 ≤ 1 (10)
Further, either a swimmer is assigned or not assigned. Thus, the value of Xij will be 0
or 1 only.
Combining the objective function, all 10 constraints and hidden constraint, the
mathematical program will be an integer program.
Solved problem 11
Mr. Rakesh has rupees four million to invest and three opportunities available to
him. Each opportunity requires deposits in rupees 1 million amount. Mr. Rakesh
may allocate all the money to one opportunity or split the money between them. The
expected returns are tabulated as below:

Amount invested (in millions)

0 1 2 3 4
Return from 0 2 5 6 7
opportunity
1
Return from 0 1 3 6 7
opportunity
2
Return from 0 1 4 5 8
opportunity
3

How much money should be invested in each opportunity to obtain the greatest total
return?
Solution:
The objective is to maximize total return denoted by Z, which is the sum of the
returns from each opportunity. All investments are restricted to be integral
multiples of unit 1 million.
Letting fi (x) (i = 1. 2. 3) denote the return (in 1 million rupees units) from
opportunity i when x units of money are invested in it. We can rewrite the return
table as given below.

f x 0 1 2 3 4
f1 (x) 0 2 5 6 7
f2 (x) 0 1 3 6 7
f3 (x) 0 1 4 5 8
 Defining xi (i = 1, 2, 3) as the number of units of money invested in opportunity i, we
can formulate the objective as:
Maximize Z = f1 (x1) + f2 (x2) + f3 (x3) (1)
Since the individual has only four units of money to invest
x1 + x2 + x3 ≤ 4 (2)
Augmenting (1) and (2) with the hidden condition that x 1, x2 and x3 has to be
nonnegative and integral, we obtain the mathematical program:
Maximize Z = f1 (x1) + f2 (x2) + f3 (x3)
Subject to the constraint
x1 + x2 + x3 ≤ 4
And x1, x2 and x3 ≥ 0 and has to be integral

1.13 Limitations of Operations Research

The limitations are more related to the problems of model building, time and money
factors. The limitations are:
• Magnitude of computation: Modern problems involve a large number of variables.
The magnitude of computation makes it difficult to find the interrelationship.
• Intangible factors: Non-quantitative factors and human emotional factors cannot be
taken into account.
• Communication gap: There is a wide gap between the managers and the operations
researchers.
• Time and money factors: When the basic data is subjected to frequent changes, then
incorporation of them into OR models is a costly affair.
• Human Factor: Implementation of decisions involves human relations and
behaviour.
REVIEW QUESTIONS
Formulate mathematical programs for the problems given below:
1. Mr. Harsha has developed two types of handcrafted adult games that he sells to
department stores throughout India. Although the demand of these games
exceeds his capacity to produce them, Mr. Harsha continues to work alone and
limits his week work to 50 hours. Game 1 takes 3.5 hours to produce and gives a
profit of rupees 140, while game 2 requires 4 hours to complete and brings a
profit of rupees 155. How much games of each type should Mr. Harsha produce
weekly, if his objective is to maximize the total profit?
Ans: maximize Z = 140 X1 + 155 X2
 Subject to the constraints
3.5 X1 + 4 X2 ≤ 50
And X1, X2 ≥ 0

2. A pharmacy has determined that a healthy person should receive 70 units of

proteins, 100 units of carbohydrates and 20 units of fat daily. If the store carries the
six types of health food with their ingredients as shown in the table below, what
blend of foods satisfies the requirements at minimum cost to the pharmacy?

Foods Protein units Carbohydrates Fat units Cost per unit

units
A 20 50 4 2
B 30 30 9 3
C 40 20 11 5
D 40 25 10 6
E 45 50 9 8
F 30 20 10 8

Ans: Minimize Z = 2X1 + 3 X2 + 5 X3 + 6 X4 + 8 X5 + 8 X6

Subject to the constraints:
20 X1 + 30 X2 + 40 X3 + 40 X4 + 45 X5 + 30 X6 ≥ 70
50 X1 + 30 X2 + 20 X3 + 25 X4 + 50 X5 + 20 X6 ≥ 100
4 X1 + 9 X2 + 11 X3 + 10 X4 + 9 X5 + 10 X6 ≥ 20
And X1, X2 X3, X4, X5, X6 ≥ 0

3. A local manufacturing firm produces four different metal products, each of which
must be machined, polished and assembled. The specific time requirements (in
hours) for each product are as follows:

Machining, hours Polishing, hours Assembling, hours

Product I 3 1 2
Product II 2 1 1
Product III 2 2 2
Product IV 4 3 1

The firm has available to it on weekly basis, 480 hours of machining time, 400 hours
of polishing time and 400 hours of assembling time. The unit profits on the product
are rupees 360, rupees 240, rupees 360 and rupees 480, respectively. The firm has a
contract with a distributor to provide 50 units of product I, and 100 units of any
combination of products II and III each week. Through other customers the firm
can sell each week as many units of products I, II and III as it can produce, but only
a maximum of 25 units of product IV. How many units of each product should the
firm manufacture each week to meet all contractual obligations and maximize its
total profit? Assume that any unfinished pieces can be finished the following week.
4. A caterer must prepare from five fruit drinks in stoke 500 liters of a punch
containing at least 20 % orange juice, 10 % grapes juice and 5% cranberry juice. If
inventory data are as shown below, how much of each fruit drink should the caterer
use to obtain the required composition at the minimum total cost?

Orange Grapes Cranberry Supply, Cost Rs. /

juice, % juice, % juice, % liters liter
Drink A 40 40 0 200 90
Drink B 5 10 20 400 45
Drink C 100 0 0 100 120
Drink D 0 100 0 50 105
Drink E 0 0 0 800 15

5. A town has been budgeted with rupees 15 million for the development of new
rubbish disposal areas. Seven sites are available, whose projected capacities and
development costs are as given below. Which sites should the town develop?

site A B C D E F G
Capacity 20 17 15 15 10 8 5
tons /
week
Cost per 8700 5520 4200 4200 2160 840 2820
1000

6. A semi conductor corporation produces a particular solid state module that it

supplies to four different television manufacturers. The module can be produced at
each of the corporation’s three plants, although the production cost varies due to
differing production efficiencies at the plants. Specifically, it costs rupees 66 to
produce a module at plant A, rupees 57 at plant B and rupees 61.8 at plant C.
Monthly production capacities of the plants are 7500, 10,000 and 8100 modules,
respectively. Sales forecasts project monthly demand at 4200, 8300, 6300 and 2700
modules for television manufacturer I, II, III and IV, respectively. If the cost of
transportation of a module from a factory to a manufacturer is as shown below, find
a production schedule that will meet all needs at minimum total cost.
Manufacturer I Manufacturer II Manufacturer III Manufacturer IV
Plant A 6.6 7.8 5.4 11.0
Plant B 7.2 9.6 6.0 8.4
Plant C 8.4 7.2 7.2 9.0
 7. The manager of a supermarket meal department finds she has 200 kg of round
steak, 800 kg of chuck steak and 150 kg of pork in stock on Saturday morning,
which she will use to make hamburger meat, picnic patties and meat loaf. Demands
for each of these items always exceed the supermarket’s supply. Hamburger meat
must be at least 20 percent ground round and 50 percent ground chuck (by weight),
picnic patties must be at least 20 percent ground pork and 50 percent ground chuck
and meat loaf must be at least 10 percent ground round, 30 percent ground pork
and 40 percent ground chuck. The reminder of each product is an inexpensive
nonmeat filler, which the store has in unlimited supply. How many kg of each
product should be made if the manager desires to minimize the amount of meat that
must be stored in the supermarket over Sunday?
8. A legal firm has accepted five new cases, each of which can be handled
adequately by any one of its five junior partners. Due to differences in experience
and expertise, however, the junior partners would spend varying amount of time on
the cases. A senior partner has estimated the time requirement (in hours) as shown
below in the table. Determine an optimum assignment of cases to lawyer such that
each junior partner receives a different case and the total hours expended by the
firm can be minimized.

Case 1 Case 2 Case 3 Case 4 Case 5

Lawyer 1 145 122 130 95 115
Lawyer 2 80 63 85 48 78
Lawyer 3 121 107 93 69 95
Lawyer 4 118 83 116 80 105
Lawyer 5 97 75 120 80 111

9. A recreational motor manufactures golf carts and snow mobiles at its three
plants. Plant A produces 40 golf carts and 35 snow mobiles daily. Plant B produces
65 golf carts daily but no snow mobiles and plant C produces 53 snow mobiles daily
but no golf carts. The cost of operating plants A, B and C are respectively rupees
12.6 million, 11.4 million and 10.92 million per day. How many days (including
Sundays and holidays) should each plant operate during September to fulfill a
production schedule of 1500 golf carts and 1100 snow mobiles at minimum cost?
Assume that labour contracts require that once a plant is opened, workers must be
paid for entire day.
10. The Apollo fertilizer corporation produces two types of farm fertilizers, Apollo
regular and Apollo deluxe. Apollo regular is composed of 25% active ingredients
and 75% inert ingredients, while Apollo deluxe contains 40% active ingredients and
60% inert ingredients. Godown storage capacity limit inventories to 500 tons of
active ingredients and 1200 tons of inert ingredients, and they are completely
replenished once a week.
Apollo regular is similar to other fertilizers on the market and competitively priced
at rupees 1500 per ton. At this price, company has had no difficulty in selling all the
Apollo regular it produces. Apollo deluxe however, has no competition, and so there
are no constraints on its price. Of course, demand dose depend on price, and
through past experience the company has determined that price P (in rupees) and
demand D (tons) are related by P = 36000 – 60D. How many tons of each type of
fertilizer should Apollo Fertilizer Corporation produce weekly in order to maximize
revenue?

CHAPTER 2
LINEAR Programming
2.1 Introduction
Minimal time effort and maximum benefit coupled with best possible output or set of
outputs is the formula of decision maker. Today decision makers or managements have to
tackle the issue of allocating limited and scarce resources at various levels in an
organization in the best possible manner. Man, machine, time and technology are some of
these common resources. The management’s task is to obtain the best possible output (or
set of outputs) from these given resources.
The outputs can be measured from factors, such as profits, the costs, the social welfare,
and the overall effectiveness. In several situations, output can be expressed as a linear
relationship among several variables. The amount of available resources can also be
expressed as a linear relationship among the various system variables. The management
dilemma is to optimize (maximize or minimize) the output or the objective function
subject to the set of constraints. Optimization of resources in which both the objective
function and the constraints are represented by a linear form is known as a linear
programming problem (LPP).
2.2 Linear programming
Linear programming (LP) is a mathematical technique designed to help managers in their
planning and decision making. It is usually used in an organization that is trying to make
most effective use of its resources. Resources typically include machinery, manpower,
money, time, warehouse space, or raw materials.
A few examples of problems in which LP has been successfully applied are:
1. Developments of production schedule that will satisfy future demands for a firm’s
product and at the same time minimize total production and inventory costs.
2. Establishment of an investment portfolio from a variety of stocks or bonds that will
maximize a company’s return on investment.
3. Allocation of a limited advertising budget among radio, TV, and newspaper in order
to maximize advertising effectiveness.
4. Determination of a distribution system that will minimize total shipping cost from
several warehouses to various market locations.
5. Selection of the product mix in a factory to make best use of machine and man hours
available while maximizing the firm’s profit.

2.3 Properties of an LP problem

(a) All LP problems seek to maximize or minimize some quantity (usually profit or cost).
We refer to this property as the objective of an LP problem.
(b) The second property that LP problems have in common is the presence of restrictions,
or constraints, that limit the degree to which we can pursue the objective. Therefore
 maximization or minimization of a quantity (the objective function) subject to limited
resources (the constraints) is the requirement.
2.4 Requirements of linear programming problem
The common requirements of LPP are as follows:
• Decision variables and their relationship
• Well defined objective function
• Existence of alternative courses of action.
• Non-negative conditions on decision variables.
2.5 Basic assumptions of linear programming
The basic assumptions of LPP are:
• Linearity: Both objective function and constraints must be expressed as linear
inequalities.
• Deterministic: All coefficients of decision variables in the objective and constraint
expressions should be known and finite.
• Additivity: The value of objective function for the given values of decision variables
and the total sum of the resources used must be equal to sum of the contribution
earned from each decision variables and the sum of the resources used by the decision
variables respectively.
• Divisibility: The solution of the decision variables and resources can be any non-
negative values including fractions.
2.6 Mathematical formulation of LPP
The procedure for mathematical formulation of a linear programming problem consists of
the following major steps:
 Study the given situation to find the key decisions to be made
Step 1

Identify the variables involved and designate them by symbols

Step 2 Xj (j = 1, 2, ........n )

State the feasible alternatives which generally are X j ≥ 0, for all

Step 3 j
Identify the constraints in the problem and express them as
linear inequalities or equations. LHS of which are linear
Step 4 functions of the decision variables.

Identify the objective function and express it as a linear

Step 5 function of the decision variables

Fig 2.1: procedure of mathematical formulation of a linear programming problem

General Linear Programming Model
The Linear Programming Problem (LPP) is a class of mathematical programming in
which the functions representing the objectives and the constraints are linear.
Optimization refers to the maximization or minimization of the objective functions.
The general linear programming model is usually defined as follows:-
Maximize or minimize
Z = C1 X1 + C2 X2 + -------Cn Xn
Subject to the constraints:
a11 X1 + a12 X2 + ------- a1n Xn ~ b1
a21 X1 + a22 X2 + ------- a2n Xn ~ b2
---------------------------------------------
am1 X1 + am2 X2 + ------ amn Xn ~ bn
where Cj , bj and aij (i = 1,2,3-----m, j = 1,2,3 -----n) are constants. X j (j = 1,2,3 -----n) are
the decision variables. Hare ~ is either ≤, = or ≥.
The interpretation is that:
bi = Available amount of resource i
aij = The amount of resource i to be allocated to each unit of activity j.
Cj = the cost or worth per unit of activity
Z = the objective to be minimized or maximized.
Solved problem 1
A company has three operational departments (weaving, processing and packing) with
capacity to produce three different types of cloths namely, suiting, shirting and woolens
yielding a profit of Rupees 2, Rupees 4 and Rupees 3 per meter respectively. One meter
of suiting requires 3 minutes in weaving, 2 minutes in processing and 1 minute in
packing. Similarly one meter of shirting requires 4 minutes in weaving, 1 minute in
processing and 3 minutes in packing. One meter of woolen requires 3 minutes in each
department. In a week, total run time of each department is 60, 40 and 80 hours for
weaving, processing and packing respectively.
Formulate the linear programming problem to find the product mix to maximize the
profit.
Solution:
Step 1: The key decision is to determine the weekly rate of production for the three types
of cloths.
Step 2: Let us designate the weekly production of suiting’s, shirting’s and woolens by X 1
meters, X2 meters and X3 meters respectively.
Step 3: Since it is not possible to produce negative quantities, feasible alternatives are
sets of values of X1, X2 and X3 satisfying X1, X2 and X3 ≥ 0.
Step 3: The constraints are the limited availability of three operational departments. One
meter of suiting requires 3 minutes of weaving. The quantity being X1 meters, the
requirement for suiting alone will be 3X1 units. Similarly, X2 meters of shirting and X3
meters of woolen will require 4X2 and 3X3 minutes respectively. Thus the total
requirement of weaving will be 3X1 + 4X2 + 3X3, which should not exceed the available
3600 minutes. So the man-hour constraint for weaving becomes:
3X1 + 4X2 + 3X3 ≤ 3600
Similarly, the constraints for the processing department and packing departments are:
2X1 + X2 + 3X3 ≤ 2400
X1 + 3X2 + 3X3 ≤ 4800
Step 5: The objective is to maximize the total profit from sales. Assuming that whatever
is produced is sold in the market, the total profit is given by the linear relation
Z = 2X1 + 4X2 + 3X3
The linear programming can thus be put in the following algebraic format:
Find X1, X2 and X3 so as to maximize
Z = 2X1 + 4X2 + 3X3
 Subject to the constraints:
3X1 + 4X2 + 3X3 ≤ 3600
2X1 + X2 + 3X3 ≤ 2400
X1 + 3X2 + 3X3 ≤ 4800
And X1, X2 and X3 ≥ 0
Canonical forms
The general linear programming problem (LPP) can always be put in the standard form as
given below, called as “canonical form”.
Maximize Z = C1 X1 + C2 X2 + -------Cn Xn
Subject to the constraints:
a11 X1 + a12 X2 + ------- a1n Xn ≤ b1
a21 X1 + a22 X2 + ------- a2n Xn ≤ b2
---------------------------------------------
am1 X1 + am2 X2 + ------ amn Xn ≤ bn
Where Cj, bj and aij (i = 1, 2, 3-----m, j = 1, 2, 3 -----n) are constants. X j (j = 1,2,3 -----n) are
the decision variables and has to be ≥ 0.
The characteristics of this form are:-
• All decision variables are non-negative.
• All constraints are of ≤ type.
• The objective function is of the maximization type.
Any LPP can be put in the canonical form by the use of five elementary transformations:
1. The minimization of a function is mathematically equivalent to the maximization of
the negative expression of this function. That is,
Minimize Z = C1 X1 + C2 X2 + -------Cn Xn is equivalent to
Maximize - Z = - C1 x1 – C2 X2 - ------- - Cn Xn
2. Any inequality in one direction (≥ or ≤) may be changed to an inequality in the
opposite direction by multiplying both sides by -1.
For example 2X+3Y ≥ 5 is equivalent to -2X-3Y ≤ -5.
3. An equation can be replaced by two inequalities in opposite direction.
For example, 2X+3Y = 5 can be written as 2X + 3Y ≤ 5 and 2X +3Y ≥ 5 or -2X – 3Y ≤
-5
 4. An inequality constraint with its left hand in the absolute form can be changed into
regular inequalities.
For example: |2X + 3Y| ≤ 5 is equivalent to 2X + 3Y ≤ 5 and 2X + 3Y ≥ -5 or -2X - 3Y ≤
-5.
5. The variable which is unconstrained in sign (i.e., ≤ 0, ≥ 0 or 0) is equivalent to the
difference between two non-negative variables.
For example, if X is unconstrained in sign then X = (X+ - X-) where X+ ≥ 0, X- ≤ 0.

Solved problem 2:
A firm engaged in producing 2 models, viz., model A and model B, performs only three
operations painting, assembly and testing. The relevant data are as follows:
Table 2.1: Unit sale price and hours required for each unit

Unit sale price Hours required for each unit

Assembly Painting Testing
Model A Rupees 50/- 1.0 0.2 0.0
Model B Rupees 80/- 1.5 0.2 0.1

Total numbers of hours available each week are for assembly 600, painting 100, and testing
30. The firm wishes to determine the weekly product mix so as to maximize revenue.
Solution:
Let Z: total revenue
X: Number of units of model A
Y: Number of units of model B
X and Y: are known as decision variables
b1: weekly hours available for assembly = 600 hours
b2: weekly hours available for painting = 100 hours
b3: weekly hours available for testing = 30 hours
Since the objective (goal) of the firm is to maximize its revenue,
Therefore, the goal is maximize Z = 50X + 80Y
Thus, the model can be stated as below:
Maximize Z = 50X + 80Y
Subject to the constraints:
1.0X + 1.5Y ≤ 600 (assembly constraint)
 0.2X + 0.2Y ≤ 100 (painting constraint)
0.0X + 0.1Y ≤ 30 (Testing constraint)
X, Y ≥ 0 (the non-negativity conditions)
Solved problem 3:
A milk distributor supplies milk in bottles to houses in three areas A, B and C in a city. His
delivery charges per bottle are Rupees 0.30 in area A, Rupees 0.40 in area B and Rupees 0.50
in area C. He has to spend on an average 1 minute to supply one bottle in area A, 2 minutes
per bottle in area B and 3 minutes per bottle in area C. He can spare only 2 hours 30 minutes
for this milk distribution but not more than 1 hour 30 minutes for area A and B together. The
maximum number of bottles he can deliver is 120. Find the number of bottles that he has to
supply in each area so as to earn the maximum.
Solution:
The decision variables of the model can be defined as follows:
X: number of bottles of milk which the distributor supplies in area A.
Y: Number of bottles of milk which the distributor supplies in area B.
Z: Number of bottles of milk which the distributor supplies in area C.
The objective:
Maximize W = 0.30X + 0.40Y + 0.50Z
Constraints:
1. Maximum number of milk bottles is 120, that is, X + Y + Z ≤ 120.
2. Since he requires one minute per bottle in area A, 2 minutes in area B and 3 minutes
in area C and he cannot spend more than 150 minutes for the work
1X + 2Y + 3Z ≤ 150
3. Further, since, he cannot spend more than 90 minutes for areas A and B. therefore:
1X + 2Y ≤ 90
4. The non-negativity X, Y ≥ 0
5. The problem can now be stated in the standard LP form as:
Maximize W = 0.30X + 0.40Y + 0.50Z
Subject to the constraints:
X + Y + Z ≤ 120
X + 2Y + 3Z ≤ 150
X + 2Y ≤ 90
And X, Y ≥ 0
Solved problem 4:
An oil company has two units A and B which produce three different grades of oil, super
fine, medium and low grade oil. The company has to supply 12, 8 and 24 barrels of super
fine, medium and low grade oils respectively per week. It costs the company Rupees 1000/-
and Rupees 800/- per day to run the units A and B respectively. On a day unit A produces 6,
2 and 4 barrels and the unit B produces 2, 2 and 12 barrels for different grades respectively.
The manager has to decide on how many days per week should each unit be operated in order
to meet the requirement at minimum cost.
Solution:
The given data can be presented in summary as follows:
Table 2.2: Capacity and requirements details of an oil industry

Product Capacity Requirements

Unit A Unit B
Super fine 6 2 12
Medium 2 2 8
Low grade 4 12 24
Cost Rupees 1,000 Rupees 800 -

Let X and Y are the number of days the units A and B be operated per week respectively.
Then the objective of the manager is to:
Minimize the cost function Z = 1000X + 800Y
Subject to the constraints: 6X + 2Y ≥ 12 (super fine)
2X + 2Y ≥ 8 (medium)
4X + 12Y ≥ 24 (low grade)
X, Y ≥ 0
Solved problem 5:
A toy company manufactures two types of doll, a basic version doll A and a deluxe version
doll B. Each doll of type B takes twice as long to produce as one of type A, and company
would have time to make a maximum of 2000 per day. The supply of plastic is sufficient to
produce 1500 dolls per day (Both A and B combined). The deluxe version requires a fancy
dress of which there are only 600 per day available. If the company makes a profit of Rupees
3 and Rupees 5 per doll, respectively on doll A and B, then how many of each doll should be
produced per day in order to maximize the total profit. Formulate this problem.
Solution:
Let X and Y are the number of dolls produced per day of type A and B respectively.
So the objective function:
Maximize Z = 3X + 5Y
Let the doll A requires t hrs.
Thus, doll B requires 2t hours.
So the total time to manufacture X and Y dolls should not exceed 2000 t hours.
Therefore, t X + 2tY ≤ 2000t
Or X + 2Y ≤ 2000 (time constraint)
Other constraints:
X + Y ≤ 1500 (plastic constraint)
Y ≤ 600 (dress constraint)
Further, the dolls cannot be manufactured partially as they cannot be sold unfinished,
thus, the hidden constraint will be
X, Y ≥ 0
Thus the model can be stated as below:
Maximize Z = 3X + 5Y
X + 2Y ≤ 2000
X + Y ≤ 1500
Y ≤ 600
X, Y ≥ 0

Graphical analysis of linear programming Problems.

Once the linear programming model has been formulated on the basis of the given objective
and the associated constraint functions, the next step is to solve the problem and obtain the
best possible or the optimal solution. Various mathematical and analytical techniques can be
employed for solving the linear programming model. The graphical solution is one of the
methods of solving two variable linear programming problems.
Example:
Maximize Z = 700X1 + 500X2
Subject to:
4X1 + 3X2 ≤ 210
2X1 + X2 ≤ 90
And X1 ≥ 0, X2 ≥ 0.
Let horizontal axis represent X1 and the vertical axis X2. First draw the line 4X1 + 3X2 = 210
(by replacing the inequality symbols by the equality) which meets the X 1 axis at point A
(52.50, 0) (put X2 = 0 and solve for X1 in 4X1 + 3X2 = 210) and the X2 axis at the point B (0,
70) (put X1 = 0 in 4X1 + 3X2 = 210 and solve for X2).

80
60
40
20
0
0.5 1 1.5 2 2.5 3 3.5
Fig 2.1
Any point on the line 4X1 + 3X2 = 210 or inside the shaded portion will satisfy the restriction
of the inequality, 4X1 + 3X2 ≤ 210. Similarly, the line 2X 1 + X2 = 90 meets the X1 axis at the
point (45, 0) and the X2 axis at the point (0, 90).

100
80
60
40
20
0
0.5 1 1.5 2 2.5 3 3.5
Fig 2.2
Combining the two graphs will give the shape as follows:

100
80
60
40 Y-Values
20
0
0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5
 Fig 2.3
The area inside the two graphs (shaded portion) will satisfy both the constraints. This region
is called feasible region.

Fig 2.4
Some basic definitions:

 Feasible region: Any non – negative value of (X 1, X2) (i.e., X1 ≥ 0, X2 ≥ 0) is a

feasible solution of the LPP if it satisfies all the constraints. The collection of all
feasible solution is known as the feasible region.

 Convex: A set X is convex if for any points X 1, X2 in X, the line segment joining
these points is also in X. That is X 1, X2 Є X, 0 ≤ λ ≤ 1 implies that λX 2 + (1 – λ)X1 Є
X.
By convention, a set containing only a single point is also a convex set.
λX2 + (1 – λ)X1 (where 0 ≤ λ ≤ 1) is called a convex combination of X1 and X2.
Appoint X on a convex set is said to be an extreme point if there does not exist X1,
X2 Є X (X1 ≠ X2) such that X = λX2 + (1 – λ)X1 for some λ with 0 ≤ λ ≤ 1.

 Half plane: A linear inequality in two variables is known as half plane. The
corresponding equality or line is known as the boundary of the half plane.

 Convex polygon: A convex polygon is a convex set formed by the intersection of

finite number of closed half planes.

 Redundant constraint: A constraint which does not affect the feasible region.

 Basic solution: A basic solution of a system of m equations and n variables (m < n)

is a solution, where at least n – m variables are 0.

 Basic feasible solution: A basic feasible solution of a system of m equations and n

variables (m < n) is a solution where m variables are non - negative (≥ 0) and n – m
variables are 0.

 Optimal feasible solution: Any feasible solution that optimizes the objective
function is called an optimal feasible solution.
 Fig 2.5: convex region

Fig 2.6: Non-convex region

Note: The objective function is maximized or minimized at one of the extreme points
referred to as optimum solution. Extreme points are referred to as vertices or corner
points of the convex regions.
Methodology for graphical solution of Linear Programming Problems.
Solving the LPP with two decision variables through graphical representation is easy.
Consider X1 X2 – plane, where plot the solution space enclosed by the constraints. The
solution space is a convex set bounded by a polygon; since a linear function attains extreme
(maximum or minimum) values only on the boundary of the region. You can consider the
vertices of the polygon and find the value of the objective function in these vertices. Compare
values of the objective function at these vertices to obtain the optimum feasible solution of
the problem.
The method of solving a LPP on the basis of the above analysis is known as graphical
method.
Working rule

 Step 1: Write down the equations by replacing the inequality symbols by the
equality symbols in the given constraints.

 Step 2: Plot the straight lines represented by the equations obtained in step 1.

 Step 3: Identify the convex polygon region relevant to the problem. Decide on which
side of the line the half plane is located.

 Step 4: Determine the vertices of the polygon and find the values of the given
objective function Z at each of these vertices. Identify the greatest and least of these
values. These are respectively the maximum and minimum values of Z.

 Step 5: Identify the values of (X1, X2) which correspond to the desired extreme
values of Z. This is an optimal solution of the problem.
Classification of the feasible points
1. Boundary point. The point which are on the boundary of the feasible region.
Collection of boundary points belonging to one set is called boundary line (segment).
2. Vertices. The intersections of boundary lines are called the vertices.
3. Interior points. All points inside the feasible region.
4. Corner point. The vertices if feasible are called corner points.
Solved problem 2:
Solve the given LPP using the graphical method
Maximize Z = 50X1 + 80X2
Subject to the constraints
1.0X1 + 1.5X2 ≤ 600
0.2X1 + 0.2X2 ≤ 100
0.0X1 + 0.1X2 ≤ 30
And X1 ≥ 0, X2 ≥ 0
Solution:
The horizontal axis represents X1 and the vertical axis X2. Plot the constraints lines and mark
the feasibility region as shown below:
X2

C
B
D

X1
A E

Fig. 2.7

Any point on the thick line or inside the shaded portion will satisfy all the restrictions of the
problem. The ABCDE is the feasibility region carried out by the constraints operating on the
objective function. This depicts the limits within which the values of the decision variables
are permissible.
The intersection point C can be solved by linear equation 0.1X 2 = 30 and X1 + 1.5X2 = 600.
That is C (150, 300).
The intersection point D can be solved by the linear equations X 1 + 1.5X2 = 600 and 0.2X1 +
0.2X2 = 100. That is D (300, 200).
The next step is to maximize revenues subject to the above shaded area. This can be worked
out at different corner points as tabulated below:

Points (coordinates) Z = 50X1 +80X2

A (0, 0) 0
B (0, 300) 24000
C(150, 300) 31500 (Maximum)
D (300, 200) 31000
E (500, 0) 25000

Thus Maximum Z = 31500.

Solved problem 3:
Solve the given LPP using the graphical method
Minimize Z = 20X + 10Y
Subject to the constraints
X + 2Y ≤ 40
3X + Y ≥ 30
4X + 3Y ≥ 60
And X, Y ≥ 0
Solution:
The horizontal axis represents X and the vertical axis Y. Plot the constraints lines and mark
the feasibility region as shown below:
Any point on the thick line or inside the shaded portion will satisfy all the restrictions of the
problem. The ABCD is the feasibility region carried out by the constraints operating on the
objective function. This depicts the limits within which the values of the decision variables
are permissible.
The point A is (15, 0)
The intersection point B can be solved by linear equation 3X + Y = 30 and 4X + 3Y = 60.
That is B (6, 12).
The intersection point C can be solved by the linear equations X + 2Y = 40 and 3X + Y = 30.
That is C (4, 18).
The point D is (40, 0)
The next step is to maximize revenues subject to the above shaded area. This can be worked
out at different corner points as tabulated below:

Points (coordinates) Z = 20X + 10Y

A (15, 0) 300
B (6, 12) 240 (minimum)
C(4, 18) 360
D (40, 0) 800

Thus Minimum Z = 240 at .

Some exceptional cases

 Unique optimal solution: In this case only one solution will be obtained in a
graphical solution.

 Multiple optimal solutions / Alternative solutions: When the objective function is

parallel to a binding constraint (i.e., a constraint that is satisfied as an equation by the
optimal solution), the objective function will assume the same optimal value at more
than one solution point. For this reason they are called alternative optima.

 Unbounded solution: When the values of the decision variables may be increased
indefinitely without violating any of the constraints, the solution space (feasible
region) is unbounded. The value of the objective function, in such cases, may increase
(for maximization) or decrease (for minimization) indefinitely. Thus both the solution
space and the objective function value are unbounded.

 No solution / infeasible solution: When the constraints are not satisfied

simultaneously, the linear programming problem has no feasible solution. This
problem can never occur if all the constraints are of the ‘≤’ type.
Solved problem 4:
Solve the given LPP using the graphical method
Maximize Z = 2X + 4Y
Subject to the constraints
X + 2Y ≤ 5
X+Y≤4
And X, Y ≥ 0
Solution:
The horizontal axis represents X and the vertical axis Y. Plot the constraints lines and mark
the feasibility region as shown below:

Any point on the thick line or inside the shaded portion will satisfy all the restrictions of the
problem. The ABCD is the feasibility region carried out by the constraints operating on the
objective function. This depicts the limits within which the values of the decision variables
are permissible.
The point A is (0, 0)
The point B is (0, 2.5).
The intersection point C can be solved by linear equation X + Y = 4 and X + 2Y = 5. That is
C (3, 1).
The point D is (4, 0)
The next step is to maximize the value of Z subject to the above shaded area. This can be
worked out at different corner points as tabulated below:

Points (coordinates) Z = 2X + 4Y
A (0, 0) 0
B (0, 2.5) 10 (maximum)
C(3, 1) 10 (maximum)
D (4, 0) 8
Since the maximum value for Z is same from point B till point C (Z = 10) on the line
depicting the equation X + 2Y = 5, the problem has got multiple optimal solution / or
alternative solutions.
Thus Maximum Z = 10.
Solved problem 5
Solve the given LPP using the graphical method
Maximize Z = 3X1 + 2X2
Subject to the constraints
X1 - X2 ≤ 1
X1 + X2 ≥ 3
And X1 ≥ 0, X2 ≥ 0
Solution;

Solved problem 6:
Solve the given LPP using the graphical method
Maximize Z = 3X1 - 2X2
Subject to the constraints
X1 + X2 ≤ 1
2X1 + 2X2 ≥ 4
And X1 ≥ 0, X2 ≥ 0
Solution:
Important geometric properties of LPP
 The region of feasible solution has convexity property of geometry, provided the
feasible solution of the problem exists.
 Convexity means that the region of the feasible solution has no holes in them. That is,
they are solids, and they have no cuts (like) on the boundary.
 The line joining any two points in the region also lies in the region.
 The boundaries of the regions are lines or plains.
 There are corners or extreme points on the boundary, and there are edges joining
various corners.
 The objective function can be represented by a line or a plane for any fixed value of z.
 At least one corner of the region of feasible solutions will be an optimal solution
wherever the maximum or minimum value z is finite.
 If the optimal solution is not unique, there are points other than corners that are
optimal but in any case at least one corner is optimal.
 The different situation is found when the objective function can be made arbitrarily
large. Of course no corner is optimal in that case.
 Infeasible region
Solved problem 7:
Use the graphical method to solve the LPP.
Maximize Z = 5X1 + 3X2
Subject to constraints:
3X1 + 5X2 ≤ 15
 5X1 + 2X2 ≤ 10
And X1, X2 ≥ 0
Solution:

Solved problem 8:
A company making the cold drinks has two bottling plants located at two towns T 1 and T2.
Each plant produces three drinks A, B and C and their production capacity per day is shown
below:

Cold drinks Plant at

T1 T2
A 6000 2000
B 1000 2500
C 3000 3000

The marketing department of the company forecasts a demand of 80,000 bottles of A,

22000 bottles of B and 40000 bottles of C in the month of June. Operating costs per day of
plant at T1 and T2 are Rupees 6000 and Rupees 4000 respectively. Find the number of days
for which each plant must be run in June so as to minimize the operating costs while meeting
the market demand.

Solution:

Let plant T1 work for X days

And plant T2 work for Y days

Then objective function:

Minimize Z = 6000X + 4000Y

The constrains are for the demands of cold drinks A, B and C. thus,
Subject to the constraints:

6000X + 2000Y ≥ 80000 (1)

1000X + 2500Y ≥ 22000 (2)

3000X + 3000Y ≥ 40000 (3)

And X, Y ≥ 0

C (0, 40

6000X + 2000Y = 80000

B (12, 4)
A (22, 0)

Thus, the LPP is to minimize the objective function subject to the constraints (1), (2), and (3).
The solution space is unbounded. The constraint (3) is dominated by the constraints (1) and
(2) and hence does not affect the solution space. Such a constraint 3000X + 3000Y = 40000
is called the redundant constraint. The value of the convex region A, B, and C are A (22,0),
B (12, 4) and C (0, 40). The values of the objective function Z at the vertices are

Zat A = 132000

Zat B = 88000

Zat C = 160000

Thus, the minimum value of Z is Rupees 88,000 and it occurs at B. hence the optimal
solution to the problem is:

Plant T1 to work for 12 days and plant T 2 to work for 4 days and minimum operating
cost will be Rupees 88,000.
Solved problem 9:
Solve the given LPP using the graphical method
Maximize Z = 4X1 + 3X2
Subject to the constraints
X1 - X2 ≤ - 1
- X1 + X2 ≤ 0
And X1 ≥ 0, X2 ≥ 0
Solution:

Solved problem 10:

Solve the given LPP using the graphical method
Maximize Z = 100X1 + 40X2
Subject to the constraints
10X1 + 4X2 ≤ 2000
3X1 + 2X2 ≤ 900
6X1 + 12X2 ≤ 3000
And X1 ≥ 0, X2 ≥ 0
Solution:
Solved problem 11:
Solve the given LPP using the graphical method
Maximize Z = 7X1 + 3X2
Subject to the constraints
X1 + 2X2 ≥ 3
X1 + X2 ≤ 4
0 ≤ X1 ≤ 5/2
0 ≤ X2 ≤ 3/2
And X1 ≥ 0, X2 ≥ 0
Solution:

Solved problem 12:

By using the graphical method, find the minimum and maximum values of the function Z = X
- 3Y, where X and Y are non - negative and subject to the following conditions:
3X + 4Y ≥ 19
 2X – Y ≤ 9
2X + Y ≤ 15
X - Y ≥ -3
Solution:

16

14

12

10
Series 1
8 Series 2
Series 3
6 Series 4

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 7 7.5

Solved problem 13:

For conducting a practical examination, the chemistry department of a college requires 10, 12
and 7 units of three chemicals X, Y and Z respectively. The chemicals are available in two
types of boxes: box A and box B. Box A contains 3, 2, and 1 units of X, Y and Z respectively
and costs Rupees 300. Box B contains 1, 2, and 2 units of X, Y and Z respectively and costs
Rupees 200. Find how many boxes of each type should be bought by the department so that
the total cost is minimal.
Solution:
Let X1 boxes of type A to be purchased.
And, X2 boxes of type B to be purchased.
Putting complete problem in tabular form:

Chemical X Chemical Y Chemical Z Costs (rupees)

Box A: X1 3 2 1 300
Box B: X2 1 2 2 200
requirements 10 12 7

Thus, the mathematical model becomes;

Minimize Z = 300 X1 + 200 X2
Subject to the constraints:
3 X1 + X2 ≥ 10 (chemical A constraint)
2 X1 + 2 X2 ≥ 12 (chemical B constraint)
X1 + 2 X2 ≥ 7 (chemical C constraint)
And, X1, X2 ≥ 0
Plot the constraints lines and mark the feasibility region as shown below:

The LPP is to minimize the objective function subject to the constraints (1), (2), and (3). The
solution space is unbounded.
The point A is (7, 0)
The intersection point B can be solved by linear equation X 1 + X2 = 6 and X1 + 2 X2 = 7. That
is B (5, 1).
The intersection point C can be solved by linear equation X1 + X2 = 6 and 3X1 + X2 = 10.
That is C (2, 4).
The point D is (0, 10)
The next step is to minimize the value of Z subject to the above shaded area. This can be
worked out at different corner points as tabulated below:

Points (coordinates) Z = 300X1 + 200 X2

A (7, 0) 2100
B (5, 1) 1700
C(2, 4) 1400 (minimum)
D (0, 10) 2000
Since the minimum value for Z is 1400 at point C (2, 4), the solution for the given problem
is;
2 x Box A to be purchased and 4 x box B to be purchased. The minimum cost will be
rupees 1400.

Limitations of LPP
In spite of wide areas of applications, some limitations are associated with linear
programming techniques. These are stated below:

 In some problems objective functions and constraints are not linear. Generally in real
life situations concerning business and industrial problems constraints are not linearly
treated to variables.

 There is no guarantee of getting integer valued solutions, for example, in finding out
how many males and females would be required to perform the job of air hostess in a
passenger aircraft. Rounding off the solution to the nearest integer will not give an
optimal solution. Integer programming deals with such problems.

 A linear programming model does not take into consideration the effect of time and
uncertainty. Thus the model should be defined in such a way that any change due to
internal as well as external factors can be incorporated.

 Sometimes large scale problems cannot be solved with linear programming

techniques even when the computer facility is available. Such difficulty may be
removed by decomposing the main problem into several small problems and then
solving them separately.

 Parameters appearing in the model are assumed to be a constant. But in real life
situations they are neither constant nor deterministic.

 Linear programming deals with only single objective whereas in real life situations
problems come across with multi – objectives. Goal programming and multi –
objective programming deal with such problems.
Advantages of LPP
The advantages of linear programming techniques may be outlined as follows:

 Linear programming technique helps in making the optimum utilization of resources.

It also indicates how a decision maker can employ productive factors most effectively
by choosing and allocating these resources.

 Quality of decisions may also be improved by linear programming technique. The

user of this technique becomes more objective and less subjective.

 Linear programming technique provides practically applicable solutions because there

might be other constraints operating outside the problem. These constraints must also
be taken into consideration. Just because so many units must be produced does not
 mean that all those can be sold. So the necessary modification of its mathematical
solution is required for the sake of convenience to the decision maker.

 In production processes, high lighting of bottlenecks is the most significant advantage

of this technique. For example, when bottlenecks occur, some machines cannot meet
the demand while others remain idle for some time.
2.6 Application of LP model
Linear programming is a powerful tool for selecting alternatives in a decision problem and,
consequently, has been applied in a wide variety of problem settings. Some of the
applications are indicated covering the major functional areas of a business organization.

 Finance: The problem of the investor could be a portfolio-mix selection problem. In

general, the number of different portfolios can be much larger than the example
indicates, more and different kinds of constraints can be added. Another decision
problem involves determining the mix of funding for a number of products when
more than one method of financing is available. The objective may be to maximize
total profits where the profit for a given product depends on the method of financing.
For example, funding may be done with internal funds, short term debt, or immediate
financing (amortized loans). There may be limits on the availability of each of the
funding options as well as financial constraints requiring certain relationships between
the funding options so as to satisfy the terms of bank loans intermediate financing.
There may also be limits on the production capacity for the products. The decision
variables would be the number of units of each product to be financed by each
funding option.

 Production and operation management: quite often in the process industries a

given raw material can be made into a wide variety of products. For example, in the
oils industry, crude oil is refined into gasoline, kerosene, home-heating oil, and
various grades of engine oil. Given the present profit margin on each product, the
problem is to determine the quantities of each, the problem is to determine the
quantities of each product that should be produced. The decision is subject to
numerous restrictions such as limits on the capacities of various refining operations,
raw material availability, demands of each product, and any government imposed
policies on the output of certain products. Similar problems also exist in the chemical
and food processing industries.

 Human resources: personal planning problems can also be analyzed with linear
programming. For example, in the telephone industry, demands for the services of
installer-repair personnel are seasonal. The problem is to determine the number of
installer-repair personnel and line repair personnel to have on the work force each
month where the total cost of hiring, layoff, overtime, and regular time wages are
minimized. The constraints set includes restrictions on the service demands that must
be satisfied, overtime usage, union agreements, and the availability of skilled people
for hire. This example runs contrary to the run of divisibility; however, the workforce
level for each month would normally be large enough that rounding to the closest
integer in each case would not be detrimental, provided the constraints are not
violated.

 Marketing: linear programming can be used to determine the proper mix of media to
use in advertising campaign. Suppose that the available media are radio, television,
 and newspapers. The problem is to determine how advertisements to place in each
medium. Of course, the cost of placing an advertisement depends on the medium
chosen. Objective is to minimize the total cost of advertising campaign, subject to a
series of constraints. Since each medium may provide a different degree of exposure
of the target population, there may be a lower bound on the total exposure from the
campaign. Also each medium may have a different efficiency rating in producing
desirable results; there may thus be a lower bound on efficiency. In addition, there
may be limits on the availability of each medium advertising.

 Distribution: Another application of linear programming is in the area of distribution.

Consider a case in which there are m factories that must ship goods to n ware houses.
A given factory could make shipments to any number of ware houses. Given the cost
to ship one unit of product from each factory to each ware house, the problem is to
determine the shipping pattern (number of units that each factory ships to each ware
house) that minimizes total costs. This decision is subject to the restrictions that
demand at each factory cannot ship more products than it has the capacity to produce.

REVIEW QUESTIONS
Question 1.
Three grades of coal A, B, and C contain ash and phosphorous as impurities. In a particular
industrial process a fuel obtained by blending the above grades containing not more than 25%
ash and 0.03% phosphorous is required. The maximum demand of fuel is 100 tons. The
percentage impurities and costs of the various grades of coal are shown below. Assuming that
there is an unlimited supply of each grade of coal and there is no loss in blending, formulate
the blending problem to minimize the cost.

Coal grade % ash % phosphorous Cost per ton in

rupees
A 30 0.02 240
B 20 0.04 300
C 35 0.03 280

Question 2.
A firm manufactures two products; the net profit on product 1 is Rupees 3 per unit and
Rupees 5 per unit on product 2. The manufacturing process is such that each product has to
be processed in two departments D1 and D2. Each unit of product1 requires processing for 1
minute at D1 and 3 minutes at D2; each unit of product 2 requires processing for 2 minutes at
D1 and 2 minutes at D2. Machine time available per day is 860 minutes at D1 and 1200
minutes at D2. How much of product 1 and 2 should be produced every day so that total
profit is maximum. Formulate this as a problem in LP.
Question 3.
Use the graphical method to solve the LPP
Maximize Z = 2X1 + 4X2
Subject to constraints:
X1 + 2X2 ≤ 5
X1 + X2 ≤ 4
And X1, X2 ≥ 0
(Ans: Any point on the line segment BC gives the maximum value (Z = 10) of the
objective function, there exists an alternative solution.)
Terminal question 4.
Use the graphical method to solve the LPP.
Maximize Z = 6X1 + X2
Subject to:
2X1 + X2 ≥ 3
X1 - X2 ≥ 0
And X1, X2 ≥ 0
(Ans: unbounded solution)

CHAPTER – 3
SIMPLEX METHOD
Introduction
The Simplex method is “a systematic procedure for generating and testing candidate
vertex solutions to a linear program.” (Gill, Murray and Wright)
It begins at an arbitrary corner of the solution set. At each iteration, the Simplex
method selects the variable that will produce the largest change towards the minimum (or
maximum) solution. That variable replaces one of its compatriots that is most severely
restricting it, thus moving the Simplex method to a different corner of the solution set and
closer to the final solution. In addition, simplex method can determine if no solution
actually exists. Note that the algorithm is greedy since it selects the best choice at each
iteration without needing information from previous or futuristic iteration.

 The Simplex method provides an efficient technique, which can be applied for solving
LPP of any magnitude involving two or more decision variables.

 It is an iterative procedure for finding the optimal solution.

 Simplex method, according to its iterative search, selects the optimal solution from
among the set of feasible solutions to the problem.

 It considers minimum number of feasible solutions to obtain an optimal solution.

Standard form of LPP
The characteristics of the standard form of LPP are:

 All constraints are equations. The right hand side element of each constraint equation
is non-negative.

 All the variables are non-negative

 The objective function is of maximization type.

Slack variables, surplus variables and artificial variables. The inequality constraints
of equations are obtained by adding or subtracting the left hand side of each constraint by
a non-negative variable.
Slack variable. A linear constraint of the form Σ a ij Xj ≤ bi can be converted into an
equality by adding a new, nonnegative variable to the left hand side of the inequality.
Such a variable is numerically equal to the difference between right hand side and left
hand side of the inequality and is known as slack variable. It represents the wastage of
resource or unutilized resource in the model as per that constraint.
Example: consider two models of washing machines are to be assembled. Model 1
takes 7 hour to assemble and model 2 takes 5 hours to assemble. The total assembly
hours available in a week are 40 hours.
Defining the variables,
Let X1 = number of model 1 to be assembled
X2 = number of model 2 to be assembled
Then time constrained model becomes
7X1 +5X2 ≤ 40
The left hand side of the inequality models the total number of hours used to assemble all
washing machines in a week, while the right hand side is the total number of hours
available for assembly in a week. But, availability is higher than the usage. Therefore,
there will be some wastage, though as small as possible, in a week. This inequality is
transformed into equation by adding a new variable S1 to the left hand side.
7X1 +5X2 + S1 = 40
Here S1 represents the number of assembly hour available but not used. Therefore it
is a wastage included in the model.
Surplus variable. A linear constraint of the form Σ aij Xj ≥ bi can be converted into
equality by subtracting a new, nonnegative variable from the left hand side of the
inequality. Such a variable is numerically equal to the difference between right hand side
and left hand side of the inequality and is known as surplus variable. It represents the
excess requirement of the resource over and above the right side quantity.

 Introduce slack variables (S1) for “≤” type of constraints.

 Introduce surplus variables (Si) and artificial variables (Ai) for “≥” type of constraints.

 Introduce only artificial variable for “=” type of constraints.

 Cost (Cj) of slack and surplus variables will be zero and that of artificial variable will
be “M”.
Solved problem 1:
Consider the LPP
Minimize Z = 4X1 + X2
Subject to 3X1 + X2 = 3
4X1 + 3X2 ≥ 6
X1 + 2X2 ≤ 3
X1, X2 ≥ 0
Initial basic feasible solution of a LPP
Consider a system of m equations in n unknowns X1, X2, X3, …… Xn
a11X1 + a12X2 + ……a1nXn = b1
a21X1 + a22X2 + ……a2nXn = b2
………………………………….
am1X1 + am2X2 + amnXn = bm
Where m ≤ n
To solve this system of equations, give n – m variables the value 0.

 The variables assigned the value zero are called “non – basic variables”.

 Remaining variables are called “basic variables”.

 Solve the equations.

 If one or more values of basic variables are also zero, then solution is said to
degenerate.

 A basic solution satisfying all the constraints is said to be feasible solution.

Solved problem 2:
Consider the system of equation
2X1 + X2 - X3 = 2
3X1 + 2X2 + X3 = 3
X1, X2, X3 ≥ 0
Procedure to solve problem by Simplex method:

 ZJ = sum of [cost of variable X its coefficient in the constraints – profit or cost

coefficient of the variable].

 Convert the problem into standard form

 Find ZJ – CJ for each variable.

 Slack and artificial variables will form basic variable for the first simplex table.
Surplus variable will never become basic variable for the first simplex table.

 Select the most negative value of Zj – Cj. That column is called key column. The
variable corresponding to the column will become basic variable for the next table.

 Divide the quantities by the corresponding values of the key column to get ratios.
Select the minimum ratio. This becomes the key row. The basic variable
corresponding to this row will be replaced by the variable found in previous step.

 The element that lies both on key column and key row is called pivotal element.

 Once an artificial variable is removed as basic variable, its column will be deleted
from next iteration.

 For maximization problem, decision variables coefficient will be same as in objective

function. For minimization problem, decision variable coefficients will have opposite
signs as compared to objective function.

 Values of artificial variables will always be M for both maximization and

minimization problems.

 The process is continued till all ZJ – CJ ≥ 0

The Simplex algorithm:
Steps:
1. Locate the most negative number in the last (bottom) row of the Simplex table,
excluding that of last column and call the column in which this number appear as the
work column.
2. Form ratios by dividing each positive number in the work column, excluding that of
the last row into the element in the same row and last column. Designate that element
in the work column that yields the smallest ratio as the pivot element. If more than
one element yields the same smallest ratio, choose arbitrarily one of them. If no
element in the work column is non negative, the program has no solution.
3. Use elementary row operations to convert the pivot element to 1 and then reduce all
other elements in the work column to 0.
4. Replace the X variable in the pivot row and first column by X variable in the first row
and pivot column. The variable, which is to be replaced is called the outgoing variable
and the variable that replaces is called the incoming variable. This new first column is
the current set of basic variables.
5. Repeat steps 1 through 4 until there are no negative numbers in the last row excluding
the last column.
6. The associated optimal value of the objective function is the number in the last row
and last column for a maximization program but the negative of this number for a
minimization problem.

Solved problem 3
Maximize Z= X1 + 9X2 + X3
Subject to:
X1 + 2X2 + 3X3 ≤ 9
3X1 + 2X2 + 2X3 ≤15
X1, X2, X3 ≥ 0
Solution:
Converting the inequalities into standard equation form by introducing the slack
variables (since both inequalities are ≤ type)
X1 + 2X2 + 3X3 + S1 = 9 (1)
3X1 + 2X2 + 2X3 + S2 = 15 (2)
Here S1 and S2 are slack variables.
The contribution of slack variables in objective function is always 0.
Thus, the objective function becomes:
Z = X1 + 9X2 + X3 +0S1 + 0S2
Converting objective function into standard form with right hand side a digit (Z j – Cj Xj =
0):
Z - X1 - 9X2 - X3 +0S1 + 0S2 = 0 (3)
Write the same in matrix form:
1 2 3 1 0 X1 9
3 2 2 0 1 X2 15
-1 -9 -1 0 0 X3 0
S1
S2
Now we can write the same into table form
Taking S1 and S2 as basic variables and others as non basic variables for the initial table:

X1 X2 X3 S1 S2 bi ratio
S1 1 2 3 1 0 9 9/2 Key row

S2 3 2 2 0 1 15 15/2
Zj - Cj -1 -9 -1 0 0 0
Pivot element
Key column
If we look at the table, the least number in the last row is -9, which is in X 2 column. Thus
X2 column is the key column. Now, dividing b i column by X2 column, we get the ratios.
The least ratio is 9/2, which lies in S 1 row. Thus, S1 row is the key row. Therefore, X2 will
become basic variable and incoming in place of S 1, which becomes non-basic variable
and outgoing.
Using elementary row operations to convert the pivot element to 1 and then reduce all
other elements in the key column to 0. For first row divide complete row by pivot element
itself to make pivot element 1. For second row, subtract first row from second row to
make key column, second row number to 0, and for bottom row, add first x 9/2 to bottom
row. Thus, the new table becomes

X1 X2 X3 S1 S2 bi
X2 1/2 1 3/2 1/2 0 9/2
S2 2 0 -1 -1 1 6
Zj - Cj 7/2 0 25/2 9/2 0 81/2

Since all elements in the bottom row are positive, it is an optimal solution.
Therefore maximum Z = 81/2 with X1 = 0, X2 = 9/2 and X3 = 0.
S2 = 6 depicts that for maximum value, 6 units of resource b2 will be non utilized.

Solved problem 4:
Maximize Z= 2X1 + 4X2 + X3 + X4
Subject to: X1 + 3X2 + X4 ≤ 4
2X1 + X2 ≤ 3
X2 + 4X3 + X4 ≤ 3
X1, X2, X3, X4 ≥ 0
Solution:
Converting into equation form by introducing slack variables
X1 + 3X2 + X4 + S1 = 4
 2X1 + X2 + S2 = 3
X2 + 4X3 + X4 + S3 = 3
And Z - 2X1 - 4X2 - X3 - X4 + 0S1 + 0S2 + 0S3 = 0
Writing into table form

X1 X2 X3 X4 S1 S2 S3 bi ratio
S1 1 3 0 1 1 0 0 4 4/3
S2 2 1 0 0 0 1 0 3 3
S3 0 1 4 1 0 0 1 3 3
Zj- Cj -2 -4 -1 -1 0 0 0 0

Seeing the table, it is observed that first row is key row and second column is key
column. A12 (the element on the intersection of key row and key column ) is the pivot
element.
Completing the elementary operations to convert pivot element as 1 and other elements as
0 in the key column, we get the new table as

X1 X2 X3 X4 S1 S2 S3 bi ratio
X2 1/3 1 0 1/3 1/3 0 0 4/3 -
S2 5/3 0 0 -1/3 -1/3 1 0 5/3 -
S3 -1/3 0 4 2/3 -1/3 0 1 5/3 5/12
Zj- Cj -2/3 0 -1 1/3 4/3 0 0 16/3

Seeing the table, it is observed that third column is the key column and third row is the
key row. The pivot element is A33. Thus S3 is the outgoing variable and X3 is the
incoming variable.
Conducting the elementary operations, we get the new table as

X1 X2 X3 X4 S1 S2 S3 bi ratio
X2 1/3 1 0 1/3 1/3 0 0 4/3 4
S2 5/3 0 0 -1/3 -1/3 1 0 5/3 1
X3 -1/12 0 1 1/6 -1/12 0 1/4 5/12 -
Zj- Cj -3/4 0 0 1/2 5/4 0 1/4 23/4

Seeing the table, it is observed that first column is the key column and second row is the
key row. The pivot element is A21. Thus S2 is the outgoing variable and X1 is the
incoming variable.
Conducting the elementary operations, we get the new table as

X1 X2 X3 X4 S1 S2 S3 bi ratio
X2 0 1 0 2/5 2/5 1/5 0 1
 X1 1 0 0 -1/5 -1/5 3/5 0 1
X3 0 0 1 3/20 -1/10 1/20 1/4 1/2
Zj- Cj 0 1 0 7/20 11/10 9/20 1/4 13/2

Since all elements in the bottom row are positive, it is an optimal solution.
Therefore maximum Z = 13/2 with X1 = 1, X2 = 1, X3 = ½ and X4 = 0 .

Solved problem 5:
A manufacturing firm has discontinued production of a certain unprofitable product line.
This created considerable excess production capacity. Management is considering
devoting this excess capacity to one or more of three products 1, 2 and 3. The available
capacity on the machines which might limit output is given below:

Machine type Available time (in machine hours per week)

Milling machine 250
Lathe 150
Grinder 50

The number of machine hours required for each units of the respective product is given
below;

Machine type Productivity (in machine hours per unit)

Product 1 Product 2 Product 3
Milling machine 8 2 3
Lathe 4 3 0
Grinder 2 0 1

The unit profit would be rupees 20, Rupees 6 and rupees 8 for products 1, 2 and 3
respectively. Find how much of each product the firm should produce in order to
maximize profit?
Solution:
The mathematical program for this problem will be
Maximize Z = 20 X1 + 6 X2 + 8 X3
Subject to the constraints
8 X1 + 2 X2 + 3 X3 ≤ 250
4 X1 + 3 X2 + 0 X3 ≤ 150
2X1 + 0 X2 + 1 X3 ≤ 50
X1, X2, X3 ≥ 0
Solution
Converting into equation form by introducing slack variables
8 X1 + 2X2 + 3 X3 + S1 = 250
4 X1 + 2 X2 + S2 = 150
2 X1 + 1 X3 + S3 = 50
And Z – 20 X1 – 6 X2 – 8 X3 + 0S1 + 0S2 + 0S3 = 0
Writing into table form

X1 X2 X3 S1 S2 S3 bi ratio
S1 8 2 3 1 0 0 250 125/4
S2 4 3 0 0 1 0 150 75/2
S3 2 0 1 0 0 1 50 25
Zj- Cj -20 -6 -8 0 0 0 0
The table after first iteration

X1 X2 X3 S1 S2 S3 bi ratio
S1 0 2 -1 1 0 -4 50 25
S2 0 3 -2 0 1 -2 50 50/3
X1 1 0 1/2 0 0 1/2 25 -
Zj- Cj 0 -6 2 0 0 10 500

The table after second iteration

X1 X2 X3 S1 S2 S3 bi ratio
S1 0 0 1/3 1 -2/3 -8/3 50/3 50
X2 0 1 -2/3 0 1/3 -2/3 50/3 -
X1 1 0 1/2 0 0 1/2 25 50
Zj- Cj 0 0 -2 0 2 6 600

The table after third iteration

X1 X2 X3 S1 S2 S3 bi ratio
S1 0 0 0 1 -1/2 -3 25
X2 1 1 0 0 1/4 0 75/2
X3 2 0 1 0 0 1 50
Zj- Cj 2 0 0 0 2 8 625

Since all elements in the bottom row are positive, it is an optimal solution.
Therefore maximum Z = 625 with X1 = 0, X2 = 75/2, and X3 = 50 .
Penalty cost method or big-M method:
Consider a LPP when at least one of the constraints is of the type ≥ or =. While
expressing in the standard form, add a non negative artificial variable to each of such
constraints. In objective function, incorporate them with large positive coefficients for
minimization program and with very large negative coefficients in maximization
program. These coefficients are denoted by ± M.
The modification in procedure:
1. The last row of the simplex table is decomposed into two rows, the first of which
involves those terms not containing M, while the second involves those containing M.
2. The step 1 of the Simplex method is applied to the last row created in the above
modification and followed by steps 2, 3 and 4 until this row contains no negative
elements. Then step 1 of simplex algorithm is applied to those elements next to the
last row that are positioned over zero in the last row.
3. Whenever an artificial variable ceases to be basic, it is removed from the first column
of the table as a result of step 4. It is also deleted from the top row of the table as is
the entire column under it.
4. The last row is removed from the table whenever it contains all zeroes.
5. If non zero artificial variables are present in the final basic set, then the program has
no solution. In contrast, zero valued artificial variables in the final solution may exist
when one or more of the original constraint equations are redundant.
Solved problem 6:
Use penalty cost method to
Maximize Z = 2X1 + 5X2
Subject to:
X1 ≤ 40
X2 ≤ 30
X1 + X2 ≥ 60
X1, X2 ≥ 0
Solution
Converting into equation form by introducing slack, surplus and artificial variables
X1 + S1 = 40 (1)
X2 + S2 = 30 (2)
X1 + X2 – S3 + A1 = 60 (3)
And Z = 2X1 + 5X2 + 0S1 + 0S2 + 0S3 – MA1
As per equation (3)
A1 = 60 - X1 - X2 + S3
Putting the value of A1 in the objective function
Z = 2X1 + 5X2 + 0S1 + 0S2 + 0S3 – M (60 - X1 - X2 + S3)
Or Z – (M + 2)X1 – (M + 5)X2 + 0S1 + 0S2 + MS3 = - 60M
Writing into table form

X1 X2 S1 S2 S3 A1 bi ratio
S1 1 0 1 0 0 0 40 -
S2 0 1 0 1 0 0 30 30
A1 1 0 0 -1 1 60 60
1
Zj- Cj -2 -5 0 0 0 0 0
-M -M 0 0 M -60M

Table after first iteration

X1 X2 S1 S2 S3 A1 bi ratio
S1 1 0 1 0 0 0 40 40
X2 0 1 0 1 0 0 30 -
A1 1 0 0 -1 -1 1 30 30
Zj- Cj -2 0 0 5 0 0 150
-M 0 0 M M 0 -30M

Thus, X1 is the entering variable and A1 is the leaving variable.

Table after second iteration

X1 X2 S1 S2 S3 bi ratio
S1 0 0 1 1 1 10 10
X2 0 1 0 1 0 30 -
X1 1 0 0 -1 -1 30 -
Zj- Cj 0 0 0 3 -2 210
0 0 0 0 0 0

Thus, S1 is the leaving variable and S3 is the entering variable.

Table after third iteration

X1 X2 S1 S2 S3 bi ratio
S3 0 0 1 1 1 10
 X2 0 1 0 1 0 30
X1 1 0 1 0 0 40
Zj- Cj 0 0 2 5 0 230
0 0 0 0 0 0

Since all numbers in last two rows are positive, this is an optimum solution.
The solution is:
Z = 230
X1 = 40
X2 = 30
And S3 = 10

Two phase method

The drawback of the penalty cost method is the possible computational error that
could result from assigning a very large value to the constant m. To overcome this
difficulty, use of M is eliminated by solving the problem in two phases:
Phase 1.
Formulate the problem by eliminating the original objective function by the sum of
the artificial variables for a minimization problem and the negative of the sum of the
artificial variables for maximization problem. The resulting objective function is
optimized by the Simplex method with the constraints of the original problem. If the
problem has the feasible solution, the optimal value of the new objective function is zero
(which indicates that all artificial variables are zero). Then we proceed to phase2.
Otherwise, if the optimal value of the new objective function is non zero, the problem has
no solution and the method terminates.
Phase 2.
Use the optimum solution of the phase 1 as the starting solution of the original
problem. Then the objective function is taken without the artificial variables and is solved
by simplex method.
Solved problem 7:
Use the two phase method to
Maximize Z = 3X1 – X2
Subject to: 2X1 + X2 ≥ 2
X1 + 3X2 ≤ 2
X2 ≤ 4
 X1, X2 ≥ 0
Solution;
Rewriting in the standard form
Maximize Z = 3X1 – X2 + 0S1 + 0S2 + 0S3 – MA
Subject to the constraints
2X1 + X2 – S1 + A= 2
X1 + 3X2 + S2 = 2
X2 + S3 = 4
Phase 1: consider the new objective
Maximize Z* = -A = 2X1 + X2 – S1 – 2
Or Z* - 2X1 - X2 + S1 = – 2
Subject to the constraints
2X1 + X2 – S1 + A= 2
X1 + 3X2 + S2 = 2
X2 + S3 = 4
X1, X2, S1, A, S2, S3 ≥ 0
Table below depicts the initial simplex table arrived through the simplex method

X1 X2 S1 S2 S3 A bi ratio
A 2 1 -1 0 0 1 2 1
S2 1 3 0 1 0 0 2 2
S3 0 1 0 0 1 0 4 -
Zj- Cj -2 -1 1 0 0 0 -2

The first iteration gives the following table:

X1 X2 S1 S2 S3 A bi ratio
X1 1 1/2 -1/2 0 0 1/2 1
S2 0 5/2 1/2 1 0 -1/2 1
S3 0 1 0 0 1 0 4
Zj- Cj 0 0 0 0 0 1 0

Since there are no negative elements in the last row, phase 1 is complete.
The optimal solution to the new objective is Z* = 0
Phase II
 Consider the original objective function
Maximize Z = 3X1 – X2 + 0S1 + 0S2 + 0S3
Or Z - 3X1 + X2 + 0S1 + 0S2 + 0S3 = 0
Subject to X1 + ½ X2 – ½ S1 = 1
5/2 X2 + ½ S1 + S2 = 1
X 2 + S3 = 4
X1, X2, S1, S2, S3 ≥ 0
Table below depicts the corresponding simplex table

X1 X2 S1 S2 S3 bi ratio
X1 1 1/2 -1/2 0 0 1 1
S2 0 5/2 1/2 1 0 1 -
S3 0 1 0 0 1 4 -
Zj- Cj -3 1 0 0 0 0

After first iteration, the table is:

X1 X2 S1 S2 S3 bi ratio
X1 1 1/2 -1/2 0 0 1 -
S2 0 5/2 1/2 1 0 1 2
S3 0 1 0 0 1 4 -
Zj- Cj 0 5/2 -3/2 0 0 3

Proceeding to next iteration, we get the following table:

X1 X2 S1 S2 S3 bi ratio
X1 1 3 0 1 0 2 -
S2 0 5 1 2 0 2 -
S3 0 1 0 0 1 4 -
Zj- Cj 0 10 0 3 0 6

Since all the elements of the last row are non-negative, the current solution is optimal.
The maximum value of the objective function is Z = 6 which is attained for X1 = 2, X2 = 0
Solved problem 8:
Maximize Z = 3X1 + 2X2
Subject to 2X1 + X2 ≤ 2
3X1 + 4X2 ≥ 12
 X1, X2 ≥ 0
Solution:
Rewriting in the standard form
Maximize Z = 3X1 + 2X2 + 0S1 + 0S2 – MA
Subject to the constraints
2X1 + X2 + S1 = 2
3X1 + 4X2 - S2 + A = 12
X1, X2, S1, S2, A1 ≥ 0
Phase I:
Consider the new objective
Maximize Z* = -A = 3X1 + 4X2 – S2 – 12
Or Z* - 3X1 - 4X2 + S2 = – 12
Subject to the constraints
2X1 + X2 – S1 = 2
3X1 + 4X2 - S2 + A = 12
X1, X2, S1, A, S2 ≥ 0
Table below depicts the initial simplex table arrived through the simplex method

X1 X2 S1 S2 A bi ratio
S1 2 1 1 0 0 2 2
A 3 4 0 -1 1 12 3
Zj- Cj -3 -4 0 1 0 -12

The first iteration gives the following table:

X1 X2 S1 S2 A bi
X2 2 1 1 0 0 2
A -5 0 -4 -1 1 4
Zj- Cj 5 0 4 1 0 -4

Since all the elements of the last row are non-negative, the procedure is complete. But the
existence of non-zero artificial variables in the basic set indicates that the problem is
infeasible.
Solved problem 9:
Minimize Z = 5X1 + 6X2
Subject to:
2X1 + 5X2 ≥ 1500
3X1 + X2 ≥ 1200
X1, X2 ≥ 0
Solution:
Introducing the surplus and artificial variables in the constraints
Minimize Z = 5X1 + 6X2 + 0S1 + 0S2 + MA1 + MA2
Subject to
2X1 + 5X2 – S1 + A1 = 1500
3X1 + X2 - S2 + A2 = 1200
Since the objective function should not involve artificial variables A 1 and A2, we find the
values of A1 and A2 from the constraint equations and put in the objective function.
A1 = 1500 - 2X1 - 5X2 – S1

Terminal question 1:
Maximize Z = 3X1 – X2
Subject to:
X1 + 3X2 ≤ 3
2X1 + X2 ≥ 2
X2 ≤ 4
X1, X2 ≥ 0
Terminal question 1:
Maximize Z = 6X1 + 7X2
Subject to:
X1 + 3X2 ≥ 12
3X1 + X2 ≥ 12
X1 + X2 ≥ 8
X1, X2 ≥ 0
 Sensitivity analysis

In linear programming, all models are assumed to be constant. However; in real life
situations, the decision environment is always dynamic. Therefore, it is important for the
management to know how profit would be affected by an increase or decrease in the
resource level, by change in the cost of raw materials. Such an investigation is known as
sensitivity analysis or post optimality analysis.
The management of a company rarely restricts its interest to the numerical values of
an optimal solution. Actually it is interested in finding out the impact of changes in input
parameters in the optimal solution. Such analysis of impact of amendment in parameters
is called as sensitivity analysis.
The results of sensitivity analysis establishes upper and lower bounds for input
parameters values within which they can vary without causing major changes in the
optimal solution.
Sensitivity analysis allows to figure out which data offers a significant impact on the
results. This in turn allows to concentrate on getting accurate data for those items or at
least running through several scenarios with various values of the crucial data to get an
idea of the range of possible outcomes.
Sensitivity analysis is important because of the following reasons:

 Values of LP parameters might change

 LP parameters have an uncertainty factor attached to them.
Any change in a LP’s parameter affects the optimality of the basic variable. This is
because of the following two reasons.

 When a variable (or number of variables) in row Zj – Cj has a negative coefficient, we

can obtain a better basic feasible solution of larger Z value by pivoting a non-basic
variable with a negative coefficient in row Z j – Cj. If this occurs, the basic variable
becomes a sub-optimal basis.
 A constraint (or number of constraints) may have a negative RHS. In such a case, at
least one number of basic variable will now be negative and will no longer yield a
basic feasible solution. If this occurs, it can be said that the basic variable is now an
infeasible basis.
The following six types of changes in LP’s parameters cause changes in the optimal solution:

 Changing the objective function coefficient of a non-basic variable

 Changing the objective function coefficient of a basic variable
 Changing the right hand side of of a constraint
 Changing the column of a non-basic variable.
 Adding a new variable or activity
 Adding a new constraint.
Example.
Luminous lamp produces three types of lamps A, B and C. These lamps are processed on
three machines M1, M2 and M3. The full technology and input restrictions are as given
below:-

product machines Profit per unit

M1 M2 M3
A 10 7 2 12
B 2 3 4 3
C 1 2 1 1
Available time 100 77 80

Let X1, X2 and X3 be the number of units to be produced. Then the Linear Programming
model for the above problem will be as given below:
Maximize Z = 12X1 + 3X2 + X3
Subject to constraints:
10X1 + 2X2 + X3 ≤ 100
7X1 + 3X2 + 2X3 ≤ 77
2X1 + 4X2 + X3 ≤ 80
X1, X2, X3 ≥ 0
On solving the above problem by the simplex, we get
X1 = 73/8, X2 = 35/8 and X3 = 0 and the optimal value Z = 981/8. The final table of
simplex table is as given below:
Table 3.7: The final simplex table

Basic X1 X2 X3 S1 S2 S3 Solution
variables values bi
X1 1 0 -1/16 3/16 -1/8 0 73/8
X2 0 1 13/16 -7/16 5/8 0 35/8
S3 0 0 -17/8 11/8 -9/4 1 177/4
Zj - C j 0 0 11/16 15/16 3/8 0 981/8
Cj* 12 3 27/16 15/16 3/8 0 -

Changes in profit contribution (Cj) of a non basic variable. A non basic variable can
be brought into basic only if its contribution rate becomes attractive. Therefore, it is
needed to determine the upper limit of the profit contribution (C j) of each non basic
variable. The reverse is true for a minimization problem. Consider the above final
simplex table.
From the above, it can be seen that profit contribution for product C is Rupees 1, which is
not greater than Cj*. Therefore, to bring X3 into the basis, its profit contribution rate must
exceed Rupees 27/16 to make Zj – Cj value negative or zero. That is Zj – Cj ≤
0.specifically,
  If Cj* - C > Zj – Cj, then a new optimal solution must be derived
 If Cj* - C = Zj – Cj, then alternative optimal solution exists
 If Cj* - C > Zj – Cj, then current optimal solution remains unchanged.
In case of Z3 = 1 and Z3 – C3 = 11/16
C3* - 1 ≥ 11/16
C3* ≥ 11/16 + 1 = 27/16
X3 can be introduced in the basis, if its contribution rate C3* increases at least Rupees
27/16. If it increases beyond that, then the current solution will no longer be optimal.
Changes in profit contribution (Cj) of a basic variable. The optimal solution will
remain insensitive till some range of profit contribution. If it is below minimum value, the
profit will be very less and re-optimization to be done. Similarly, if it is above the
maximum value, the dominance of the product is much higher that it may replace the
existing product as basic variable. Thus, reoptimisation to be carried out. To find out the
lower and upper limits for contribution, the following rule is applicable.
-Min [(Zj – Cj) / Yij; if Yij > 0] ≤ C* - Z ≤ min [(Zj – Cj) / Yij; if Yij < 0]
Where i to correspond to basic variable for product and not for slack
variables, and j to correspond to non basic variables only.
Let us see the case of product A.
In X1 row X3 and X5 columns have got the values -1/16 and -1/8. Therefore,
Min [(11/16) / -(-1/16), (3/8) / -(-1/8)] = min (11,3) = 3
Corresponding to non-basic variable X4, Y14 = 3/16. Therefore,
Min [15/16/3/16] = 5
-5 ≤ C1* - 12 ≤ 3
Or 7 ≤ C1* ≤ 15
Thus we can say that the optimal solution is insensitive as long as the changed profit
contribution for product A remains between Rupees 7 and Rupees 15 per product.
Change in available resources. If the available resources change, it is needed to investigate
whether a previous optimal solution remains feasible or not. For long term planning, it is
important to know the bounds within which each available resource (for instant machine
hours) can vary without causing violent changes in the current optimal solution.
To illustrate this, divide bi column by the corresponding coefficient in the slack variable
column for particular machinery. For example, for machine M1 divide by S1 column.

bi S ratio
73/8 3/16 146/3
35/8 -7/16 -10
177/4 11/8 354/11
The least positive ratio (354/11) indicates the number of hours of machine M 1 that can be
decreased. The least negative ratio (-10) indicates the increase in the machine hours.
Calculating the range.
(Total availability)i – least positive ratio = lower limit
(Total availability)i – least negative ratio = upper limit
For machine M1
Lower limit = 100 – 354/11 = 746/11
Upper limit = 100 – (-10) = 110
Therefore range of hours for machine M1 is 746/11 to 110.
The range of hours for machine M2 and M3 can be calculated in the same manner.

CHAPTER – 4
DUALITY IN LINEAR PROGRAMMING
4.1 Introduction
Every Linear Programming Problem (LPP) is associated with another linear programming
problem involving the same data and optimal solutions. Such two problems are said to be
duals of each other. One problem is called the primal problem, while the other problem is
called the dual.
Primal dual
The dual formulation is derived from the same data and solved in a manner similar to the
original ‘primal’ formulation. In other words, it can be said that dual is inverse of the
primal formulation because of the following reasons.
Reason.

 If more constraints and very less number of variables, the efforts can be reduced

 Interpretation of the dual variable from economic point of view is extremely useful.

 The concept of duality is useful to obtain additional information about the variation in
the optimal solution. These changes could be effected in the constraints coefficient,
in resource availabilities and / or objective function coefficient. This effect is termed
as post optimality or sensitivity analysis.
Characteristics of dual solutions:
If the primal problem possesses a unique non – degenerate, optimal solution, then the
optimal solution to the dual is unique. The other cases are:
 When the primal has a degenerate optimal solution, the dual has multiple optimal
solutions

 When the primal has multiple optimal solutions, the dual will have degenerate optimal
solution.

 When the primal is unbounded, the dual is infeasible.

 When the primal problem is infeasible, the dual is unbounded or infeasible.

Formulation of dual problems:
Consider the following LPP:
Maximize Z= C1 X1 + C2 X2 + -------CN XN
Subject to the constraints:
a11 X1 + a12 X2 + ------- a1n Xn ≤ b1
a21 X1 + a22 X2 + ------- a2n Xn ≤ b2
---------------------------------------------
am1 X1 + am2 X2 + ------ amn Xn ≤ bm
X1, X2, ----- Xn ≥ 0.
Guide lines:

 The maximization problem in primal becomes minimization problem in dual and vice
versa.

 (≤) types of constraints in the primal become (≥) type of constraints in the dual and
vice versa.

 The coefficients C1, C2, ------, CN in the objective function of the primal become b 1, b2,
------, bm in the constraints of the dual.

 The constants b1, b2, ------, bm in the constraints of primal become C1, C2, ------, Cn in
the objective function of the dual.

 If the primal has n variables and m constraints, the dual will have m variables and n
constraints.

 The variables in both primal and dual are non negative.

Thus the dual problem will be:
Minimize W= b1 Y1 + b2 Y2 + -------bm Ym
Subject to the constraints:
a11 Y1 + a21 X2 + ------- am1 Ym ≥ C1
a12 Y1 + a22 Y2 + ------- am2Ym ≥ C2
 ---------------------------------------------
a1N Y1 + a2N Y2 + ------ amn Ym ≥ Cn
Y1, Y2, ----- Ym ≥ 0.
The primal LPP should be in canonical form, and then it is easy to form dual LPP.
Solved problem 1:
Write the dual of
Maximize
Z= 4X1 + 5X2
Subject to:
3X1 + X2 ≤ 15
X1 + 2X2 ≤ 10
5X1 + 2X2 ≤ 20
X1, X2 ≥ 0
Solution:
The given problem is in standard form. Therefore, the dual is:
Minimize W = 15Y1 + 10Y2 + 20Y3
Subject to the constraints;
3Y1 + Y2 + 5Y3 ≥ 4
Y1 + 2Y2 + 2Y3 ≥ 5
Y1, Y2, Y3 ≥ 0
Solved problem 2:
Write the dual of
Minimize Z= 10X1 + 12X2
Subject to 2X1 + 3X2 ≥ 10
5X1 + 6X2 ≥ 20
X1 + 2X2 ≥ 15
2X1 + 3X2 ≥ 12
X1, X2 ≥ 0
Solution:
The given problem is in the standard form. Therefore its dual problem is:
Maximize
W = 10Y1 + 20Y2 + 15Y3 + 12Y4
Subject to the constraints:
2Y1 + 5Y2 + Y3 + 2Y4 ≤ 10
3Y1 + 6Y2 + 2Y3 + 3Y4 ≤ 12
Y1, Y2, Y3 and Y4 ≥ 0
Solved problem 3:
Write the dual of
Maximize Z= 100X1 + 200X2
Subject to:
3X1 - 10X2 ≥ 15
4X1 + 15X2 ≤ 20
X1, X2 ≥ 0
Solution:
First convert the problem into standard form.
3X1 – 10X2 ≥ 15 is to be converted into “≤” type constraint. Thus
-3X1 + 10X2 ≤ -15
Therefore its dual problem is:
Minimize W = -15Y1 +20Y2
Subject to the constraints:
-3Y1 + 4Y2 ≥ 100
10Y1 + 15Y2 ≥ 200
Y1, Y2 ≥ 0
Solved problem 4:
Write the dual of
Maximize Z= 40X1 + 30X2
Subject to:
10X1 + 6X2 ≤ 15
 5X1 - 7X2 ≥ -10
X1 + X2 = 9
X1, X2 ≥ 0
Solution:
Converting the second equation into “≤” type of constraint.
-5X1 + 7X2 ≤ 10
The third equation X1 + X2 = 9 is changed into
X1 + X2 ≤ 9
and X1 + X2 ≥ 9 this is same as –X1 –X2 ≤ -9
therefore the given problem is:
Maximize Z = 40X1 + 30X2
Subject to the constraints:
10X1 + 6X2 ≤ 15
-5X1 + 7X2 ≤ 10
X1 + X2 ≤ 9
–X1 –X2 ≤ -9
X1, X2 ≥ 0
Therefore the dual is:
Minimize W = 15Y1 + 10Y2 + 9Y3’ – 9Y3’’
Subject to the constraints:
10Y1 – 5Y2 + Y3’ – Y3’’ ≥ 40
6Y1 + 7Y2 + Y3’ – Y3’’ ≥ 30
Let Y3’ – Y3’’ = Y3
Then the dual problem becomes;
Minimize W = 15Y1 + 10Y2 + 9Y3
Subject to the constraints:
10Y1 – 5Y2 + Y3 ≥ 40
6Y1 + 7Y2 + Y3 ≥ 30
Y1, Y2 ≥ 0
Y3 is unrestricted in sign
Solved problem 5:
Write the dual of
Minimize Z= 12X1 + 15X2
Subject to:
5X1 + 3X2 ≥ 10
X1 + X2 ≤ 5
X1 ≥ 0
X2 is unrestricted in sign
Solution:
Since X2 is unrestricted in sign, it is converted as
X2 = X2’ – X2’’
Therefore its standard form is:
Minimize Z= 12X1 + 15X2’ – 15X2’’
Subject to:
5X1 + 3(X2’ – X2’’) ≥ 10
-[X1 + (X2’ – X2’’)] ≤ -5
Or -X1 - X2’ + X2’’ ≤ -5
Therefore the dual becomes
Maximize W = 10Y1 – 5Y2
Subject to the constraints:
5Y1 – Y2 ≤ 12 (1)
3Y1 – Y2 ≤ 15 (2)
-3Y1 + Y2 ≤ -15 (3)
No. (3) constraint is 3Y1 – Y2 ≥ 15 (4)
Constraints (2) and (4) give
3Y1 – Y2 = 15
The dual problem is:
Maximize W = 10Y1 – 5Y2
Subject to the constraints:
5Y1 – Y2 ≤ 12
3Y1 – Y2 = 15
Y1 ≥ 0 and Y2 is unrestricted in sign.

Solved problem 6:
Write the dual of
Minimize Z= 3X1 - 2X2 + 4 X3
Subject to:
3X1 + 5X2 + 4X3 ≥ 7
6X1 + X2 + 3X3 ≥ 4
7X1 - 2X2 - X3 ≤ 10
X1 - 2X2 + 5X3 ≥ 3
4X1 + 7X2 - 2X3 ≥ 2
X1, X2, X3 ≥ 0
Solution:
Since the problem is of maximization type all constraints should be ≥ type. Thus,
constraint equation no. (3) is to be changed. It becomes
- 7X1 + 2X2 + X3 ≥ -10
Thus, the dual problem becomes:
Maximize W = 7Y1 + 4Y2 – 10Y3 + 3Y4 + 2Y5
Subject to the constraints;
3Y1 + 6Y2 – 7Y3 + Y4 + 4Y5 ≤ 3
5Y1 + Y2 + 2Y3 - 2Y4 + 7Y5 ≤ -2
4Y1 + 3Y2 + Y3 + 5Y4 - 2Y5 ≤ 4
Y1, Y2, Y3, Y4, and Y5≥ 0
Solved problem 7:
Write the dual of
Maximize Z= 12X1 + 10X2
Subject to:
 2X1 + 3X2 ≤ 18
2X1 + X2 ≤ 14
X1, X2 ≥ 0
Solution:
Since the given problem is already in standard form, the dual for this problem will be:
Minimize W = 18Y1 + 14Y2
Subject to the constraints:
2Y1 + 2Y2 ≥ 12
3Y1 + Y2 ≥ 10
And Y1, Y2 ≥ 0
Solved problem 8:
Write the dual of
Maximize Z= 2X1 + 5X2 + 6X3
Subject to:
5X1 + 6X2 – X3 ≤ 3
-2X1 + X2 + 4X3 ≤ 4
X1 – 5X2 +3X3 ≤ 1
-3X1 – 3X2 + 7X3 ≤ 6
X1, X2, X3 ≥ 0
Solution:
Since the given problem is already in the standard form, the dual of the given problem is:
Minimize W = 3Y1 + 4Y2 + Y3 + 6Y4
Subject to the constraints:
5Y1 – 2Y2 + Y3 – 3Y4 ≥ 2
6Y1 + Y2 - 5Y3 – 3Y4 ≥ 5
-Y1 + 4Y2 + 3Y3 + 7Y4 ≥ 6
Y1, Y2, Y3, Y4 ≥ 0
Solved problem 9:
Write the dual of
Minimize Z= 5X1 - 6X2 + 4X3
Subject to: 3X1 + 4X2 + 6X3 ≥ 9 (1)
X1 + 3X2 + 2X3 ≥ 5 (2)
7X1 – 2X2 - X3 ≤ 10 (3)
X1 – 2X2 + 4X3 ≥ 4 (4)
2X1 + 5X2 - 3X3 ≥ 3 (5)
X1, X2, X3 ≥ 0
Solution:
Since the problem is of minimization, all constraints in standard form have to be “≥” type.
Therefore, constraint no. (3) has to be converted. It becomes:
-7X1 + 2X2 + X3 ≥ -10
Thus, the dual of this problem will be:
Maximize W = 9Y1 + 5Y2 – 10Y3 + 4Y4 + 3Y5
Subject to the constraints:
3Y1 + Y2 – 7Y3 + Y4 + 2Y5 ≤ 5
4Y1 + 3Y2 + 2Y3 - 2Y4 + 5Y5 ≤ -6
6Y1 + 2Y2 + Y3 + 4Y4 - 3Y5 ≤ 4
Y1, Y2, Y3, Y4, Y5 ≥ 0

Economic interpretation of duality

The dual variable Y1 represents the worth per unit of resource I (Also called as shadow
prices or dual prices). In an industry, the problem is of two kind, maximization of profit
or minimization of allocation of resources. Thus, The linear programming problem can
also be thought of as a resource allocation model where its objective is to maximize
revenue or profit subject to limited resources. The associated dual problem offers
interesting economic interpretation of the LP resource allocation model. Consider the
general representation of the general primal and dual problems where primal takes the
role of a resource allocation model.

Primal Dual
Maximize Minimize

Z = Σj=1n CjXj W = Σi=1m biYi

Subject to Subject to
 Σj=1n aijXj ≤ bi, I = 1, 2, ……, m Σi=1m aijYi ≤ Cj, j = 1, 2, ……, n

Xj ≥ 0, j = 1, 2, ….., n Yi≥ 0, I = 1, 2, ……., m

In the above resource allocation model, the primal problem has ‘n’ economic activities
and ‘m’ resources. The coefficient Cj in the primal represents the profit per unit of activity
j. Resource I, whose maximum availability is bi, is consumed at the rate aij units per unit
of activity j.

 If Z=W, we have optimal solution.

 If Z ≤ W, solution is non-optimal.

 If Z ≥ W, the solution is non-stable.

Thus, for any pair of feasible primal and dual solutions,
(Objective value in the maximization problem) ≤ (Objective value in the minimization
problem)
At the optimum, the relationship holds as a strict equation. Here the sense of optimization
is very important.
Hence for any two primal and dual feasible solutions, the value of the objective functions,
when finite, must satisfy the following inequality.

Z = Σj=1n CjXj ≤ Σi=1m biYi = W

The strict equality Z = W, holds true when both the primal and dual solutions are optimal.
Consider the optimal condition Z = W. given that the primal problem represents a resource
allocation model, you can think of Z as representing the profit in rupees. b i represents the
number of units available of the resource i. therefore we can express the equation Z = W as
profit = Σ (units of resource) x (profit per unit resource of i)

This means that the dual variable Yi, represents the worth per unit of resource i. variable Y i
are also called as dual prices, shadow prices and simplex multipliers.

With the same logic, the inequality Z < W associated with any two feasible primal and dual
solution is interpreted as profit < (worth of resources).

This relationship implies that as long as the total return from all the activities is less than the
worth of the resources, the corresponding primal and dual solution are not optimal.
Optimality is achieved only when the input equals the output (profit). Economically, the
system is said to be unstable (non optimal) when input (worth of resources) exceeds the
output (return). Stability occurs only when the two quantities are equal.

Terminal questions
Write the dual of following problems:
Q1.
Maximize Z = 7X1 + 5X2
Subject to:
X1 + 2X2 ≤ 6
4X1 + 3X2 ≤ 12
X1, X2 ≥ 0
Q2.
Maximize Z= 3X1 + 4X2
Subject to:
5X1 + 4X2 ≤ 200
3X1 + 5X2 ≤ 150
5X1 + 4X2 ≥ 100
8X1 + 4X2 ≥ 80
X1, X2 ≥ 0
Q3.
Maximize Z= 2X1 + X2
Subject to: 4X1 + 3X2 ≤ 12
4X1 + X2 ≤ 8
4X1 - X2 ≤ 8
X1, X2 ≥ 0
Answers to terminal questions.
Q1. Minimize
W = 6Y1 + 12Y2
Subject to constraints:
Y1 + 4Y2 ≥ 7
2Y1 + 3Y2 ≥ 5
Y1, Y2 ≥ 0
Q2. Minimize
W = 200Y1 + 150Y2 – 100Y3 – 80Y4
Subject to the constraints:
5Y1 + 3Y2 – 5Y3 – 8Y4 ≥ 3
4Y1 + 5Y2 – 4Y3 – 4Y4 ≥ 4
And Y1, Y2, Y3, Y4, ≥ 0
Q3. Minimize
W = 12Y1 + 8Y2 +8Y3
Subject to the constraints:
4Y1 + 4Y2 + 4Y3 ≥ 2
3Y1 + Y2 - Y3 ≥ 1
Y1, Y2, Y3 ≥ 0

CHAPTER – 5
TRANSPORTATION MODEL
5.1 Introduction
Many practical problems in operations research can be broadly formulated as linear
programming problem, for which the simplex method is a powerful technique for solution
of these problems. But this is a general method and cannot be used for specific types of
problems like,

 Transportation model
 Transshipment model
 The assignment models.
The above models are also basically allocation models. We can adopt the complex
techniques to solve them, but easier algorithms have been developed for solution of such
problems. The following sections deal with the transportation problems and their
streamlined procedure for solution.
5.2 transportation model
Transportation model is an important class of linear programs. For a given supply at each
source and a given demand at each destination, the model studies the minimization of the
cost of transporting a commodity from a number of sources to several destinations.
The transportation problem involves m sources, each of which has available a i (I = 1, 2,
…., m) units of homogeneous product and n destinations, each of which requires b j (j = 1,
2, …., n) units of products. Here a i and bj are positive integers. The cost Cij of transporting
one unit of the product from the i th source to the jth destination is given for each I and j. the
objective is to develop an integral transportation schedule that meets all demands from
the inventory at a minimum total transportation cost. It is assumed that the total supply
and total demand are equal.
 ∑i=1m ai = ∑j=1n bJ
The condition is generated by creating either a fictious destination with a demand equal to
surplus if total demand is less than the total supply or a dummy source with a supply
equal to the shortage if total demand exceeds total supply. The cost of transportation from
the fictitious destination to all sources and from all destinations to the fictitious sources
are assumed to be zero so that total cost of transportation will remain the same.
5.3 Formulation of transportation problem
The standard mathematical model for transportation problem is as follows.
Let Xij be the number of units of the homogeneous product to be transported from sources
i to the destination j.
Then objective is to:
Minimize Z = ∑i=1m ∑j=1n CiJ XiJ
Subject to:
∑j=1n XiJ = ai , i = 1,2,------m
∑i=1m XiJ = bJ ; J = 1,2,------n
With all XiJ ≥ 0 and integers.
Theorem: A necessary and sufficient condition for the existence of a feasible solution to
the transportation problem is that:
∑i=1m ai = ∑j=1n bJ
5.4 Transportation Algorithm (MODI Method)
The transportation algorithm is the Simplex method specialized to the format of table. It
involves:

 Finding an initial basic feasible solution.

 Testing the solution for optimality.

 Improving the solution, when it is not optimal.

 Repeating steps 2 and 3 until the optimal solution is obtained.

The solution to T.P is obtained in two stages:
In the first stage, find the basic feasible solution using any of the following methods:-

 North-west corner rule

 Matrix minimum method or least cost method

 Vogel’s approximation method (VAM)

In the second stage, test the basic feasible solution for its optimality by following
method:-

 Modi method

 Stepping stone method.

Table 5.1: Transportation table

D1 D2 Dn Supply
S1 C11 C12 C1n a1
X11 X12 X1n
S2 C21 C22 C2n a2
X21 X22 X2n
S3 C31 C32 C3n a3
X31 X32 X3n

Sm Cm1 Cm2 Cmn am

Xm1 Xm2 Xmn
Demand b1 b2 bn Σai = Σbj

5.3 The Initial Basic Feasible Solution

Let us consider a TP involving m-origins and n-destinations. Since the sum of origin
capacities equals the sum of destination requirements, a feasible solution always exists.
Any feasible solution satisfying m + n – 1 of the m + n constraints is a redundant one and
hence it can be deleted. This also means that a feasible solution to a TP can have only m +
n – 1 positive component, otherwise the solution will degenerate.
It is always possible to assign an initial feasible solution to a TP, satisfying all the rim
requirements. This can be achieved either by inspection or by following some simple
rules. It can be started by imagining that the transportation table is blank that is initial X ij
= 0. The simplest procedures for allocation are discussed in the following section.

North west corner rule.

 Step 1. Make first assignment in cell (1,1), the upper left hand (north west corner) of
table. The amount allotted is
X11 = min ( a1, b1 )
Either the capacity of origin O1 is used up or the requirement at destination D1 is
satisfied or both. This value of X11 is entered in the upper left hand corner of cell (1, 1) in
the transportation table.

 Step 2. If b1 > a1, the capacity of origin O1 is exhausted and the requirement at
destination D1 is still not satisfied. Then at least one variable in the first column will
have to take on a positive value. Move down vertically to the second row and make
the second allocation of magnitude:
 X21 = min ( a2, b1 – X21) in the cell (2, 1).
This either exhausts the capacity of origin O 2 or satisfies the remaining
demand at destination D1.
If a1 > b1, the requirement at destination D1 is satisfied, but capacity of origin O1 is not
completely exhausted. Move to right in a horizontal position to the second column to
make the second allocation of the magnitude.
X12 = min (a1 – X11, b2) in the cell (1, 2).
This either exhausts the remaining capacity of origin O1 or satisfies the demand at
destination D2.
If b1 = a1, the capacity of origin O1 is completely exhausted as well as the requirement
at destination is completely satisfied, then there is a tie at the second allocation. An
arbitrary tie breaking choice is made. Make the second allocation of the magnitude
X12 = min (a1 – a2, b2) = 0 in the cell (1, 2).
X21 = min ( a2, b1 – X21) = 0 in the cell (2, 1).

 Step 3. Start from the new North West corner of the transportation table satisfying
the destination requirements and exhausting the origin capacities one at a time,
moving down towards the lower right corner of the transportation table until all the
rim requirements are satisfied.
Solved problem 1
Determine initial basic feasible solution to the following transportation problem using the
North West corner rule.
D1 D2 D3 D4 Available
O1 6 4 1 5 14
O2 8 9 2 7 16
O3 4 3 6 2 5
required 6 10 15 4 35

Solution:

 Start with the top most left corner.

 Allocate the maximum possible amount here.
 Move to the right hand cell if there is still any available quantity, otherwise move to
down cell. Again perform step 2 and continue until all the available quantity is
exhausted.
In this method we start from the North West corner cell i.e., (1, 1) cell and allocate there
maximum possible. Thus 6 which is min (a 1 = 14, b1 = 6), can be allocated in this. After
this allocation 8 units are available at O 1, so we move to the next cell of the first row and
allocate 8 units balance at cell (1, 2).
But still second column demand is not satisfied so we allocate more in the second
column. We move to the cell (2, 2) and allocate 2 units where sum of 2 nd column is 10.
Still there are 14 units balance available at O2. We send these 14 units to the next
destination that is cell (2, 3).
But still there one unit balance at destination D3 thus we move to cell (3, 3) and allocate
this one unit there. Now there are 4 units balance at O3. We move to cell (3, 4) and
allocate these 4 units there. The resulting feasible solution is as shown in the following
table.

D1 D2 D3 D4 Available
O1 6 4 1 5 14
(6) (8)
O2 8 9 2 7 16
(2) (14)
O3 4 3 6 2 5
(1) (4)
required 6 10 15 4 35

The total transportation cost = 6x6 + 4x8 + 9x2 + 2x14 + 6x1 + 2x4 = 128
Solved problem 2
Determine initial basic feasible solution to the following transportation problem using the
North West corner rule.
D1 D2 D3 D4 Available
O1 1 2 1 4 30
O2 3 3 2 1 50
O3 4 2 5 9 20
required 20 40 30 10 100

Based on the solution for solved problem 1, we start allocating from cell (1, 1) and move
right and down sequentially. Till the entire rim requirements are met. The resulting
feasible solution is as shown in the following table.

D1 D2 D3 D4 Available
O1 1 2 1 4 30
(20) (10)
O2 3 3 2 1 50
(30) (20)
O3 4 2 5 9 20
(10) (10)
required 20 40 30 10 100

The total transportation cost = 20x1 + 2x10 + 3x30 + 2x20 + 5x10 + 9x10 = 310
Solved problem 3
 Solve the following transportation problem using north west corner rule.

Destination
origin P Q R supply
A 5 7 8 70
B 5 4 6 30
C 6 7 7 50
demands 65 42 43 150

Solution
Since supply = 150 = demands, the problem is balanced. Using the north west corner
rule for the allocation, we get the following allocation

Destination
origin P Q R supply
A 5 (65) 7 (5) 8 70
B 5 4 (30 6 30
C 6 7 (7) 7 (43) 50
demands 65 42 43 150

The total cost of transportation by this method will be

Total cost = 65 x 5 + 5 x 7 + 30 x 4 + 7 x 7 + 43 x 7 = 830
Note: the solution obtained by the north west corner rule is a basic feasible solution. In
this method we do not consider the unit cost of transportation. Hence, the solution
obtained may not be an optimum solution. But this will serve as an initial solution which
can be improved.
Matrix minimum method.

 Step 1. Allocate up to the full capacity/ requirement of the cell having lowest cost. XiJ
= min ( ai, bJ ).

 Step 2. If XiJ = ai , cross off the ith row and decrease bi by ai , go to step 3.
If XiJ = bJ, cross off the jth column of the table and decrease ai by bj. Go to step 3.

 Repeat step 1 and 2 for the resulting reduced table until all the rim requirements are
satisfied. Whenever the minimum is not unique, make an arbitrary choice among the
minimum.
Solved problem 4
Obtain an initial basic feasible solution to the following transportation problem using matrix
minimum method.

D1 D2 D3 D4 CAPACITY
O1 1 2 3 4 6
O2 4 3 2 0 8
 O3 0 2 2 1 10
DEMAN 4 6 8 6 24
D

Solution:
The transportation table of given TP has 12 cells. Following the matrix minimum
method.

D1 D2 D3 D4 CAPACITY
O1 1 2 3 4 6
(6) 0
O2 4 3 2 0 8
(2) (6) 0
O3 0 2 2 1 10
(4) (6) 0
DEMAN 4 6 8 6 24
D 0 0 0 0

Thus, The total transportation cost = 0 x 6 + 0 x 4 + 2 x 6 + 2 x 2 + 2 x 6 = 28

Vogel’s approximation method.
The Vogel’s approximation method (VAM) takes into account not only the least cost C ij, but
also the cost that just exceeds Cij. This method is based on the ‘difference’ associated with
each row and column in the matrix giving unit cost of transportation C ij. This ‘difference’ is
defined as the arithmetic difference between the smallest and next to the smallest cost in that
row or column. This difference in a row or column indicates the minimum unit penalty
incurred in failing to make an allocation to the smallest cost cell. It is a measure of proper
priorities for making allocations to the respective rows and columns. In other words, if we
take a row, we have to allocate to the cell having the least cost and if we fail to do so, extra
cost will be incurred for wrong choice, which is called penalty. The minimum penalty is
given by this procedure.
 Step 1. For each row of the transportation table, identify the least cost and next
smallest cost cells. Determine the difference between them and for each row and
column. Display them alongside the transportation table by enclosing them in
parenthesis against the respective rows and the respective columns.
 Step 2. Identify the row or column with the largest difference among all the rows and
the columns. If a tie occurs, use any arbitrary tie breaking choice. Let the greatest
difference correspond to the ith row and let Cij be the smallest cost in the ith row.
Allocate the maximum feasible amount XiJ = min (ai , bj) in (I, j) cell and cross off the
ith row or jth column in the usual manner.
 Step 3. Recompute the column and row differences for the reduced transportation
table and go to step 2. Repeat till all the rim requirements are satisfied.
Remarks
  A row or column “difference” indicates the minimum unit penalty incurred by failing
to make an allocation to the least cost cell in that row or column.
 It is clear that VAM determines an initial basic feasible solution, which is very close
to the optimum solution, but the number of iterations required to reach the optimum
solution are small.
Area of Application

 It is used to compute transportation routes in such a way as to minimize

transportation cost for finding out location of warehouses.
 It is used to find out locations of transportation corporations depots where
insignificant total cost difference may not matter.
Solved problem 5
A company has three production facilities S1, S2 and S3 with production capacity of 7, 9 and
18 units (in 100s) per week of a product respectively. These units are to be shipped to four
ware houses D1, D2, D3 and D4 with requirement of 5, 8, 7 and 14 units (in 100s) per week.
The transportation costs (in Rupees) per unit between factories to ware houses are given
below. Determine the initial basic feasible solution using Vogel’s approximation method.

D1 D2 D3 D4 Capacity
S1 19 30 50 10 7
S2 70 30 40 60 9
S3 40 8 70 20 18
Demand 5 8 7 14 34

Solution:
The difference (penalty costs) for each row and column has been calculated as depicted in
table given below. The maximum penalty, 22 occurs in column D 2. Thus and cell (S3, D2) is
having the least cost in that column. Thus it is selected for allocation. The maximum possible
allocation in this cell can be 8

D1 D2 D3 D4 Capacity Row difference

S1 19 30 50 10 7 9
S2 70 30 40 60 9 10
S3 40 8 (8) 70 20 18 10 12
Demand 5 80 7 14 34
Column 21 22 10 10
defferenc
e
The new row and column penalties are calculated except column D 2 because its demand
has been satisfied. The second round allocation is made in column D 1 with target
penalty 21 in the same way as in the first round as depicted in the cell (S 1, D1) in the
table given below.

D1 D2 D3 D4 Capacity Row difference

S1 19 (5) 30 50 10 72 9
S2 70 30 40 60 9 20
S3 40 8 (8) 70 20 18 10 20
Demand 5 (0) 80 7 14 34
Column 21 X 10 10
defferenc
e

In the third round, the maximum penalty 50 occurs at row S 3. The maximum possible
allocation of 10 units is made in cell (S 3, D4) having least transportation cost 20 as
depicted in table given below.

D1 D2 D3 D4 Capacity Row difference

S1 19 (5) 30 50 10 72 40
S2 70 30 40 60 9 20
S3 40 8 (8) 70 20 (10) 18 10 0 50
Demand 5 (0) 80 7 14 4 34
Column X X 10 10
defferenc
e

In the fourth round, the maximum penalty occurs in column D 4. The maximum possible
allocation of 2 units is made in cell (S 1, D4). The process is continued with new
allocations till all rim requirements are fulfilled. The initial solution using vogel’s
approximation method is depicted in table given below.

D1 D2 D3 D4 Capacity Row difference

S1 19 (5) 30 50 10 (2) 720 40
S2 70 30 40 (7) 60 (2) 90 20
 S3 40 8 (8) 70 20 (10) 18 10 0 X
Demand 5 (0) 80 70 14 4 2 34
0
Column X X 10 50
defferenc
e

The total transportation cost associated with this method is calculated as follows;
Total cost = 5 x 19 + 2 x 10 + 7 x 40 + 2 x 60 + 8 x 8 + 10 x 20 = Rs. 779
Moving towards optimality and optimality test
The rationale behind the optimality
Given an initial non-negative basic feasible solution, a question may arise as to how we can
find a successively better basic feasible solution. This means that we have to select an
entering variable, a leaving basic variable and identify the corresponding solution. In the
transportation model, selecting an entering variable means selecting a new cell, in which to
make a non-zero allocation in the transportation matrix.
The principle of selecting the variable which improves the value of the objective function
(minimizes the total transportation cost), at the fastest rate is still valid. So we select an
unoccupied cell and make a unit allotment, so that the total cost of transportation will
decrease by greatest amount.
Optimality test
For conducting the optimality test to any feasible solution of a (mix) transportation problem,
the following two conditions must be satisfied.
1. It consists of exactly (m + n – 1) individual cells being allocated.
2. These allocations are in independent positions.
The first condition is normally satisfied in many problems. If the first condition is not
satisfied then it results in a state known as degeneracy, in the transportation model. The
discussion of degeneracy and its resolution is explained in a later section.
A set of allocations comprising a feasible solution is said to be in independent positions if
it is impossible to increase or decrease any or violating the row or column restrictions. In
other words, a simple criterion for independence is that, it is impossible to travel from any
allocation back to itself, by a series of alternating horizontal and vertical jumps from one
occupied cell to another, without a direct reversal of route.
Definition. A loop is the sequence of cells in the table such that:

 Each pair of consecutive cells lies either in the same row or column.

 No three consecutive cells lie in the same row or column.

  The first and the last cells of the sequence lie in the same row or column.

 No cell appears more than once in the sequence.

Consider the non-basic variable corresponding to the most negative of the quantities C iJ –
Ui - VJ. Calculated in the test for optimality. It is made to the incoming variables. Construct a
loop consisting exclusively of this incoming variable (cell) and current basic variables (cells).
Then allocate to the incoming cell as many units as possible such that, after appropriate
adjustments have been made to the other cells in the loop, the supply and demand constraints
are not violated. All allocations remain non-negative and one of the old basic variables is
reduced to 0. (where upon it ceases to be basic)
Modified distribution method / Modi method / u – v method.
 Step 1. Under this method we construct penalties for rows and column by subtracting
the least value of row / column from the next least value.
 Step 2. Select the highest penalty constructed for both row and column. Enter that
row / column and select the minimum cost and allocate min (ai , bJ).
 Step 3. Delete the row or column or both if the rim availability / requirements is met.
 Step 4. Repeat step 1 to 3 till all allocations are over.
 Step 5. Depict all rows as U1, U2…..Ui and all columns as Vj
 Step 6. For allocating cell, form the equation U i + VJ = Cij . Set one of the dual
variable Ui / Vj to 0 and solve for others.
 Step 7. Use these values to find ∆Ij = CiJ – Ui – VJ of all ∆Ij ≥ 0, then it is the optimal
solution.
 Step 8. If any ∆ij ≤ 0, select the most negative cell and form loop. Starting point of the
loop is positive and alternative corner of the loop are negative and positive. Examine
the quantities allocated at –ve places. Select the minimum. Add it at + ve places and
subtract from –ve places.
 Step 9. Form new table and repeat steps 5 to 8 till ∆ij ≥ 0.
Solved problem 6
Solve the following transportation problem with cost coefficients, demands and supplies as
given in the following table.

D1 D2 D3 Supply
A 6 8 4 14
B 4 9 8 12
C 1 2 6 5
Demand 6 10 15 31

Solution
Since total demand = 31 = total supply, the problem is balanced. The initial basic
feasible solution is obtained by Vogel’s approximation method. Table below depicts the
initial solution.

rows D1 D2 D3 Supply
U1 A 6 8 4 (14) 14
U2 B 4 (6) 9 (5) 8 (1) 12
U3 C 1 2 (5) 6 5
Demand 6 10 15 31
columns V1 V2 V3

Now moving towards optimality

For allocated cells
U1 + V3 = 4
U2 + V1 = 4
U2 + V2 = 9
U2 + V3 = 8
U3 + V2 = 2
Set U2 as 0 and solve for others. We get
U1 = -4 V1 = 4
U2 = 0 V2 = 9
U3 = -7 V3 = 8
Compute ∆Ij = CiJ – Ui – VJ for unallocated cells
∆11 = 6 – (-4) – 4 = 6
∆12 = 8 – (-4) – 9 = 3
∆31 = 1 – (-7) – 4 = 4
∆33 = 6 – (-7) – 8 = 5
Since all ∆Ij for unallocated cells are positive, the solution is optimum solution.
The optimum allocations are; X13 = 14; X21 = 6; X22 = 5; X23 = 1 and X32 = 5
The minimum transportation cost is
Total cost = 14 x 4 + 6 x 4 + 5 x 9 + 1 x 8 + 5 x 2 = 143
Unbalanced Transportation problem
When the supply and demands are not equal, it is called as unbalanced transportation problem
(Σai ≠ Σbj). we introduce either a dummy row or a dummy column with zero costs for each
cell.s
Solved problem 7
A car company is faced with an allocation problem resulting from a rental agreement that
allows cars to be returned to locations other than those from where they were originally
rented. At the present time there are two locations, with 15 and 13 cars respectively and 4
locations requiring 9, 6, 7 and 9 cars respectively. The unit transportation cost (in Dollars )
between the locations is as given below:

D1 D2 D3 D4
S1 45 17 21 30
S2 14 18 19 31

Obtain a minimum cost schedule.

Terminal question 1:
Solve the following transportation problem.

A B C D Availability
I 21 16 25 13 11
II 17 18 14 23 13
III 2 27 18 41 19
Demand 6 10 12 15 43

Terminal question 2:
A company has three cement factories located in cities I, II, III which supply cement to four
projects located in towns A, B, C, D. Each plant can supply daily 6, 1, 10 truckloads of
cement respectively and the daily cement requirements of the projects are respectively 7, 5, 3,
2 truck loads. The transportation cost per truck load of cement (in hundreds of rupees) from
each plant to each project site is as follows:

A B C D
I 2 3 11 7
II 1 0 6 1
III 5 8 15 9

Terminal questions 3
Solve the following transportation problem.
 A B C D E F Availability
I 9 12 9 6 9 10 5
II 7 3 7 7 5 5 6
III 6 5 9 11 3 11 2
IV 9 8 11 2 2 10 9
Demand 4 4 6 2 4 2 22

Terminal question 4
Determine initial basic feasible solution to the following transportation problem using the
north west corner rule.

D1 D2 D3 D4 Availability
O1 6 4 1 5 14
O2 8 9 2 7 16
O3 4 3 6 2 5
Requirement 6 10 15 4 35

Terminal question 5
Obtain an initial basic feasible solution to the following transportation problem using
Vogel’s approximation method.

D1 D2 D3 D4
A 5 1 3 3 34
B 3 3 5 4 15
C 6 4 4 3 12
D 4 -1 4 2 19
21 25 17 17 80

CHAPTER – 6
ASSIGNMENT PROBLEM
The objective is to minimize the cost or time of completing a number of jobs by a number of
persons and maximize efficiently revenue, sales etc.
There are n jobs for a factory and factory has n machines to process the jobs. A job i = 1, 2,
---- n, when processed by machine j = 1, 2, --- n is assumed to incur a cost C ij. The
assignment is to be made in such a way that each job can associate with one and only one
machine. Determine an assignment of jobs to machines so as to minimize the overall cost.
Mathematical formulation of the problem
Let XIJ be a variable defined by
XiJ = 0 if the ith job is not assigned to the jth machine.
= 1 if the ith job is assigned to the jth machine.
then, since only one job is to be assigned to each machine, we have
∑i=1n XiJ = 1 AND ∑j=1n XiJ = 1
Also the total assignment cost is given by
Z = ∑i=1n ∑j=1n CiJ XiJ
Thus the assignment problem takes the following mathematical form
Determine XiJ ≥ 0 (i, j = 1, 2, ----n)
So as to
Minimize Z = ∑i=1n ∑j=1n CiJ XiJ
Subject to the constraints
∑i=1n XiJ = 1 j = 1, 2, --- n AND ∑j=1n XiJ = 1 i = 1, 2, ---- n
With all XiJ = 0 OR 1
Hungarian method: Algorithm

 Step 1: prepare row ruled matrix by selecting the minimum values for each row and
subtract it from other element of the row.

 Step 2: prepare column reduced matrix by subtracting minimum value of the column
from the other values of that column.

 Step 3. First row wise assign a zero. If there is only one zero in the row and cross (x)
other zeros in that column.
 Step 4. Now assign column wise if there is only one zero in that column and cross
other zeros in that row.
Repeat step 3 and 4 till all zeros are either assigned or crossed. If the number of
assignments made is equal to number of rows present, then it is the optimal solution
otherwise proceed as follows

 Step 5: Tick mark the row which is not assigned. Look for crossed zero in that row.
Mark the column containing the crossed zero. Look for assigned zero in that column.
Mark the row containing assigned zero. Repeat this process till all markings are over.
  Step 6: Draw straight line through unmarked rows and marked column. The number
of straight lines drawn will be equal to number of assignments made.

 Step 7. Examine the uncovered element. Select the minimum.

o Subtract it from uncovered elements.

o Add it at the point of intersection of lines.

o Leave the rest as it is.

o Prepare a new table.

 Step 8 Repeat step 3 to 7 till optimum assignment is obtained.

 Step 9. Repeat step 5 to 7 till number of allocations = number of rows.

Solved problem 1.

Find the optimum assignment so as to minimize the cost.

1 2 3 4 5
PERSONS
A 8 4 2 6 1
B 0 9 5 5 4
C 3 8 9 2 6
D 4 3 1 0 3
E 9 5 8 9 5

Preparing the row ruled matrix

1 2 3 4 5
PERSONS
A 7 3 1 5 0
B 0 9 5 5 4
C 1 6 7 0 4
D 4 3 1 0 3
E 4 0 3 4 0
Now preparing the column ruled matrix

1 2 3 4 5
PERSONS
A 7 3 0 5 0
B 0 9 4 5 4
C 1 6 6 0 4
 D 4 3 0 0 3
E 4 0 2 4 0

Now, assigning row wise and column wise

1 2 3 4 5
PERSONS
A 7 3 0 5 0
B 0 9 4 5 4
C 1 6 6 0 4
D 4 3 0 0 3
E 4 0 2 4 0

Since all 5 jobs are assigned to five persons, the solution to the problem is:
Job 1 to person B 0
Job 2 to person E 5
Job 3 to person D 1
Job 4 to person C 2
Job 5 to person A 1
Total cost = 9

Unbalanced A.P
If the numbers of rows are not equal to number of columns then it is an unbalanced AP. We
introduce either a dummy row or column and proceed as usual.
Solved problem 2

Operations M1 M2 M3 M4
O1 10 15 12 11
O2 9 10 9 12
O3 15 16 16 17

Solution:
Here number of operations are less than number of machines. Thus, introducing
operation 4 (O4) as fourth row with costs as 0 for each machine.

Operations M1 M2 M3 M4
 O1 10 15 12 11
O2 9 10 9 12
O3 15 16 16 17
O4 0 0 0 0

Now preparing row reduced matrix

Operations M1 M2 M3 M4
O1 0 5 2 1
O2 0 1 0 3
O3 0 1 1 2
O4 0 0 0 0

Since all column has got 0, thus, there is no change in column reduced matrix
Assigning row wise

Operations M1 M2 M3 M4
O1 0 5 2 1
O2 0 1 0 3
O3 0 1 1 2
O4 0 0 0 0

Assigning column wise

Operations M1 M2 M3 M4
O1 0 5 2 1
O2 0 1 0 3
O3 0 1 1 2
O4 0 0 0 0

Since all operations are assigned, the solution to the problem is

O1 to M1 10
O2 to M3 9
O3 to M4 17
O4 to M2 0
The total cost = 36

Infeasible assignments
Certain jobs cannot be assigned to certain machines or operators.
For example:
Computer operator cannot be assigned to repair an automobile.
A drilling machine cannot cut gears
A welder cannot be assigned the job to manufacture printed circuit boards (PCBs).
For such jobs at the assignment point, We insert a high value α at these cells and apply
Hungarian method.
Solved problem 3

Territory P1 P2 P3 P4
T1 20 - 32 27
T2 15 20 17 18
T3 16 18 - 20
T4 - 20 18 24

Solution:
Assigning very high cost for impossible assignments (α)

Territory P1 P2 P3 P4
T1 20 α 32 27
T2 15 20 17 18
T3 16 18 α 20
T4 α 20 18 24

Now preparing row reduced matrix

Territory P1 P2 P3 P4
T1 0 α 12 7
T2 0 5 2 3
T3 0 2 α 4
T4 α 2 0 6

Now preparing the column reduced matrix

Territory P1 P2 P3 P4
T1 0 α 12 4
T2 0 3 2 0
T3 0 0 α 1
T4 α 0 0 3

Now assigning row wise (yellow shade depicts the assignment)

Territory P1 P2 P3 P4
T1 0 α 12 4
 T2 0 3 2 0
T3 0 0 α 1
T4 α 0 0 3

Since all territories are already assigned, no need of assigning column wise.
Thus, the solution to the problem is
T1 to P1 20
T2 to P4 18
T3 to P2 18
T4 to P3 18
The minimum cost of assignment is = 74
Maximization in A.P
Hungarian method cannot be applied to maximization problems. Therefore we reduce it to
minimization problem by subtracting every value from the highest value of the matrix and
then proceed as usual.
Solved problem 4.

Classes P1 P2 P3 P4
C1 70 75 80 65
C2 80 70 65 75
C3 75 70 85 80
C4 60 65 70 90

Solution:
Converting to minimization problem by subtracting each cell value from the highest
value (90)

Classes P1 P2 P3 P4
C1 20 15 10 25
C2 10 20 25 15
C3 15 20 5 10
C4 30 25 20 0

Preparing row reduced matrix

Classes P1 P2 P3 P4
C1 10 5 0 15
C2 0 10 15 5
C3 10 15 0 5
C4 30 25 20 0
Preparing the column reduced matrix

Classes P1 P2 P3 P4
C1 10 0 0 15
C2 0 5 15 5
C3 10 10 0 5
C4 30 20 20 0

Assigning the classes row wise (box with yellow shade depicts assignment).

Classes P1 P2 P3 P4
C1 10 0 0 15
C2 0 5 15 5
C3 10 10 0 5
C4 30 20 20 0

Assigning column wise (box with yellow shade depicts assignment).

Classes P1 P2 P3 P4
C1 10 0 0 15
C2 0 5 15 5
C3 10 10 0 5
C4 30 20 20 0

Since all classes are assigned, the solution to the problem is:
C1 to P2 75
C2 to P1 75
C3 to P3 85
C4 to P4 90
The maximum value = 325
Solved problem 5.

Five different machines can do any of the required five jobs with different profits resulting
from each assignment as given below:

MACHINES
A B C D E
1 30 27 40 28 40
2 40 24 27 21 36
JOBS 3 40 32 33 30 35
4 25 38 40 36 36
5 29 62 41 34 39
Solution:
Since it is a profit problem, it has to be maximized. Thus, it is a maximization problem.
Converting it to minimization problem by subtracting each cell value from the highest value.

A B C D E
1 32 35 22 34 40
2 22 38 35 41 26
3 22 30 29 32 27
4 37 24 22 26 26
5 33 0 21 28 23

Preparing row reduced matrix

A B C D E
1 10 13 0 12 18
2 0 16 13 19 4
3 0 8 7 10 5
4 15 2 0 4 4
5 33 0 21 28 23

Now preparing column reduced matrix

A B C D E
1 10 13 0 8 14
2 0 16 13 15 0
3 0 8 7 6 1
4 15 2 0 0 0
5 33 0 21 24 19

Now assigning row wise (yellow shade depicts the assignment)

A B C D E
1 10 13 0 8 14
2 0 16 13 15 0
3 0 8 7 6 1
4 15 2 0 0 0
5 33 0 21 24 19

Now assigning column wise

A B C D E
1 10 13 0 8 14
2 0 16 13 15 0
3 0 8 7 6 1
4 15 2 0 0 0
 5 33 0 21 24 19

Since, number of assignment are only 4 and not 5, we have to proceed further

A B C D E
1 10 13 0 8 14
2 0 16 13 15 0
3 0 8 7 6 1
4 15 2 0 0 0
5 33 0 21 24 19

Now marking the lines through unmarked rows and marked column

A B C D E
1 10 13 0 8 14
2 0 16 13 15 0
3 0 8 7 6 1
4 15 2 0 0 0
5 33 0 21 24 19

On examining the uncovered elements it is seen that minimum is 1 (A35).

Completing step six of hingarian method and preparing the new table

A B C D E
1 11 13 0 8 14
2 0 15 12 14 0
3 0 7 6 5 0
4 16 2 0 0 0
5 34 0 21 24 19

Assigning row wise

A B C D E
1 11 13 0 8 14
2 0 15 12 14 0
3 0 7 6 5 0
4 16 2 0 0 0
5 34 0 21 24 19

all jobs are assigned, the solution to the problem is

Job 1 to machine C 40
Job 2 to machine E 36
Job 3 to machine A 40
 Job 4 to machine D 36
Job 5 to machine B 62
The maximum profit is = 214
Terminal question 1

Four jobs are to be done on four different machines. The cost in rupees of producing ith on the
jth machine is given below. Assign the jobs to different machines so as to minimise the total
cost.

Jobs M1 M2 M3 M4
J1 15 11 13 15
J2 17 12 12 13
J3 14 15 10 14
J4 16 13 11 17

Terminal question 2

A marketing manager has 5 salesmen and 5 sales districts. Considering the capabilities of
the salesmen and the nature of districts, the marketing manager estimates that sales per month
(in hundred rupees) for each salesman in each district would be as follows. Find the
assignment of salesman to districts that will result in maximum sales.

A B C D E
1 32 38 40 28 40
2 40 24 28 21 36
3 41 27 33 30 37
4 22 38 41 36 36
5 29 33 40 35 39

Terminal question 3

In a plant layout there are 5 vacant places. The plant orders four machines to be installed in
these places. The cost of installing is as follows. Find the optimum assignment

A B C D E
M1 9 11 15 10 11
M2 12 9 - 10 9
M3 - 11 14 11 7
M4 14 8 12 7 8

Terminal question 4

Find the assignment that maximizes the total sale.

1 2 3 4
 M1 42 35 28 21
M2 30 25 20 15
M3 30 25 20 15
M4 24 20 16 12

CHAPTER – 7
INTEGER PROGRAMMING PROBLEM

The integer programming problem is a special case of the linear programming problem where
all or some variables are constrained to assume non-negative integer values.
Binary, all and mixed integer programming problem
Determine the value of unknown Xi (i = 1,2,3, ---- n) so as to optimize
Z = ∑i=1n Ci Xi
Subject to the constraints
Z = ∑i=1n ∑j=1n aij Xij = bi
And XJ ≥ 0 (J = 1, 2, ----n)
Where XJ being an integer value for j = 1,2,-----n

 Pure. If all the variables are constrained to only integer value, it is called an “all
integer programming problem.
 Mixed. If only some of the variables are restricted to take integer value, then it is
“mixed integer programming problem”
 Zero – one. If Xj is 0 or 1, it is called as binary integer programming problem.
Gomory’s method
Invented by Ralph E. Gomory.

 Get solution by simplex method.

 If all variables have integer value, then it is optimal solution.

 Else increase new constraint called Gomory’s constraint.

 Solve problem by dual simplex method

 Go to step 2
Algorithm
  Step 1: Obtain the optimum solution of the given problem by using Simplex method.

 Step 2 : Test the integrality of the solution. If all integer variables, then it is an
optimal solution. Else go further.

 Step 3. Examine the constraints equation. Choose the largest fraction from b i.
Assume it as fbi = max (fb1, fb2, ----- fbn)

 Step 4. Express each of the negative fractions if any, in the kth row as the sum of a
negative integer and non-negative fraction.

 Step 5. Find the Gomory’s constraint.

∑j=0n’ fkj XJ ≥ fk0

 Add the equation GI = -fKO + ∑J = 0n’ fkj XJ to the current set of constraints equations.

 Step 6. Find new optimum solution by simplex method.

 Step 7. If this new optimum solution is an integer solution, then it is an optimum

solution. Otherwise, return to step 3 and repeat the process until an optimum feasible
integer solution is obtained
Solved problem 1
Find the optimum integer solution to the following all IPP
Maximize Z = X1 + X2
Subject to:
3X1 + 2X2 ≤ 5
X2 ≤ 2

X1, X2 ≥ 0 and are integers.

Solution:

Branch and bound technique

Algorithm
Step 0: If the master list is not empty, choose an l.p.p. out of it. Otherwise stop the
process. Go to step 1.
Step 1 : Obtain the optimum solution to the chosen problem. If either

 It has no feasible solution. Or

 The resulting optimum value of the objective function z is less than or equal to Z (T).
Then let Z(T+1) = ZT and return to step 0, otherwise go to step 2.
Step 2. If the so obtained solution satisfies the integer constraints then record it. Let
Z(T+1) be associated value of Z, return to step 0. Otherwise move to step 3.
Step 3. Select any variable XJ J = 1, 2, -------P, that does not have an integer value in the
obtained solution to the l.p.p. chosen in step 0. Let X J* denote this optimum value of XJ,
add two LPP’s to the master list. These LPP’s are identical with the LPP chosen in step 0,
except that in one, the lower bound on XJ is replaced by [XJ*] + 1. Let Z(T+1) = ZT, return to
step 0.
Solved problem 2
Find the optimum integer solution to the following all i.p.p using the branch and bound
technique
Maximize Z = 7X1 + 9X2
Subject to the constraints
-X1 + 3X2 ≤ 6
7X1 + X2 ≤ 35
0 ≤ X1, X2 ≤ 7
X1, X2 are integers
Solved problem 3
Find the optimum integer solution to the following all i.p.p
Maximize Z = X + 2Y
Subject to:
X+Y≤7
2X ≤ 11
2Y ≤ 7
X,Y ≥ 0 and are integers.
Terminal question 1
Use branch and bound technique to solve the following problem
Maximize Z = 3X1 + 3X2 + 13X3
Subject to the constraints
-3X1 + 6X2 + 7X3 ≤ 8
6X1 - 3X2 + 7X3 ≤ 35
 0 ≤ X1, X2 , X3 ≤ 5
X1, X2 and X3 are integers

CHAPTER – 8
INFINITE QUEUING MODEL

Queuing theory based on probability concepts gives an indication of the capability of

a given system and of the possible changes in its performance with modification to the
system.
Formation of queue occurs whenever the current demand for a service exceeds the
current capacity to provide that service.
Examples of queues

 Queue for the bus

 At the office premises for the lift.

 Cars waiting at the patrol pumps for the petrol.

 Customers waiting in the bank

 Purchase of ticket at cinema hall.

 Mobile subscribers waiting for SIM card.

 Aircraft queuing for free run way.

The common feature.
Customers arrive at the service centre and wait for the service.
Reasons for the queue.

 Increased arrival rate of customers

 Insufficient service stations.

 Multiple service channels- series, parallel

 Enhanced service time per unit customer.

 Server not available for complete time.

- If
  Arrival rate of customer > service time per unit of customer. – long queue
 Arrival rate of customer ≤ service time per unit of customer. – no queue or
controllable queue
Analysis of queuing process

 Step 1 preliminary study.

 Step 2 exploration of the various alternatives – by introducing the changes in the

constituents of the queuing system, it should be possible to effect improvement.
o Alter the arrival pattern

 Withdraw the facility to certain catagories of customers.

 Introducing appointment system.
 Improve time taken for sevice
 Increase no. Of service stations.
o Provide additional service facilities to relieve congestion.

o Priority may be given to arrivals involving high cost of waiting time.

o Separate channel of service for different types of customers.

 Step 3 – collection of data and analysis.

 Step4 - evaluation of alternatives.

 Step 5 - implementation.
Constituents of queuing system

 Arrival pattern. Average rate at which customers arrive as well as statistical

pattern of customers.
o Regular pattern

o Completely random arrivals.

 Service facilities. When service is available, number of customers that can be

served at a time and the statistical pattern of time taken for services.

 Queue discipline.
o First come first serve

o Last come first out.

Properties of the queuing system

Arrivals.
 Regular arrival. Appointment cases. Not very common and difficult to handle
mathematically

 Random arrival. They follow the poisson distribution

Service facility

 Availability of service.

 Number of service centers

 Single service channel

 Multi service channel

 Service channels in series.

Duration of service.
Length of time taken to serve a customer.

 Constant service time

 Random service time

o The server does not distinguish between the various arrivals.

o The server does not deliberately change the duration of service on the basis of
the time taken to serve the previous arrival.
o The server forgets the time for which he has been serving a customer.

 Service time follows exponential distribution.

Queue discipline

 First come first served.

 Last come first out

Dynamic queuing discipline

 Service in random order

 Priority service

 Pre – emptive priority

Customer Behaviour.
 Balking: arriving customers are said to “balk” if they do not join a queue because of
their reluctance to wait.
 Collusion: customers may be in collusion in the sense that only one person would
join the queue but would demand service on behalf of several customers.
 Reneging: impatient customers who would not wait beyond a certain time and leave
the queue are said to renege.
 Jockeying: some customers keep on switching over from one queue to another in a
multiple service centre. This is called jockeying.
Terminal question 1
In a particular single-server system, the arrival rate λ = 5 per hour and service rate µ = 8
per hour. Assume the conditions for use of the single channel queuing model, find out:
(a) The probability that the server is idle.
(b) The probability that there are at least two customers in the system.
(c) Expected time that a customer is in the queue.
Terminal question 2
Customers arrive at the first class ticket counter of a theater at the rate of 12 per hour.
There is one clerk serving the customer at a rate of 30 per hour. Assuming the condition
of the single channel queuing model, evaluate:
(a) The probability that an arriving customer has to wait for the service.
(b) The expected number of customers in the system.
(c) The average waiting time of the customer in the system.
Terminal question 3
In a bank every 15 minutes one customer arrives for cashing the chaque. The clerk takes
10 minutes to service. Assuming the usual conditions, find:
(a) The ideal time of clerk in 8 working hours.
(b) Expected number of customers in the queue.
(c) The probability that a customer has to wait for 15 minutes or more to receive the
service.
Application of queuing theory

 Traffic control

 Determining the sequence of computer operations

 Predicting computer performance

 Health services (e.g., control of hospital bed assignments)

 Airport traffic, airline ticket sales

 Layout of manufacturing systems

 Telecommunications
 Mathematical Analysis of Queuing Theory
Introduction
Queuing theory is the collection of mathematical models of various queuing systems that
take as inputs parameters of the above elements and that provide quantitative parameters
describing the system performance
Statistical Equilibrium. The state in which the probability distribution remains same for
arrivals is called the steady state and the system is said to have acquired a state of
statistical equilibrium
In the steady state, there will be variations in the queue from time to time but the
probability distributions representing the queuing process will remain the same and are
independent of the time at which the system is examined.
Properties of the queuing system
Arrivals.

 Customers are discrete entities.

 Population - finite/ infinite.

 No simultaneous arrivals.

 Pattern of arrivals in a time period t0 follows Poisson distribution with average arrival
rate λ
Prob [number of arrivals =X] = [e-λ t0(λ t0)X]/X!
X = 1, 2, 3, -------
Service

 Single serve channel/multiple channel.

 Single queue/ infinite capacity

Pattern discipline

 First come first served.

 When number of arrivals follow Poisson distribution, the inter arrival time, i.e., the
time between arrivals follow exponential distribution (continuous).
The queuing systems with which we are concerned are denoted by m/m/1 and m/m/c
where m stands for exponential inter - arrival and exponential service time distribution,
the third figure indicates the number of channels available (1 or c).
Service system
Configuration of the service system
 Single server - single queue

 Single server - several queues

 Several (parallel) servers - single queue

 Several servers – several queues

 Speed of service
o The service rate describes the number of customers serviced during a
particular time period
o The service time indicates the amount of time needed to service a customer

o Service rates and times are reciprocal of each other and either of them is
sufficient to indicate the capacity of the facility.

MULTIPLE SERVERS IN A SERIES

Single channel models
Poisson arrivals, exponential services, single channel, infinite population

 PN = ρn (1 – ρ) ρ = λ/µ

 P0 = (1 – ρ)

 E (m) = λ2/µ(µ - λ)

 E (m/ m>0) = µ/(µ - λ)

 E (n) = λ/(µ - λ)

 E (w) = λ/µ(µ - λ)

 E (w/w>0) = 1/(µ - λ)

 P (W = 0) = 1- ρ

 E (v) = 1/(µ - λ)

 F(w) = λ (1-ρ)e – w (µ-λ)

 F(v) = (µ-λ)e - v (µ-λ)

Poisson arrivals, exponential services, single channel, number of customers limited
to N
 PN = P0 ρn ρ = λ/µ

 P0 = (1 – ρ)/1 – ρn+1
n = 1, 2, 3, ----N

 P (W=0) = P0 = (1 – ρ)/1 – ρn+1

Solved problem 1
Patrons arrive at a small post office at the rate of 30 per hour. Service by the clerk on duty
takes an average of 1 minute per customer.
(a) calculate the mean customer time
(i) Spent waiting in line
(ii) Spent receiving or waiting for service
(b) Find the mean number of persons
(ii) In line
(iii) Receiving or waiting for services
Solved problem 2
The tool company’s quality control department is manned by a single clerk, who takes an
average of 5 minutes in checking parts of each of the machines coming for inspection.
The machine arrive once in every 8 minutes on the average. One hour of machine is
valued at Rupees 15 and clerk’s time is valued at Rupees 4 per hour. What is the average
hourly queuing system costs associated with the quality control department?
Solved problem 3
A television repairman finds that the time spent on his job has an exponential distribution
with a mean of 30 minutes. If he repairs the sets in the order they come in, and if the
arrival of set follows the Poisson distribution with an average rate of 10 per 8 hour day,
what is the repairman’s expected idle time each day? How many jobs are ahead of the
average set just brought in.
Solved problem 4
A hospital emergency room can accommodate at most m = 5 patients. The patients arrive
at a rate of 4 per hour. The single staff physician can only treat only 5 patients per hour.
Any patient overflow is directed to another hospital.
(a) Determine the probability distribution for the number of patients either waiting for or
receiving treatment at any given time.
(b) Determine the mean values for the number of patients in the emergency room, and the
number of patients to see the doctor.
Solved problem 5
Cars arrive at a toll gate on a frequency according to Poisson distribution with mean 90
per hour. Average time for passing through the gate is 38 seconds. Drivers complain of
long waiting time. Authorities are willing to decrease the passing time through the gate to
30 seconds by introducing new automatic devices. This can be justified only if under the
old system, the number of waiting cars exceeds 5. In addition the percentage of gate’s idle
time under the new system should not exceed 10 seconds. Can the new device be
justified?
Solved problem 6
Customers arrive at one-window drive-in bank according to a Poisson distribution with
mean 10 per hour. Service time per customer is exponential with mean 5 minutes. The
space in front of the window, including that for the serviced car can accommodate a
maximum of 3 cars. Other cars can wait outside this space.
(a) What is the probability that an arriving customer can drive directly to the space in
front of the window?
(b) what is the probability that an arriving customer will have to wait outside the
indicated space?
(c) how long is an arriving customer expected to wait before starting services?
(d) How many spaces should be provided in front of the window so that all the arriving
customers can wait in front of the window at least 20% of the time?
Multiple service channel models
Poisson arrivals, exponential services, multi channel, infinite population

 PN = P0 ρn/n! for n<c ρ = λ/µ

 PN = P0 ρn/c!(cn-c) for n> or =c

 P0 = 1/[∑n=0 c-1 {ρn/n!} + ρc/c!(1-ρ/c)]

 E (m) = [{λµ (λ/µ)c}/(c-1)! (cµ - λ)2]p0

 E (n) = [{λµ (λ/µ)c}/(c-1)! (cµ - λ)2]p0 + λ/µ

 OR E(n) = E(m) + λ/µ

 E (w) = E(m)/λ

 E (v) = E(m)/λ + 1/µ = E(w) + 1/µ

Solved problem 7
Ships arrive at a port at a rate of one in every 3 hours, with a negative exponential
distribution of interval time. The time a ship occupies a berth for unloading and loading
has a negative exponential distribution with an average of 12 hours. If the average delay
of ships waiting for berth is to be kept below 6 hours, how many berths should there be at
the port?
Solved problem 8
A tool crib is operated by m servers, and demands for service arrive randomly at mean
rate of 1.4 per minute. The mean service time per server is 1.25 minutes, and times are
exponentially distributed. Finite population effects may be ignored. If the average hourly
pay rate of the tool crib operator is rs. 2 per hour and the average hourly pay rate of
production employee is rs. 4 per hour, determine the optimum value of m. (you may
assume an 8 hour working day).
Solved problem 9
A bank has two counters for withdrawals. Counter one handles withdrawals of value less
than Rupees 300 and counter two Rupees 300 and above. Analysis of service time shows
a negative exponential distribution with mean service time of 6 minutes per customer for
both the counters. Arrival of follows poisson distribution with mean 8 per hour for
counter one and 5 per hour for counter two.
(a) What are the average waiting times per customer of each counter?
(b) If each counter could handle all withdrawals irrespective of their value, how would
the average waiting time change?
Erlang family of distribution of service time
Poisson arrivals, service with k phases, each phase having a negative exponential
distribution

 E (m) = (K + 1)λ2/ 2K.µ(µ - λ)

 E(n) = E(m) + λ/µ

 E(w) = E(m)/ λ

 E(v) = E(m)/ λ + 1/µ = E(w) + 1/µ

For constant service time, equating k= ∞

 E (m) = λ2/ 2µ(µ - λ)

 E(n) = E(m) + λ/µ

 E(w) = E(m)/ λ

 E(v) = E(m)/ λ + 1/µ = E(w) + 1/µ

Solved problem 10
In a cafeteria at a bus depot the customers will have to pass through three arrive at one-
window counters. At the first counter the customers buy coupons at the second they select
and collect the snacks to be taken and at the third they collect coffee or tea as required.
The server at each counter takes on an average two minutes although the distribution of
the time of service is approximately exponential. If arrivals of the customers to the
cafeteria are approximately poisson at an average of six per hour, what is the average time
spent by a customer waiting in the cafeteria? What is the average time of getting the
service?
Solved problem 11
Arrivals at telephone booth are considered to be poisson with an average time of 10
minutes between one arrival and the next. The length of the phone call is assumed to be
distributed exponentially, with mean 3 minutes.

 What is the probability that a person arriving at the booth will have to wait?

 The telephone department will install a second booth when convinced that an arrival
would expect waiting for at least 3 minutes for a phone call. By how much the flow of
arrivals increase in order to justify a second booth?

 What is the average length of the queue that forms time to time?

 What is the probability that it will take him more than 10 minutes altogether to wait
for the phone and complete his call?

 What is the most probable time spent in getting the service?

Limitations of mathematical approach
Classic queuing is too mathematically restrictive to be able to model all real world
situations.
Terminal question 1
A stenographer has 5 persons for whom she performs stenographic work. Arrival rate is
poisson and service times are exponential. Average arrival rate is 4 per hour with an
average service time of 10 minutes. Cost of waiting is Rupees 8 per hour while the cost of
servicing is Rupees 2.50 each. Calculate:

 The average waiting time of an arrival

 The average length of the waiting line

 The average time which an arrival spends in the system

 The minimum cost service rate.

Terminal question 2
A ticket issuing office is being manned by a single server. Customers arrive to purchase
tickets according to a poisson process with a mean rate of 30 per hour. The time required
to serve a customer has an exponential distribution with a mean of 90 seconds. Find the
value of pn, ls and ws denote the expected line length and waiting time in the system
respectively.

CHAPTER – 9
FINITE QUEUING MODEL
Unlike the case of infinite population, there will be cases, where the possible number of
arrivals will be limited and will be very small. Still there will be waiting line and queue for
requirement of service.
Examples

 Machines requiring service from repair crews.

 In an aircraft number of seats are limited, still there will be queue and waiting for
services of air hostages.

 Aircraft queuing for free run way.

Queue discipline
 FIRST COME – FIRST SERVED.
 PRIORITY e.g., Machines of high cost may be given priority for maintenance while
others may be kept waiting, even if they have broken down before.
 RANDOM e.g., In a machine shop if a single operator is attending to several
machines he may attend first to the one nearest to him.
 LAST COME - FIRST SERVED.
Notations

 N = population

 M = service channels

 T = average service time

 W = average waiting time

 U = average running time or mean time between calls for service per unit

 H = average number of units being served.

 L = average number of units waiting for services.

 J = average number of units in operation

 F = efficiency factor

 X = service factor

 D = probability that if a unit calls for service, it will have to wait.

Measures of system efficiency

Efficiency of the repair system

 F = (T + U) / (T + U + W)
 Total number of machines N = J + H + L

 Probability that a machine is being:

o Repaired = H/N

o Running = J/N

o Waiting for repair = L/N

 X = T/ (T + U) (indicator of utilization of repair crew)

 H = NT/ (T + W + U) = FNX

 L = NW / (T + W + U) = N(1-F)

 J = NU / (T + W + U) = NF(1 – X)

 F = (H + J) / (H + L + J) = (T + U) / (T + W + U)
USE OF FINITE QUEUING TABLE

 Find mean service time and mean running time.

 Find out the service factor, the indicator of utilization of service crew or effectiveness
of service crew.

 Select the table corresponding to the population n.

 For the given population, estimate the service factor value.

 Read off from tables, value of D and F for the number of service crews M.

 If necessary these values may be interpolated between relevant values of X.

 Calculate the other measures, like waiting time, number of units waiting, units being
served, units in operation, etc.
CASELET
In a chemical factory there are 5 hoppers of identical size which feed material to
grinding mills. Due to changes in the requirement of materials, there are variations in the
time taken for emptying the hoppers. On the basis of past experience this time was found
to follow negative exponential distribution with an average of 10 hours between getting
emptied. Whenever a hopper gets empty, it has to be filled by a pay loader. Although the
capacity of the hopper is same, the time taken to fill the hoppers varies due to the
different locations from which the materials are to be loaded.
The time for filling the hopper also was found to follow negative exponential
distribution with an average of 2.5 hours. The company hires the pay loaders at a cost of
Rupees 100 per hour irrespective of whether it is operated or not. If the mill has to be
stopped due to its hopper getting empty it costs Rupees 1000 per hour in terms of loss of
profits. Determine the number of pay loaders which the company should engage to
minimize overall cost.
 Terminal question 1
Customers arrive at the first class ticket counter of a theater at a rate of 12 per hour. There
is one clerk serving the customers at a rate of 30 per hour.
(i) What is the probability that there is no customer in counter (i.e., the system is
idle).
(ii) What is the probability that there are more than 2 customers in the counter?
(iii) What is the probability that there is no customer is waiting to be served?
(iv) What is the probability that a customer is being served and nobody is waiting?
Terminal question 2
Assume that a bank teller window the customers arrive in their cars at the average rate of
twenty per hour according to Poisson distribution. Assume also that the bank teller spends
average of two minutes per customer to complete a service, and the service time is
exponentially distributed. Customer who arrive from an infinite population, are served on a
first come first served basis and there is no limit to possible queue length.
(i) What is the expected waiting time in the system per customer?
(ii) What is the mean number of customers waiting in the system?
(iii) What is the probability of zero customers in the system?
(iv) What is the value of utilization factor?

CHAPTER – 10
SIMULATION
Introduction
Thus far we have seen a variety of mathematical models employed to solve physical
problems. In building these models we have made certain assumptions and the situations have
restricted uses. There are occasions where the physical system or real life system becomes
rather complex to build a mathematical model. Even if the mathematical model can be
constructed to a reasonable degree of accuracy, available quantitative techniques may not be
amenable to solve the resulting models. In this way other means are resorted to analyze the
system. One such means is simulation, which can be a valuable tool to get an answer to a
particular problem.
Generally, simulation deals with the study of (dynamic) systems over time. There are several
types of simulations: eg., scale models airplanes tested in wind tunnels, a hydraulic or a
mechanical circuit converted into an analogous electric circuit, simulating and observing the
reactions of a certain medicine with animals and then to administer the same to human
beings, to build a laboratory model for a chemical compound and if proved successful,
extending the result to a large commercial scale. Inventory, queuing, scheduling and
forecasting serve as good example for simulation studies.
In all the above situations, a mathematical model of some real system is manipulated and the
results and information are observed. These manipulations and findings are used to make
inferences about the real system. If the model involves random sampling from a probability
distribution, the procedure is called Monte Carlo Simulation.
Basic concepts of simulation
Simulation is also called experimentation in the management laboratory. While dealing with
business problems, simulation is often referred to as “Monte Carlo Analysis”. Two
American mathematicians, Von Neumann and Ulan, in the late 1940s, found a problem in the
field of neuclear physics too complex for analytical solution and too dangerous for actual
experimentation. They arrived at an approximate solution by sampling. The method they used
had resemblance to the gambler’s betting systems on the roulette table, hence the name
“Monte carlo’ has stuck.
Imagine a betting game where the stakes are based on correct prediction of the number of
heads, which occur when five coins are tossed. If it were only a question of one coin, most
people know that there is an equal likelihood of a head or a tail occurring, that is the
probability of a head is ½. However, without the application of probability theory, it would
be difficult to predict the chances of getting various numbers of heads, when five coins are
tossed.
One way of experimentation is to take five coins and toss them repeatedly, note down the
outcomes of each toss after every ten tosses, approximate the the probabilities of various
outcomes. The values of these probabilities will initially fluctuate, but they would tend to
stabilize as the numbers of tosses are increased. This approach in effect is a method of
sampling, but is not very convenient. It is also time consuming and tiresome. Instead of
actually tossing the coins, one can conduct this experiment using random numbers. Random
numbers have the property that any number is equally likely to occur, irrespective of the digit
that has already occurred.
Let us estimate the probability of tossing of different numbers of heads with five coins. For
this we take five digit ramdom numbers. Each digit in the number will depict the head or tail.
By following a convention that “even” digits signify a head and the “odd” digits represent a
tail, the tossing of a coin can be simulated. The probability of occurrence of the first digits is
½ and that of the other set is also ½- a condition corresponding to the probability of the
occurrence of a head and the probability of occurrence of a tail respectively.
We start with a set of ten five digit random numbers as given below”
78466; 78722; 06401; 04754; 97118; 71923; 78870; 61208; 05003; and 95983.
The first set of five random digits in the above list (78566) implies that the outcome of the
first toss of five coins is as depicted in the table given below:

coin 1 2 3 4 5
Random 7 8 4 6 6
number
outcome T H H H H
Hence it is 4 heads and 1 tail.
Proceeding in the same way, we can tabulate the results of the first ten losses as depicted in
the table below:

Toss no. Number of

heads tails
1 4 1
2 3 2
3 4 1
4 3 2
5 1 4
6 1 4
7 3 2
8 4 1
9 3 2
10 1 4

Based on above experimentation, the estimates of probabilities of occurrence of different

numbers of heads with respect to the theoretical probability is as given below

No. of heads (n) Estimated probability Theoretical probability (5Cn (0.5)5)

0 0.00 0.03125
1 0.30 0.15625
2 0.00 0.3125
3 0.40 0.3125
4 0.30 0.15625
5 0.00 0.03125

The same technique we can use for any event. It is only necessary to take care that the set of
random allotted to any event matches with its probability of occurrence. For instance, if we
are interested in allotting random numbers to three events A, B and C with respective
probabilities 0.24, 0.36 and 0.40, choose the two digits random numbers from 00 to 99.
The numbers 00 to 23 signify event A, 24 to 59 signify B and 60 to 99 signify event C.
Definition
Simulation is a numerical technique for conducting experiments that involve certain type of
mathematical and logical relationships necessary to describe the structure and behavior of a
complex real world system over extended period of time.
“simulation is the procedure of defining of a model of a real system and conducting
experiments with this model for the purpose of understanding the behaviour (within the limits
imposed by a criterion or a set of criteria) for the operation of a system” -Shannon
Simulation may be called as experimentation in the management laboratory.

 In the context of business problems, simulation is often referred to as “Monte - Carlo

analysis”.
  Where formulating a mathematical model is difficult, simulation is of great help for
decision making.
Types of simulations
In general, simulation can be classified into three groups:

 Analogue model
 Continuous model
 Discrete model.
Analogue model.
In the analogue model, the physical (original) system is replaced by a model using an analogy
which is easier or amenable for manipulation. For example, vibration phenomenon in
mechanics is represented by an analogue electrical system.
Continuous model.
Continuous models represent the system undergoing smooth changes in the characteristics
over a certain time period. Observation of the variations of the different characteristics with
time is the object of this type of simulation model. For example, consider an inventory
control problem where the output might include time changes of inventory levels, sales,
unsatisfied demands and back orders.
Discrete model.
If the system is simulated with a model and observed only at selected points in time, we have
the discrete models. These time points coincide with the occurrence of certain events which
play an important role to effect changes in the performance of a system. In most cases we
face situations which belong to the discrete model.
Simulation procedure

 In any simulation problem the variables to be studied will be given with associated
probabilities.

 We assign random numbers to the variables.

 Identify the relationship among the variables and run the simulation to get the results.
Why simulation
Using simulation, an analyst can introduce the constants and variables related to the problem,
set up the possible courses of action and establish criteria which act as measure of
effectiveness. The major reasons for applying simulation technique to OR problems may be
listed as follows:

 It is an appropriate tool to use in solving a problem when experimenting on the real

system:
o Would be disruptive

o Would be too expensive

 o Does not permit replication events

o Does not permit control over key variables.

 It is desirable tool for solving a business problem when a mathematical model;

o Is too complex to solve

o Is beyond the capacity of available personnel

o Is not detailed enough to provide information on all important decision

variables.

 The major reasons for adopting simulation in place of other mathematical techniques
are:
o It may be the only method available because it is difficult to observe the actual
reality
o Without appropriate assumption,, it is impossible to develop a mathematical
solution.
o It may be too expensive to actually observe the system.

o There may not be sufficient time to allow the system to operate for a very long
time.

 It provides a trial and error movement towards the optimal solution. The decision
maker selects an alternative, experiences the effect of the selection and then improves
the selection. In this way, the selection is adjusted until it approximates the optimal
solution.
Methodology of simulation

 Step 1. Identify and clearly define the problem

 Step 2. List the statements of objectives of the problem

 Step 3. Formulate the variables that influence the situation and an exact or
probabilistic description of their possible values or states.

 Step 4. Obtain a consistent set of values (or states) for the variables, i.e., a sample of
probabilistic variables, random sampling technique may be used.

 Step 5. Use the samples obtained in step 4 to calculate the value of the decision
criterion, by actually following the relationships among the variables for each of the
alternative decisions.

 Step 6. Repeat steps 4 and 5 until a sufficient number of samples are available

 Step 7. Tabulate the various values of decision criterion and choose the best policy.
Solved problem 1
A sample of 100 arrivals of customers at a retail sales depot is according to the following
distribution:

Time between arrivals (minutes) Frequency

0.5 2
1.0 6
1.5 10
2.0 25
2.5 20
3.0 14
3.5 10
4.0 7
4.5 4
5.0 2

A study of time required to service customers by adding up the bills, receiving payments,
making change and placing packages in hand trucks, yields the following distribution:

Service time (minutes) Frequency

0.5 12
1.0 21
1.5 36
2.0 19
2.5 7
3.0 5

Estimate the average percentage customer waiting time and average percentage idle time
of the server by simulation for the next 10 arrivals.
Solution:
Step 1: convert the frequency distributions of time between arrivals and service time to
cumulative probability distributions.
Step 2: allocate random numbers 00 to 99 for each of the values of time between arrivals
and service time. The range allocated to each value corresponds to the value of
cumulative probability.
Step 3: using random numbers from table, sample the random time of arrival and service
time for ten sets of random numbers.
Step 4: tabulate waiting time of arrivals and idle time of servers.
Step 5: estimate the percentage waiting time of arrivals and percentage idle time of
servers corresponding ot the ten samples.
The tables below give the allocation of random numbers based on time between
arrivals and service time.
Allocation of random numbers – time between arrivals
Time between frequency Cumulative Cumulative Random
arrivals frequency probability numbers
allocated
0.5 2 2 0.02 00 to 01
1.0 6 8 0.08 02 to 07
1.5 10 18 0.18 08 to 17
2.0 25 43 0.43 18 to 42
2.5 20 63 0.63 43 to 62
3.0 14 77 0.77 63 to 76
3.5 10 87 0.87 77 to 86
4.0 7 94 0.94 87 to 93
4.5 4 98 0.98 94 to 97
5.0 2 100 1.00 98 to 99

Allocation of random numbers – service time

Service time frequency Cumulative Cumulative Random

between frequency probability numbers
arrivals allocated
0.5 12 12 0.12 00 to 11
1.0 21 33 0.33 12 to 32
1.5 36 69 0.69 33 to 68
2.0 19 88 0.88 69 to 87
2.5 7 95 0.95 88 to 94
3.0 5 100 1.00 95 to 99

Note that the upper bound of random numbers allocated for each value of the parameter is
one less than the corresponding cumulative frequency, since you have chosen a range of
random numbers from 00 to 99.
Now we can experiment with taking ten samples each of random numbers for time
between arrival and service time. Based on time between arrival and service time we can
calculate the time to start the service, time to finish the service. Base on time of start and
time of arrival, we can calculate the waiting time of arrivals, and based on time of start
and time to finish the service of previous arrival, we can calculate the idle time of server.
The table below gives the complete calculations
Arrival and service time

arrivals service
Arriv Rando Time Time Rando Servic Tim Tim Waitin Idle
al no. m betwee of m e time e of e of g time time
numbe n arrival numbe start finis of of
r arrival s r h arrival serve
s r
1 78 3.5 3.5 54 1.5 3.5 5.0 - 3.5
2 82 3.5 7.0 24 1.0 7.0 8.0 - 2.0
3 06 1.0 8.0 51 1.5 8.0 9.5 - -
 4 04 1.0 9.0 45 1.5 9.5 11.0 0.5 -
5 97 4.5 13.5 46 1.5 13.5 15.0 - 2.5
6 71 3.0 16.5 84 2.0 16.5 18.5 - 1.5
7 79 3.5 20.0 59 1.5 20.0 21.5 - 1.5
8 61 2.5 22.5 58 1.5 22.5 24.0 - 1.0
9 07 1.0 23.5 60 1.5 24.0 25.5 0.5 -
10 95 4.5 28.0 25 1.0 28.0 29.0 - 2.5
total 1.0 14.5

The service facility is made available at clock time zero and the server has to be idle for
3.5 minutes, when the service for first arrival starts. The service is completed at 5.0
minutes and again the server is idle for 2 minutes till the second arrival joins the system.
The first three arrivals get immediate service and they don’t have to wait, as the server is
idle when they arrive. The fourth arrival that joins at 9.0 minutes has to wait for 0.5
minutes, while the service to the third is completed. Similarly the waiting time and idle
time can be computed for further arrivals.
Total elapsed time = 29 minutes
Waiting time of arrivals = 1 minutes
Percentage of waiting time = 1 x 100 / 29 = 3.4%
Idle time for server = 14.5 minutes
Percentage of idle time = 14.5 x 100 / 29 = 50%
Random numbers
In carrying out Monte Carlo simulation one needs to generate random numbers to obtain
random observations from a probability distribution. A random number is a number in a
sequence of numbers whose probability of occurrence is the same as that of any other
number in the sequence.
Definition
Random numbers have the property that any number is equally likely to occur
irrespective of the digit that has already occurred.

 The random numbers can be selected by any random process:

o Drawing chits from a hat.

o Table of random numbers prepared on the basis of some random physical

process.
o Manually generate random numbers

o To use random number table in the computer memory

o To use computer methods

 o To use an arithmetic operation to compute a sequence of random number from
a recursive equation
Manual methods are laborious and not practical except for illustrative purpose in the class
room. The devices used include rolling dice, picking the number from a box by shuffling,
use of roulette wheel where the disk is marked with numbers and tossing a coin. Tables of
random numbers are available and these are generated using some random physical
process (electric current in Rand Case) and they are considered to be truly random
numbers. The problem with using table of random numbers is making them available in a
computer programme. It would be necessary to record random numbers on a hard disk /
memory, which is more desirable one. Moreover, calling for a random number when
needed is a relatively slow computer process.
Pseudo-random numbers.
The most common method to obtain random number is to generate pseudo random
numbers through a computer programme. They are obtained using a deterministic
mathematical (arithmetic) operation. Truly random numbers cannot be generated by an
algorithm and hence random numbers generated by using a recursive equation are
referred to as pseudo random numbers.
 Mid Square method.
 Multiplicative congruential method

 Mid square method

 The operation starts with an arbitrary four digit number called the seed.
 The seed is squared and all digits except the middle four are ignored. The
middle four will depict the next random number.
 The process is repeated each time using the next random number as the new
seed.
Seed U0 = 8695
(U0)2 = (8695)2
= 75603025
Taking the middle four digits
U1 = 6030
(U1)2 = (6030)2 = 36360900
U2 = 3609
Repeating the above procedure:
U3 = 0248, U4 = 0615, U5 = 3782, U6 = 3035, U7 = 2112,
U8 = 4605, U9 = 2060, U10 = 2436
The basic disadvantage of the mid square method is that the generated numbers may start
cycling after a sort set of random numbers.

 Multiplicative congruential method

There is another method of arithmetic operation called multiplicative congruential method
which is an effective one. We define that the notation
Y = x mod (m)
Means x – y is an integer multiple of m and 0 ≤ y < m. for example 2 = 92 mod 3 thus (92
– 2) / 3 = 30 which is an integer.
The general recursive equation that is used with this method is given by
Yn = k yn-1 (mod m)
Where yn is the sequence of random numbers and k and m are the parameters. The first
number must be specified in advance along with the parameter k and m. the determination
of sequence of random numbers yn depends on the choice of k, m and y0.
For example, suppose y0 = 3, k = 7 and m = 10 so that all integers in the sequence have a
range from 0 to 9. Thus, y1 = 7 x 3 / 10 = (20 + 1) / 10 so that y 1 = 1, similarly 7 x 1 / 10 =
(0 + 7) / 10, so that y2 = 7; 7 x 7 / 10 = (40 + 9) / 10, and hence y3 = 9 and so on.
Sample size.
The larger the number of trials, the more confident we can be in our estimates. The
sample size required for a given confidence level at a specified degree of accuracy can be
worked out.
One can only say that the accuracy associated with simulation improves as the square root
of the number of trials and hence there is a need for a large number of trials.
Solved problem 2
If it is needed to be 95% certain of being correct in an experiment with marginal error of
1% of the true value, what should be the sample size?
Solution
Let p = percentage of success
N = sample size
Then standard error = [p(100 – p)/n]0.5
Standard normal deviate value corresponding to 95% confidence level from normal
table is 1.96
Margin of error = 1.96 [p(100 – p)/n]0.5 = 1
n = 1.962 p(100 – p)
p(100 – p) is maximum at p = 50
 hence n = 9600

Solved problem 3
A bread vendor buys every morning loaves of bread at 0.45 each by placing his order one
day in advance (at the time of receiving his previous order) and sells them at Rupees 0.70
each. Unsold bread can be sold the next day at Rupees 0.20 per loaf and thereafter should
be treated as of no value. The pattern of demand for bread is given below:

Fresh bread One day old bread

Daily sales Probability of Daily sales Probability of
demand demand
50 0.01 0 0.70
51 0.03 1 0.20
52 0.04 2 0.08
53 0.07 3 0.02
54 0.09
55 0.11
56 0.15
57 0.21
58 0.18
59 0.09
60 0.02

The vendor adopts the following order rules. If there is no stock with him at the end of
previous day, he orders 60 units. Otherwise, he orders 50 or 55, whichever is nearest the
actual fresh bread sale on the previous day. Starting with 0 stocks and a pending order of
55 loaves, simulate for 10 days and calculate the vendor’s profits.
Solution
Allocation of random numbers

Fresh bread One day old bread

Daily Probability Cumulative Random Daily Probability Cumulative Random
sales of demand probability number sales of demand probability number
allocated allocated
50 0.01 0.01 00 0 0.70 0.70 00 – 69
51 0.03 0.04 01 – 03 1 0.20 0.90 70 – 89
52 0.04 0.08 04 – 07 2 0.08 0.98 90 – 97
53 0.07 0.15 08 – 14 3 0.02 1.00 98 - 99
54 0.09 0.24 15 – 23
55 0.11 0.35 24 – 34
56 0.15 0.50 35 – 49
57 0.21 0.71 50 – 70
58 0.18 0.89 71 – 88
59 0.09 0.98 89 – 97
60 0.02 1.00 98 - 99
We can now construct a table to see through simulation how the stocks and sales
fluctuate.
Table of simulation result

Fresh bread One day old bread

day Receipt Random sale Closin Order Opening Random sale
of the number g stock for stock number
start of next
the day day
1 55 72 55 0 60 0 0
2 60 06 52 8 50 0 0
3 50 12 50 0 60 8 86 1
4 60 14 53 7 55 0 0
5 55 79 55 0 60 7 54 0
6 60 70 57 3 55 0 0
7 55 85 55 0 60 3 88 1
8 60 71 58 2 55 0 0
9 55 21 54 1 55 2 58 0
10 55 98 55 0 60 1 48 0
total 565 544 21 2

Estimated profit = (544 x 0.70 + 2 x 0.20) – 565 x 0.45 = 126.95

Solved problem 4
The maintenance manager of a chemical company is interested in determining a rational
policy for maintenance of pneumatic conveying equipment. The equipment is a part of the
process in line and hence affects production. It has one bearing each on the inlet side (A)
and the outlet side (B). Whenever there is a failure of any bearing, it has to be replaced
immediately. The company has a good system of maintaining records on performance of
the equipment and the failure records are available.

Number of failures
Age of failures (hours) Inlet side bearing (A) Outlet side bearing (B)
150 7 0
300 16 3
450 18 9
600 23 12
750 14 16
900 10 18
1050 7 33
1200 5 7
1350 0 2
The cost of bearing is rupees 300 each for ‘A’ and rupees 500 each for ‘B’. The cost of
downtime of equipment is Rupees 700 per hour, and it takes 2 hours to replace one
bearing either at inlet or outlet side and 3 hours to replace both the bearings.
The two maintenance policies to be evaluated are

 Replace a bearing only when it fails.

 Replace both the bearings if one fails.

Find the best alternative through simulation.
Solution
It is assumed that the failure of a bearing is independent of the maintenance policy
followed. Random numbers are allocated for different failure times.
Allocation of Random numbers

Bearing A Bearing B
Life Frequency Cumulative Random Frequency Cumulativ Random
hours of failure frequency numbers of failure e frequency numbers
150 7 7 00 – 06 0 0 -
300 16 23 07 – 22 3 3 00 – 02
450 18 41 23 – 40 9 12 03 – 11
600 23 64 41 – 63 12 24 12 – 23
750 14 78 64 – 77 16 40 24 – 39
900 10 88 78 – 87 18 58 40 – 57
1050 7 95 88 – 94 33 91 58 – 90
1200 5 100 95 – 99 7 98 91 – 97
1350 0 - - 2 100 98 - 99

We can now select random numbers from the tables and generate a set of 12
bearings for each of the bearings. Table below depicts the set of 12 bearings for each
of the bearings generated.
Set of 12 bearings for each of the bearings generated

Bearing Bearing A Bearing B

no.
1 10 300 300 99 1350 1350
2 22 300 600 96 1200 2550
3 24 450 1050 18 600 3150
4 42 600 1650 36 750 3900
5 37 450 2100 50 900 4800
6 77 750 2850 79 1050 5850
7 99 1200 4050 80 1050 6900
8 96 1200 5250 96 1200 8100
9 89 1050 6300 31 750 8850
10 85 900 7200 07 450 9300
11 28 450 7650 62 1050 10350
 12 63 600 8250 77 1050 11400

Let us compare the costs of three policies for the first 7200 hours.
Policy 1: replace a bearing only when it fails.
In 7200 hours, bearing A requires replacement 10 times and bearing B 7 times as seen
from the lives of these bearings.
Total cost = 10 x 300 + 500 x 7 + 17 x 2 700 = rs. 30, 300
Policy 2: replace both the bearings if one fails.
Table below depicts bearing failure information.
Bearing failure and replacement

Elapsed time Bearing fails first Life (hours) of Life (hours) of

(hours) newly fitted newly fitted
bearing A bearing B
300 A 300 1200
600 A 450 600
1050 A 600 750
1650 A 450 900
2100 A 750 1050
2850 A 1200 1050
3900 B 1200 1200
5100 A&B 1050 750
5850 B 900 450
6300 B 450 1050
6750 A 600 1050
7350 A - -

Total cost = 11 x 800 + 11 x 3 x 700 = 31900

Policy 1 is cheaper.
Solved problem 5
A factory produces 150 scooters. But the production rate varies with the following
distribution:

Production 147 148 149 150 151 152 153

rate
Probability 0.05 0.10 0.15 0.20 0.30 0.15 0.05

At present it calls for a track which will hold 150 scooters. Using the following random
numbers determine the average number of scooters waiting for shipment in the factory
and average number of empty space in the truck.
Random numbers: 82, 54, 50, 96, 85, 34, 30, 02, 64, 47.
Solution:
Table below depicts the production rate and probability
Production rate and probability

Production rate probability Cumulative Random numbers

probability allotted
147 0.05 0.05 00 – 04
148 0.10 0.15 05 – 14
149 0.15 0.30 15 – 29
150 0.20 0.50 30 – 49
151 0.30 0.80 50 – 79
152 0.15 0.95 80 – 94
153 0.05 1.00 95 - 99

Table below gives the simulation outcome.

Trial no. Random no. Simulated Scooter Number of

production waiting in the empty space in
rate factory the truck
1 82 152 2 -
2 54 150 2 -
3 50 150 2 -
4 96 153 5 -
5 85 152 7 -
6 34 150 7 -
7 30 150 7 -
8 02 147 4 3
9 64 151 5 -
10 47 150 7 -
total 48 3
Therefore average number of scooters waiting = 48 / 10 = 4.8 per day
Average number of empty space = 3 / 10 = 0.3 per day.
Solved problem 6
Dr. Strung is a dentist. He gives appointment to the patients every half an hour. However
he does not know the nature of illness of patients arriving at this clinic. From past records
he has the following probability distribution and also knows the exact treatment timings.
He starts his clinic at 8.00 am. Using the following information, determine the average
waiting time of the customers and idle time of the doctor.

Nature of illness Probability Time taken for treatment

Filling 0.10 50
Check up 0.30 15
Crowning 0.15 40
Cleaning 0.30. 15
Extraction 0.15 30
Random number 56, 40, 26, 66, 87, 48, 17, 22, 04, and 15
Solution:
Teble below depicts the allocation of random numbers.
Allocation of random numbers

Nature of illness Probability Cumulative Allocated random

probability numbers
Filling 0.10 0.10 00 – 09
Check up 0.30 0.40 10 – 39
Crowning 0.15 0.55 40 – 54
Cleaning 0.30. 0.85 55 – 84
Extraction 0.15 1.00 85 - 99

Table below shows the simulation outcomes

Simulation work sheet

Trial Rando Nature of Time Patient treatment Docto Patients

no. m no. illness taken s r idle waiting
arrival time time
time
starts finishe
s
1 56 Cleaning 15 08.00 08.00 08.15 - -
2 40 Crowning 40 08.30 08.30 09.10 15 -
3 26 Check up 15 09.00 09.10 09.25 - 10
4 66 Cleaning 15 09.30 09.30 09.45 5 -
5 87 Extraction 30 10.00 10.00 10.30 15 -
6 48 Crowning 40 10.30 10.30 11.10 - -
7 17 Check up 15 11.00 11.10 11.25 - 10
8 22 Check up 15 11.30 11.30 11.45 5 -
9 04 Filling 50 12.00 12.00 12.50 15 -
10 15 Check up 15 12.30 12.50 13.05 - 20
Total 55 40
Thus, doctor’s idle time = 55 minutes
Patients average waiting time = 40 / 10 = 4 minutes.

Monte Carlo method

Introduction
The technique of Monte – Carlo involves the selection of random observations within
the simulation model.
The technique is restricted for application involving random numbers to solve
deterministic and stochastic problems.
The principle
Replacement of actual statistical universe described by some assumed probability
distribution and then sampling from this theoretical population by means of random
numbers.

 The approach has the ability to develop many months or years of data in a matter of
few minutes on a computer.

 The method is generally used to solve the problems which cannot be adequately
represented by mathematical model or where solution of the model is not possible by
analytical method.
Simulation procedure

 Step 1. Define the problem

 Step 2. Construct an appropriate model.

 Step 3. Prepare the model for experimentation.

 Step 4. Using step 1 and 3, experiment with the model

 Step 5. Summarize and examine the results obtained in step 4.

 Step 6. Evaluate the results of the simulation

 Step 7. Formulate proposals for advice to management.

Step 1. Define the problem

 Identify the objectives of the problem.

 Identify the main factors which have the greatest effect on the objectives of
the problem.
Step 2. Construct an appropriate model.

 Specify the variables and parameters of the model.

 Formulate the appropriate decision rules, i.e., state the conditions under which
the experiment is to be performed.

 Identify the type of distribution that will be used – models use either
theoretical distributions or empirical distributions to state the patterns the
occurrence associated with the variables.

 Specify the manner in which the time will change.

 Define the relationship between the variables and parameters.

Step 3. Prepare the model for experimentation.

 Define the starting conditions for the simulation.

 Specify the number of runs of simulation to be made.

Step 4. Using step 1 and 3, experiment with the model

 Define a coding system that will correlate the factors defined in step 1 with the
random numbers to be generated for the simulation.

 Select a random number generator and create the random numbers to be used
in the simulation.

 Associate the generated random numbers with the factors identified in step 1
and coded in earlier (first step of step 4)
Step 5. Summarize and examine the results obtained in step 4.
Step 6. Evaluate the results of the simulation
Step 7. Formulate proposals for advice to management on the course of action to be
adopted and modify the model, if necessary.

Solved problem 7
The occurrence of rain in a city on a day is dependent upon whether or not it rained on
previous day. If it rained on previous day, the rain distribution is:

Event No rain 1 cm rain 2 cm rain 3 cm rain 4 cm rain 5 cm rain

Probabilit 0.50 0.25 0.15 0.05 0.03 0.02
y

If it did not rain the previous day, the rain distribution is:

0.75 0.15 0.06 0.04 - -

Simulate the city’s weather for 10 days and determine by simulation the total days
without rain as well as the total rain fall during the period. Use the following random
numbers.
67, 63, 39, 55, 29, 78, 70, 06, 79, 76.
For simulation, assume that for the first day of the simulation it had not rained the day
before.
Solution:
We simulate the city’s weather with and without rain fall in the following steps:
Step 1: Previous day rain distribution and random number allocated is as given in
the table:
Rain Distribution Table

event probability Cumulative Random number

probability allocated
 No rain 0.50 0.50 00 – 49
1 cm rain 0.25 0.75 50 – 74
2 cm rain 0.15 0.90 75 – 89
3 cm rain 0.05 0.95 90 – 94
4 cm rain 0.03 0.98 95 – 97
5 cm rain 0.02 1.00 98 - 99

Step 2: previous day no rain distribution is as depicted in table below:

Rain Distribution Table

event probability Cumulative Random number

probability allocated
No rain 0.75 0.75 00 – 74
1 cm rain 0.15 0.90 75 – 89
2 cm rain 0.06 0.96 90 – 95
3 cm rain 0.04 1.00 96 – 99

Step 3: simulation outcomes for 10 days using the given random numbers is as
shown in the table
Simulation Work Sheet

day Random number events Cumulative rain

1 67 No rain -
2 63 No rain -
3 39 No rain -
4 55 No rain -
5 29 No rain -
6 78 1 cm rain 1 cm
7 70 1 cm rain 2 cm
8 06 No rain 2 cm
9 79 1 cm rain 3 cm
10 76 2 cm rain 5 cm

Thus, during the simulated period it did not rain on 6 out of 10 days.
The total rain fall during the period is 5 cm.
Solved problem 8
The number of customers at a restaurant each evening is distributed as shown below:

Number of Lots Average Very few

customers
Probability 0.2 0.4 0.4
The chef refuses to work on an evening when there are very few customers and walks out.
He will not return to work until an evening when there are lots of customers although he
always comes in on Friday because he gets paid then. We are interested in fraction of
evenings that the chef is at the restaurant.
Solution:
In this example, the trial is whether the number of customers is ‘lots’; ‘average’ or
‘very few’ and it is repeated for each evening when the restaurant is open. Also,
whether the chef is ‘in’ or ‘out’ depends not only on the outcome of this trial, but
also on his whereabouts the previous evening and whether it is Friday or not.
Here we wish to simulate the level of attendance at the restaurant on each evening.
To determine whether he is ‘in’ or ‘out’ on that particular evening. We use the rules
governing the chef’s behaviour as stated in the question. Here, the sequence of ‘in’s’
and ‘out’s’ forms a simulated history of evenings for which chef is ‘in’. now we can
proceed as follows;
Step 1: to simulate the number of customers each evening.
Using the random digits and the given (prescribed) probability distribution, we
simulate the number of customers each evening as depicted in the table below:
Simulation of customers

Number of lots Average Very few

customers
Probability 0.2 0.4 0.4
digits 0, 1 2, 3, 4, 5 6, 7, 8, 9

Step 2: to generate attendance level

The attendance on a particular evening can be simulated by selecting a digit, from 0
to 9 inclusive, at random. Using the table above for simulation of customers, we can
interprete the digits as state of restaurant. On account of the number of digits
assigned to each one, the outcomes in the simulation will have the same probability
of occurrence as those in the real system.
Step 3: to determine whether the chef works or not.
After preparing an attendance level, we can determine whether the chef works or
not by the rules provided in the question. Table below depicts a specimen run of the
simulation.
The following specimen run was started with ‘Monday’ and with the chef being ‘in’,
keeping the view in mind that we are trying to reproduce the long term behaviour of
the restaurant. The initial conditions of the simulation will not affect the results we
obtain.
Simulation table

day Day of the Rando Number Chef Number Fraction

week m digit of ‘in’ or of days of days
 customer ‘out’ ‘in’ to ‘in’ to
s date date
1 Monday 3 average in 1 1.00
2 Tuesday 5 average in 2 1.00
3 Wednesday 1 lots in 3 1.00
4 Thursday 8 Very few out 3 0.75
5 Friday 6 Very few in 4 0.80
6 Saturday 7 Very few Out 4 0.67
7 Sunday 5 average Out 4 0.57
8 Monday 4 average Out 4 0.50
9 Tuesday 0 lots in 5
10 Wednesday 0.56

Step 4: when to stop the simulation run

As far as determination of any average effects is concerned, the more time periods
simulated the better. However, in actual practice we require a criterion for deciding
when the simulation to stop. In the specimen run we could continue until the
fraction at the end of each line does not change very much. At this stage there is no
point going further and we have the desired answer. In this example, it approaches
0.56 and the chef is ‘in’ 56% of all days.
Application of simulation

 Simulation can be easily understood by the users and thereby facilitates their
active involvement.

 Makes the results more reliable and also ensures easy acceptance for
implementation.

 Can be employed for wide variety of problems, viz:

o The policy for optimal maintenance

o Number of maintenance crews

o Number of equipments for handling materials

o Job shop scheduling

o Routing problem

o Stock control

 Other areas of application include:

o Dock facilities

o Facilities at airport to minimize congestion.

o Hospital appointment system.

 o Management games

 With the help of simulation, the manager tries to strike a balance between
opposing costs of providing facilities and the opportunity and other costs
of not providing them.
Simulation in networks
A number of network simulation models have also been developed, e.g., simulation of
probabilistic activity times in pert net work. The critical path and the project duration can be
found out with a randomly selected activity times for each activity. The probability
distribution of project completion time and the probability that each of given activity is on the
critical path can be obtained by repeating the process a number of times.
Solved problem 9
A project schedule consists of five activities and the duration of these activities is non –
deterministic with the following probability distribution:

Activity Days Probability

1–2 1 0.20
4 0.50
8 0.30
1–3 2 0.30
4 0.50
7 0.20
2–4 2 0.30
4 0.30
6 0.40
3–4 3 0.30
4 0.40
8 0.30
4-5 2 0.20
3 0.20
4 0.60

Simulate the duration of the project 10 times and estimate the chances of various paths to be
critical. What is the average duration of the project?
Solution:

From the PERT network, we observe that there are two paths from start to end of the project,
namely, 1 – 2 – 4 – 5 and 1 – 3 – 4 – 5. Using one digit random number corresponding to the
probability distribution of various activities, we obtain the simulated duration of each path
and hence the simulated duration of various activities and the simulation duration of the
project.
Table below depicts the random numbers allocated.
Allocation of random numbers

Activity Days Probability Cumulative Random

probability numbers
allocated
1–2 1 0.20 0.20 0,1
4 0.50 0.70 2,3,4,5,6
8 0.30 1.00 7,8,9
1–3 2 0.30 0.30 0,1,2
4 0.50 0.80 3,4,5,6,7
7 0.20 1.00 8,9
2–4 2 0.30 0.30 0,1,2
4 0.30 0.60 3,4,5
6 0.40 1.00 6,7,8,9
3–4 3 0.30 0.30 0,1,2
4 0.40 0.70 3,4,5,6
8 0.30 1.00 7,8,9
4-5 2 0.20 0.20 0,1
3 0.20 0.40 2,3
4 0.60 1.00 4,5,6,7,8,9

Table below depicts the result of the simulation

Simulation table

Activity
Simulati 1–2 1–3 2 –4 3-4 4-5 Project Critical path
on no. RN D RN D RN D RN D RN D duration
1 8 8 9 7 8 6 3 4 6 4 18 1–2–4-5
2 5 4 5 4 8 6 1 3 7 4 14 1–2–4-5
3 5 4 6 4 7 6 4 6 7 4 14 1 – 2 – 4 –5 &
1–3–4-5
4 7 8 5 4 1 2 9 8 5 4 16 1–3–4-5
5 0 1 6 4 8 6 1 3 8 4 11 1 – 2 – 4 –5 &
1–3–4-5
6 8 8 3 4 0 2 4 4 3 3 13 1 – 2 – 4 –5
7 4 4 7 4 4 4 4 4 3 3 11 1 – 2 – 4 –5 &
1–3–4-5
8 5 4 8 7 2 2 6 4 6 4 15 1–3–4-5
9 2 4 5 4 9 6 0 3 3 3 13 1 – 2 – 4 –5
10 6 4 1 2 3 4 7 8 0 2 12 1–3–4-5
On the basis of the above simulation, we observe that there is a 70% chance of the path
1 – 2 – 4 – 5 being critical. Further, the average duration of the project is 13.7 days.
Solved problem 10
A project consists of 7 activities. The time for performance of each of the activities is a
random variable with the respective probability distribution as given below:

Activity Immediate Time in days and its probability

predecessor
A - 3 4 5 - -
0.20 0.60 0.20
B - 4 5 6 7 8
0.10 0.30 0.30 0.20 0.10
C A 1 3 5 - -
0.15 0.75 0.10
D B, C 4 5 - - -
0.80 0.20
E D 3 4 5 6 -
0.10 0.30 0.30 0.30
F D 5 7 - - -
0.20 0.80
G E, F 2 3 - - -
0.50 0.50

 Draw the network diagram and identify the critical path using the expected activity
times.

 Simulate the project using the random numbers to determine the activity times. Find
the critical path and the project duration.

 Repeat the simulation four more times. Is the same path critical in all the simulations?
Solution:
(a) The network diagram is as given below:

5
D
3 4
5 1

A C E
1 D G 2
7
6

As per network diagram, following paths are visualized:

 1–3–4–6–7
 1–2–3–4–6–7
 1–3–4–5–6–7
 1–2–3–4–5–6–7
The expected time of activities can be calculated according to the probabilities:
E = ΣPiDi
For example:
Expected time for activity A = 3 x 0.20 + 4 x 0.60 + 5 x 0.20 = 4 days
Similarly, B = 4 x 0.10 + 5 x 0.30 + 6 x 0.30 + 7 x 0.20 + 8 x 0.10 = 5.9 days
C = 1 x 0.15 + 3 x 0.75 + 5 x 0.10 = 2.9 days
D = 4 x 0.80 + 5 x 0.20 = 4.2 days
E = 3 x 0.10 + 4 x 0.30 + 5 x 0.30 + 6 x 0.30 = 4.8 days
F = 5 x 0.20 + 7 x 0.80 = 6.6 days
D1 = dummy activity = 0 days
G = 2 x 0.50 + 3 x 0.50 = 2.5 days
The expected duration of the various paths;

 1 – 3 – 4 – 6 – 7 = 5.9 + 4.2 + 6.6 + 2.5 = 19.2 days

 1 – 2 – 3 – 4 – 6 – 7 = 4 + 2.9 + 4.2 + 6.6 + 2.5 = 20.2 days
 1 – 3 – 4 – 5 – 6 – 7 = 5.9 + 4.2 + 4.8 + 0 + 2.5 = 17.4 days
 1 – 2 – 3 – 4 – 5 – 6 – 7 = 4 + 2.9 + 4.2 + 4.8 + 0 + 2.5 = 18.4 days
Thus, critical path is 1 – 2 – 3 – 4 – 6 – 7 with expected duration of 20.2 days
(b) Simulation of the project
First we allocate the random numbers for duration based on the probability. The
random numbers allocated are as given below in the table:
Allocation of random numbers

activity time probability Cumulative Random

probability numbers
allocated
A 3 0.20 0.20 00 – 19
4 0.60 0.80 20 – 79
5 0.20 1.00 80 - 99
B 4 0.10 0.10 00 – 09
5 0.30 0.40 10 – 39
6 0.30 0.70 40 – 69
7 0.20 0.90 70 – 89
8 0.10 1.00 90 – 99
C 1 0.15 0.15 00 – 14
3 0.75 0.90 15 – 89
5 0.10 1.00 90 – 99
 D 4 0.80 0.80 00 – 79
5 0.20 1.00 80 – 99
E 3 0.10 0.10 00 – 09
4 0.30 0.40 10 – 39
5 0.30 0.70 40 – 69
6 0.30 1.00 70 – 99
F 5 0.20 0.20 00 – 19
7 0.80 1.00 20 – 99
G 2 0.50 0.50 00 – 49
3 0.50 1.00 50 - 99

Table below gives the outcomes of simulation

Simulation results
Simul A B C D E F G Project Critical path
ation RN D RN D RN D RN D RN D RN D RN D duration
1 68 4 13 5 09 1 20 4 73 6 07 5 92 3 18 1–3–4–5–6–
7&1–2–3–4–
5–6-7
2 99 5 93 8 18 3 20 4 22 4 07 5 29 2 19 1–3–4–6–7&
1–2–3–4–6-
7
3 57 4 33 5 49 3 65 4 92 6 98 7 00 2 20 1–2–3–4–6
-7
4 57 4 12 5 31 3 96 5 86 6 92 7 91 3 22 1–2–3–4–6
-7
5 77 4 37 5 34 3 11 4 27 4 10 5 59 3 19 1–2–3–4–6
-7

The same path is not critical in each of the runs. The average critical path length
(project duration) is 19.6 days.
Simulation of job sequencing
Solved problem 11
A job has to be processed over two machines, m 1 and m2 in that order. The
distribution of inter – arrival time of the jobs at the first machine is as follows:

Time (minutes) 1 2 3 4
Probability 0.20 0.20 0.20 0.40

The processing time at the two machines are as follows:

Machine M1 Machine M2
Time (minutes) Probability Time (minutes) Probability
1 0.10 4 0.20
2 0.20 5 0.30
3 0.30 6 0.40
4 0.30 7 0.10
5 0.10 - -
On the basis of 10 simulation runs, find out the average queue length before machine M1 and
the average queue length before machine M2.
Solution:
The various steps of simulation are;
1. simulate the inter – arrival time of the i th job on machine M1 with the help of a
random number (I = 1, 2, ….).
2. arrival time of ith job
= arrival time of (i - 1)th job (I = 1, 2, …) + inter – arrival time of ith job
Arrival time of first job = inter – arrival time of the first job.
3. Simulate the processing time on the machine M1 by random numbers
4. Departure time on machine M2
= arrival time for processing on machine M2
= max (arrival time of ith job, process completion time of (i – 1)th job on machine
M1) + processing time of ith job on machine M1
5. If arrival time of ithjob < process completion time of (i – 2) th, both ith and (i -1)th
arrivals wait.
6. Simulate the processing time of machine M2 with the help of random numbers.
7. Process completion by machine M2 of ith arrival
= max (arrival time of ith job, process completion time of (i -1) th job) + process
time by machine M2.
8. If arrival time of ith job < process completion time of (i -1) th job, the ith arrival
waits.
If arrival time of ith job < process completion time of (i -2)th job, both ith and (i
-1)th jobs wait, etc.
Using this procedure, the simulation processing data of 10 arrivals are displayed
in the simulation worksheet as given in the table below:
Simulation of arrival jobs
Simula RN Inter – Arriv R Proce Departu R Proces Complet Queu Queu
tion arrival al N ss re time N s time ion time e size e size
no. time time time on M1 on M2 on M2 on on
on on M1 M1 M2
M1
1 8 4 4 0 1 5 3 5 10 - -
2 7 4 8 6 4 12 0 4 16 - -
3 8 4 12 9 5 17 0 4 21 - -
4 9 4 16 3 3 20 0 4 25 1 1
5 4 3 19 7 4 24 0 4 29 1 1
6 4 3 22 0 1 25 2 5 34 1 1
 7 0 1 23 0 1 26 4 5 39 2 2
8 6 4 27 5 3 30 5 6 45 - 2
9 8 4 31 6 4 35 1 4 49 - 2
10 7 4 35 4 3 38 2 5 54 - 3
On the basis of simulation study, the average queue length before machine M1 is
0.5 and the average length before machine M2 is 1.2.
Advantages of simulation

 The study of very complicated system or sub – system can be done with the help of
simulation. Simulation has been described as ‘what to do when all else fails’.

 By using simulation, we can investigate the consequences for a system of possible

changes in parameters in terms of the model.

 The knowledge of a system obtained in designing and conducting the simulation is

very valuable.

 It enables us to assess the possible risks involved in a new policy before actually
implementing it.

 The simulation of complicated systems helps us to locate which variables have the
important influences on the system performance.

 Simulation methods are easier to apply than pure analytical methods.

Limitations of simulation

 Simulation generates a way of evaluating the solutions but it does not generates the
solution techniques.

 Sometimes simulation models are expensive and take a long time to develop it. For
example a corporate planning model may take a long time to develop and prove
expensive also.

 The simulation model does not produce the answers by itself. The user has to provide
all the constraints for the solution which he wants to examine.

 Not all situations can be evaluated using simulation. Only situations involving
uncertainty are considered.

 It is the trial and error approach that produces different solutions in repeated runs.
This means it does not generate optimal solutions to problems.

 A time consuming exercise.

Review questions
Question 1
Two components to be produced in m/c ‘A’ m/c ‘B’ and then finally assembled. The
time taken to assemble on two machines varies with the following probability
distribution. Using simulation technique and the ordered pairs of the random numbers,
first for m/c ‘A’ and second for m/c ‘B’, find the average time taken to produce.
 Machine ‘A’ Machine ‘B’
Production time in Probability Production time in Probability
minutes minutes
22 0.15 30 0.05
23 0.20 31 0.15
24 0.30 32 0.25
25 0.20 34 0.25
26 0.15 35 0.20
36 0.10

Random number (10, 92), 25, 83), (36, 76), (44, 15), (57, 25), (62, 67), (04, 99), (72, 53),
(81, 35), (94, 07).

Question 2
For a project comprising activities A, B, ….. H, the following information is available:
Precedence relationships:
A and B are the first activities of the project. C succeeds A while B precedes D. Both C and
D precede E and F. Activity G follows F while H is the last activity of the project and
succeeds E and G.
Time estimates and probabilities

Activit TIME DAYS

y
1 2 3 4 5 6
A 0.20 0.40 0.40
B 0.30 0.70
C 0.30 0.30 0.40
D 0.20 0.60 0.20
E 0.60 0.40
F 0.80 0.20
G 0.30 0.50 0.20
H 0.10 0.20 0.30 0.40

Simulate the project and determine, using random numbers, the activity times. Find the
critical path and the project duration. Repeat five times. Does the critical path change? State
the estimated duration of the project in each of the trials.
Question 3.
Calculate the probability of getting 2 heads and 1 tail in tossing 3 coins simultaneously.
Estimate the probability by simulation for 10 trials using random numbers for the same. What
is the difference between the two probabilities?
Question 4.
Maintenance of machines in a factory can be carried out by one man or sometimes by a two-
man crew. The times taken with one man crew are 1.5, 2.0, 2.5, or 3.0 hours with
probabilities of 0.20, 0.30, 0.35 and 0.15 respectively. A two men crew requires 0.75, 1.00,
1.50 or 2.00 hours with probabilities of 0.25, 0.35, 0.20 and 0.20 respectively. The cost of
labour is Rs. 7 per hour and the over head expenses are Rs. 3 per hour per crew. Decide
whether one or two men crew is more economical.
Question 5.
A transport corporation operates 15 long distance buses, each with one driver. The
corporation has a policy of keeping three reserve drivers on call to replace those who report
sick. If a driver is not available, the bus trip is cancelled. The probability distribution for the
daily number of sick drivers is as follows:

Number 0 1 2 3 4 5
sick
probability 0.17 0.25 0.23 0.15 0.10 0.10

Use Monte – Carlo methods to determine the utilization of reserve drivers and also the
probability that atleast one trip is cancelled because no driver is available. Compare with
exact answers.
Question 6.
The demand distribution and the lead time distribution are as given below:
Demand distribution

Demand (items / 150 200 240

week)
probability 0.25 0.50 0.25

Lead ttime distribution

Lead time(weeks) 1 2 3
probability 0.30 0.50 0.20

Simulate the demand over the lead time.

Question 7:
A bakery supplies special cakes to one of its shops each day. The number of cakes supplied
each day has the following distribution.
Supply distribution

Cakes / 15 16 17 18 19 20
day
probabilit 0.10 0.12 0.25 0.25 0.16 0.12
y
 Customer distribution

No. of 7 8 9 10
customers
probability 0.25 0.30 0.30 0.15

The probability that a buyer in need of this special cake requires 1, 2 or 3 cakes is as
follows:
Need distribution

Number of cakes to 1 2 3
a customer
probability 0.30 0.40 0.30

Determine by Monte – Carlo simulation, the average number of cakes left per day and sales
lost per day due to lack of supply. Assume that the left over cake is not used next day.
Question 8.
A firm has a single server station with the following data. Mean inter arrival time is 6 minutes
and the mean server time is 5 minutes and inter arrival time and service time have the
following probability .

Inter arrival time probability Service time probability

4 0.12 3 0.10
5 0.21 4 0.25
6 0.24 5 0.40
7 0.20 6 0.15
8 0.13 7 0.10
9 0.10

The firm starts its business at 8.00 AM and proceeds upto 10.00 AM. If the cost of waiting
for a server is Rs. 6 per hour and the customer waiting time costs Rs. 10 per hour. What will
be the total cost due to waiting by a customer and the cost of server, using Monte Carlo
simulation?

CHAPTER – 12
PROJECT SCHEDULING AND PERT – CPM
Introduction
The project may be defined as the collection of interrelated activities (or tasks), which must
be completed in a specified time according to a specified sequence and require resources,
such as personnel, money, materials, facilities etc.
The growing complexities of today’s projects have demanded more systematic and more
effective planning techniques with the objective of optimizing the efficiency of executing the
project.

 Efficiency here implies effecting the utmost reduction in the time required to
complete the project while accounting for the economic feasibility of using available
resources.
There are two analytical techniques for planning, scheduling and controlling the project:

 Project Evaluation and Review Technique (PERT).

 Critical Path Method (CPM).

History of evolution of PERT.
Pert is an acronym for “project Evaluation and review technique”. This was created as a
means to accelerate the development of the Polaris Ballistic missile. In USA the defence
department developed a nuclear missile to be launched from beneath the ocean’s surface by a
mobile submarine which would be an effective deterrent against aggression by an enemy.
This paved the way to plan how to design, develop and plan the different stages in the
production of a missile and how quickly this task could be completed. A planning and
scheduling technique named PERT gave answer to these questions.
In any new venture, uncertainties are bound to creep in. PERT incorporated these
uncertainties. There are certain statistical aspects built into the model. It has proved to be
useful, in planning and scheduling large projects consisting of numerous activities whose
completion times are uncertain and are independent of one another. PERT is an event
oriented technique. By ‘event’ we mean reaching a certain stage of completion of the project.
Another technique, Critical Path Method, abbreviated as CPM has emerged simultaneously. It
is also a network technique but it is concerned with obtaining the trade off, between cost and
completion for large projects.
CPM is not concerned with uncertain job times as in PERT. Pert is useful in research and
development projects, whereas CPM is mostly used in construction projects, or in situations
already handled, so that the details like the normal completion time, crash duration and cost
of crashing are already known.

Definition of PERT

Pert is an event oriented network giving emphasis on important stages of completion

of task.

Definition of CPM

CPM is a deterministic network to establish a trade-off for optimum balancing between

scheduled time and cost of the project.

PERT CPM
Planning scheduling and controlling
PERT

 Total project duration is regarded as a random variable and therefore associated

probabilities are calculated so as to characterize it.

 It had to cope with associated uncertainties

 It is an event oriented network.

 Normally used for projects involving non-repetitive activities. In which time estimates
are certain.

 Helps in pipelining critical areas in a project.

CPM

 Basically deterministic network.

 Suitable for establishing a trade-off for optimum balancing between schedule time
and cost of the project.

 Normally used for projects involving activities of repetitive nature.

Project control
Project control refers to re-evaluating actual progress against the plan. If significant
differences are observed then re-scheduling must be done to update and revise the
uncompleted part of the project.
Network Components
Events

 An event represents a point in time that signifies the completion of some activity and
the beginning of new one.

 Commonly represented by circles (nodes) in the network diagram.

 They do not consume time and resources.

 Identified by numbers.
Activities
Represent project operations or tasks to be conducted.
 Represented by an arrow.
 They consume time and resources.
 Identified by the numbers of their starting event and ending event.
 Normally assigned alphabets.
 ACTIVITY A
3 4

START EVENT COMPLETION EVENT

1
B F

2
Network construction

A network is defined as graphic representation with a flow of some type in its branches. It is
overall combination of events and activities represented by nodes and branches.

Rules for constructing network diagram

 Each activity is represented by one and only one arrow in the network. This means
that no single activity can be represented twice in a network.

 No two activities can be identified by the same head and tail event. This means that
there should not be loops in the network.

 Time follows from left to right. All the arrows point in one direction. Arrows pointing
in opposite direction must be avoided.

 Arrows should not cross each other.

 Every node has atleast one activity preceding to it and atleast one activity following it.
Except for the nodes at the very beginning and at the very end of the network.

 To ensure the correct precedence relationship in the network diagram, the following
questions must be answered as every activity is added to the network:

o What activities must be completed immediately before this activity can start.

o What activity must follow this activity?

 o What activity must occur concurrently with this activity?

Example:

A company is interested in preparing a budget. The details of the activities and the
departments involved are given below in the table.

The project of preparation of a production budget

Job identification alternate Job description department

A 1–2 Forecasting sales sales
B 2–4 Pricing sales Sales
C 2–3 Preparing Engineering
production schedule
D 3–4 Costing the Costing
production
E 4-5 Preparation of President
budget

Prepare the network diagram for the preparation of budget.

Solution:

The project of budgeting can be displayed in a network as given below:

2
E
3 4
D 4 1 B

A 5

From the arrow diagram, we infer that activity A is the first job. Job B and C start only
after A is over. A is called the predecessor of B and C and B and C are called as
successors of A.

Dummy activities

There is a need for dummy activity when the project contains groups of two or more jobs
which have common concurrent predecessors. The time and resources consumed by dummy
activity is zero.

C
1 2

E 3
3
B
 D

Consider an example where B and C are parallel concurrent activities and activity E is
dependent on both of them. That is, E can start only after B and C are completed. Thus, to
complete the network, a dummy activity D is put, which shows that C is also predecessor of
E alongwith B. the dummy activity is represented by dashed arrow.

Solved problem 1

Construct the arrow diagram comprising activities A, B, C … and L such that the following
relationships are satisfied:

 A, B and C the first activities of the project, can start simultaneously.

 A and B precede D

 B precedes E, F and H.

 F and C precede G

 E and H precede I and J

 C, D, F and J precede K.

 K precedes L.

 I, G and L are the terminal activities of the project.

1
8
4
C
7
2
2
4

The dummy activities D1 and D2 are used (dotted lines) to establish correct precedence
relationships. D3 is used to identify activities E and H with unique end events. The
events of the project are numbered such that their ascending order indicates the
direction of the progress in the project.

Critical path calculation

The application of PERT / CPM should ultimately yield a schedule specifying the start and
completion time of each activity. The arrow diagram is the first step towards achieving that
goal. The start and completion timings are calculated directly on the arrow diagrams using
simple arithmetic. The end result is to classify the activities as critical or non critical.

Critical activity

An activity is said to be critical if a delay in the start of the course makes a delay in the
completion time of the entire project.

Non – critical activity

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

 A non-critical activity is said to have a slack or float time.

Determination of critical path.

A critical path defines a chain of critical activities that connect the start and end events of the
network diagram.

 Critical path identifies all the critical activities of the project.

The critical path calculations are done in two phases. The first phase is called forward pass.

Forward Pass.

In this all calculations begin from the start node and move to the end node. At each node a
number is computed representing the earliest occurrence time of the corresponding event.
These numbers are shown in squares.

 We note the number of heads joining the event.

 Take the maximum earliest timings through these heads.

Let ESi = Earliest start time of all the activities emanating from event i.

DiJ = Duration of the activity (i, j)

And ESj = Earliest start time of all the activities emanating from eventj.

Then, for forward pass calculation ESj = max {ESi + Dij) for all defined (i, j) activities

Backward pass
The second phase is called the backward pass. It begins calculations from the end node and
moves to the start node. The time computed at each node is shown in a triangle Δ near the
end point, which represents the latest occurrence time of the corresponding event. In
backward pass, we see the number of tails and take the minimum value through these tails.

 The time computed represents the latest occurrence time of the corresponding event.

 The backward pass starts from the “end” event.

 The objective of the backward pass is to calculate LCi, the latest completion time for
all the activities coming into event i.

Let LCi = Latest completion time for event i,

Latest completion time.

Let LCi = Latest completion time for event i,

LCj = Latest completion time for event j,

Then for any node,

LCi = min (LCJ – Dij) for all defined activities.

An activity (i, j) lies on the critical path if it satisfies the following conditions:

 ESi = LCi

 ESJ = LCJ

 ESJ – ESi = LCJ - LCi = Dij

These conditions actually indicate that there is no float or slack time between the earliest
and latest start of the activity. Thus, the activity must be critical.

Solved problem 2

Consider a network which stands from node 1 and terminate at node 6. The time required to
perform each activity is indicated on the arrows. Define the critical path and shortest time to
complete the project.
 6
1

9 5

19 / 19

Solution:

Let us start with the forward pass with ES1 = 0

Since there is only one incoming activity (activity 1, 2) to event 2 with D12 = 3,

ES2 = ES1 + D12 = 0 + 3 = 3

Now, for end 3, there is only one activity (2, 3) to event 3 with D23 = 3

Thus, ES3 = ES2 + D23 = 3 + 3 = 6

For ES4, there are two activities approaching to event 4, (activity 3, 4 and activity 2, 4
with D24 = 2 and D34 = 0

Thus, ES4 = Max{ES2 + D24; ES3 + D34}

= Max {(3 + 2); (6 + 0)} = 6

Similarly ES5 = 13 and ES6 = 19

This completes the forward pass.

In the backward pass we have

LC6 = 19 = ES6

LC5 = LC6 – D56 = 19 – 6 = 13

LC4 = min {(LC5 – D45); (LC6 – D46)} = min{(13 – 7); (19 – 5)} = min {6; 14} = 6

Similarly LC3 = 6; LC2 = 3 and LC1 = 0

Therefore activities (1, 2), (2, 3), (3, 4), (4, 5), (5, 6) are critical and (2, 4), (4, 6) and (3, 6)
are noncritical.
Thus the activities (1, 2), (2, 3), (3, 4), (4, 5), and (5, 6) define the critical path which is
the shortest possible time to complete the project i.e., 19 days.

Determination of floats

Before showing how floats are determined, it is necessary to define two new items that are
associated with each activity. These are as follows:

 Latest start time.

LSiJ = LCJ – DiJ

 Earliest completion time

ECiJ = ESi + DiJ

There are two important types of floats namely:

 Total float (TF)

 Free float (FF)

Total float. TFiJ for activity (i, j) is the difference between the maximum time available to
perform the activity and its duration.

TFiJ = (LCJ – ESi) – DiJ = LCJ – ESi – DiJ = LCJ – ECiJ = LSiJ - ESi

Free float. The free float is defined by assuming that all the activities start as early as
possible. Free float FFiJ for activity (i, j) is the excess of available time over its duration.

FFiJ = ESJ – ESi – DiJ

For critical activity total float is zero. The free float must also be zero when total float is zero.
However, converse is not true, that is, a noncritical activity may have zero free floats.

Let us consider the solved problem 2 before the critical path calculations. The floats for
non critical activities can be summerised as depicted in the table below:

Activity Duration Earliest latest Total Free

(i, j) Dij Start Completio Start Completio float floats
ESi n ECij LSij n LCj TFij FFij
(1, 2) 3 0 3 0 3 0* 0
(2, 3) 3 3 6 3 6 0* 0
(2, 4) 2 3 5 4 6 1 1
(3, 4) 0 6 6 6 6 0* 0
(3, 5) 3 6 9 10 13 4 4
(3, 6) 2 6 8 17 19 11 11
(4, 5) 7 6 13 6 13 0* 0
(4, 6) 5 6 11 14 19 8 8
(5, 6) 6 13 19 13 19 0* 0
Note: Total float TFij = LSij - ESi

Free float FFij = ESj – ESi - Dij

*
= critical activity

Solved problem 3

A project consists of a series of tasks A, B, C, D, E, F, G, H, I with the following

relationships. (W < X, Y means X and Y cannot start until W is completed, X, Y< W means
W cannot start until both X and Y are completed). With this notation construct the network
diagram having the following constraints: A<D, E; B, D< F; C<G; B, G< H; F, G<I.

Also find the minimum time of completion of the project, the critical path, and the total floats
of each task, when the time (in days) of completion of each task is as follows:

Task: A B C D E F G H I

Time: 23 8 20 16 24 18 19 4 10

Solution:

6
H
2 (4 G
(1
5
(0) C
(0)
(

ES1 = 0, ES2 = 20, ES3 = 23, ES4 = 39, ES5 = 39, ES6 = 57, ES7 = 67

The activity table is as given below:

Activity Duration Earliest latest Total Free

(I, j) Dij Start Completio Start Completio float floats
ESi n ECij LSij n LCj TFij FFij
(1, 2) 20 0 20 18 38 18 0
(1, 3) 23 0 23 0 23 0* 0
(1, 4) 8 0 8 31 39 31 31
(2, 5) 19 20 39 38 57 18 0
(3, 4) 16 23 39 23 39 0* 0
(3, 7) 24 23 47 43 67 20 20
(4, 5) 0 39 39 57 57 10 0
(4, 6) 18 39 57 39 57 0* 0
 (5, 6) 0 39 39 57 57 18 18
(5, 7) 4 39 43 63 67 24 24
(6, 7) 10 57 67 57 67 0* 0

The critical path 1 – 3 – 4 – 6 – 7

The duration to complete the project = 67 days

Project Management – PERT

Pert was developed for the purpose of solving problems in aerospace industries. Particularly
in research and development programmes. These programmes are subject to frequent changes
and as such the times taken to complete various activities are not certain, they are changing
and not standard. This element of uncertainty is being specifically taken into account by
PERT. It assumes that the activities and their network configuration have been well defined,
but it allows for uncertainties in activity times. Thus the activity time becomes a random
variable.

Probability and cost consideration in project scheduling.

The analysis in CPM does not take into account the cases where time estimates for the
different activities are probabilistic. It also does not consider explicitly the cost of schedules.
Here in PERT we will consider both probability and cost aspects in project scheduling.

Probability considerations are incorporated in project scheduling by assuming that the time
estimates for each activity is based on three different time estimates.

If we ask an engineer, or a foreman or a worker to give a time estimate to complete a

particular task he will at once give the most probable time required to perform the activity.
This time estimate is the most likely time estimate.

But he is also asked to give two other time estimates. One of these is pessimistic time
estimate. This is the best guess of the maximum time that would be required to perform an
activity under the most adverse circumstances like:

(a) Supply of materials not in time

(b) Non – cooperation from the workers
(c) The transportation arrangements not being effective etc.

Thus the pessimistic time estimate is the longest time the activity would require.

On the other hand, if everything goes on exceptionally well or under the best possible
conditions, the time taken to complete an activity may be less than the most likely time
estimate. This time estimate is the smallest time estimate known as the optimistic time
estimate.

Thus, the three time estimates for a given activity are as follows:
Let A = The optimistic time, which will be required if the execution of the project goes
extremely well.

B = The pessimistic time, which will be required if everything goes bad.

M = The most likely time, which will be required if execution is normal.

The most likely time estimate ‘M’ may not coincide with the mid point (A + B) / 2 of A and
B.

Then, expected duration of each activity D can be obtained as the mean of (A + B) / 2 and
2M.

Therefore D = [[(A + B) / 2] + 2M] / 3 = (A+B +4M)/6

We can use this estimate to study the single estimate D in the critical path calculation. These
formulas are derived from beta distribution.

The variance of each activity denoted by V is defined by:

Variance V = {(B – A)/6}2

For calculation of critical path, the same formulas as in CPM are used. And duration can be
obtained. However, after calculation of critical path, the variance of the duration is to be
calculated.

V = ΣVk

Where, k defines the activities along the critical path.

The probability distribution of times for completing an event can be approximated by normal
distribution due to central limit theorem.

Let

E(T) = The calculated duration of the project

S (T) = the scheduled time (specified by analyst

Then Z = [S(T) – E(T)/√Vk]

From normal distribution curve table we can see the probability for calculated Z. such
probability will represent the chance that the succeeding events will occur within the
calculated duration.

Solved problem 4

A project is represented by the network shown below and has the following data:
Task A B C D E F G H I
Optimistic time 5 18 26 16 15 6 7 7 3
Pessimistic time 10 22 40 20 25 12 12 9 5
Most likely time 8 20 33 18 20 9 10 8 4

B (8)
0/0
7
1 3
A(23)
67 / 67 F (18)
I (10)

D (16)
E (24)

Determine the following:

 Expected task time and their variance.

 Earliest and latest expected times to reach each event.

 The critical path.

 The probability of an event occurring at the proposed completion data if the original
contract time of completing the project is 41.5 weeks.

 The duration of the project that will have 95% chances of being completed.

Solution:

Using the formula we can calculate expected activity times and variances in the following
table:

Duration D = (A +B + 4M) / 6

Variance V = [(B – A) / 6]2

Table 12.4: Durations and variances for each activity

Activity a b m D V
1–2 5 10 8 7.833 0.694
 1–3 18 22 20 20.00 0.444
1–4 26 40 33 33.00 5.444
2–5 16 20 18 18.00 0.444
2–6 15 25 20 20.00 2.777
3–6 6 12 9 9.00 1.000
4–7 7 12 10 9.555 0.694
5–7 7 9 8 8.00 0.111
6-7 3 5 4 4.00 0.111

Forward pass:

E1 = 0; E2 = 7.833; E3 = 20; E4 = 33; E5 = 25.833; E6 = 29; E7 = 42.555

7
20 29/3
/ 8.55
D 29. 5 7.8
E
55 33
5 3/1
H B 6.5
55

Backward pass:

L7 = 42.555; L6 = 38.555; L5 = 34.555; L4 = 33; L3 = 29.555; L2 = 16.555; L1 = 0

 The critical path is shown with thick line in the figure. The critical path is 1 – 4 – 7
and the completion time is 42.555 weeks.
 The last event will occur only after 42.555 weeks. For this we require only the
duration of critical activities.

Expected length of critical path = 33 + 9.555 = 42.555 weeks

Variance of critical path length = 6.138

Z = (Scheduled time – Estimated time) / standard deviation

= (ST – ET) / V1/2

 = (41.5 – 42.555) / 6.1381/2

= - 0.426

P(-0.426) = 0.3354

Thus, the probability that the project can be completed in less than or equal to 41.5 weeks is
33.46%. In other words, probability that the project will get delayed beyond 41.5 weeks is
66.54%.

 Given that P (Z) = 0.95 thus Z = 1.645

Then 1.645 = (ST – 42.555) / 6.1381/2

ST = 46.631 weeks.

Solved problem 5.

The following table lists the jobs of a network with their time estimates.

Jobs (I – J) Duration (days)

Optimistic Most likely Pessimistic
1-2 3 6 15
1-6 2 5 14
2-3 6 12 30
2-4 2 5 8
3-5 5 11 17
4-5 3 6 15
6-7 3 9 27
5-8 1 4 7
7-8 4 19 28

(a) Draw the network

(b) Calculate the length and variance of the critical path
(c) What is the approximate probability that the jobs on the critical path will be
completed by the due date of 42 days?
(d) What due date has about 90% chance of being met?

Solution

Before proceeding to draw the project network, let us calculate the expected time of
each activity (D) and variances V of the expected time of each activity.

Duration D = (A +B + 4M) / 6

Variance V = [(B – A) / 6]2

Jobs (I – J) Duration (days)

Optimistic (A) Most likely Pessimistic Duration Variance
 (M) (B) (D) (V)
1-2 3 6 15 7 4
1-6 2 5 14 6 4
2-3 6 12 30 14 16
2-4 2 5 8 5 1
3-5 5 11 17 11 4
4-5 3 6 15 7 4
6-7 3 9 27 4 16
5-8 1 4 7 4 1
7-8 4 19 28 18 16

(a) Project network

3 25.
3 833
33
0
0

critical activity

(b) There are three paths:

1 – 2 – 3 – 5 – 8 = 36 days

1 – 2 – 4 – 5 – 8 = 23 days

1–6–7–8 = 35 days

1 – 2 – 3 – 5 – 8 is the longest path and hence the critical path.

Expected length of the path is 36 days and the variance is = 4 + 16 + 4 + 1 = 25 days

Standard deviation = (25)1/2 = 5 days

(c) Due date = 42 days (ST)

Expected duration = 36 days (ET)

 Standard deviation = 5 days

Therefore, Z = (ST – ET) / SD = (42 – 36) / 5 = 1.2

The probability as per normal distribution curve = 0.8849 = 88.49%

(d) For 90% probability, as per normal distribution table, Z = 1.28167

thus, 1.28167 = (ST – 36) / 5

ST = 1.28167 x 5 + 36 = 42.4 days

CPM Model

For projects considered uncertain, the PERT model was developed and for projects, which
are comparatively risk free, the CPM model was developed. Both the approaches start with
the development of the network and a focal point on the critical path. The PERT approach is
‘probabilistic’, while the CPM approach is ‘deterministic’. This does not, however, mean that
in CPM analysis we work with single time estimates. Actually the main focus of CPM
analysis is on variations of activity times as a consequence of changes in resource
assignments. These variations are planned plus related to resource assignments as well as are
not caused by random factors outside the control of management as in the case of PERT
analysis. The major focus of CPM analysis is on time cost relationships and it seeks a project
schedule that minimizes either the total cost or the total duration or the both.

Assumptions.

The usual assumptions underlying CPM analysis are:

1. The costs associated with a project can be divided into two main components:
 Direct costs. Direct costs are incurred on direct materials and direct labour.
 Indirect costs. Indirect costs consist of overhead items like indirect supplies, rent,
insurance, managerial services, etc.
2. Activities of the project can be expedited by crashing, which involves employing
more resources.
3. Crashing reduces time but enhances direct costs because of the factors like overtime
payments, extra payments, and wastages. The relationship between time and direct
activity cost can be reasonably approximated by a downward sloping straight line.

Direct cost of activity

 4. Indirect costs associated with the project increase linearly with project duration.

Indirect cost of project

5.
6.

Procedure

CPM analysis seeks to examine the consequences of crashing on total cost (direct and indirect
costs). Since the behaviour of indirect project cost is well defined, the bulk of CPM analysis
is concerned with the relationship between total direct cost and project duration. The
procedure used in this respect is generally as follows:

 Step 1. Obtain the critical path in the normal network. Determine the project duration
and direct cost.
 Step 2. Examine the cost time slope of activities on the critical path obtained and
crash the activity which has the least slope.

Slope = (crash cost – normal cost) / (normal time – crash time)

 Step 3. Construct the new critical path after crashing as per step 2. Determine project
duration and cost.
 Step 4. Repeat step 2 and 3 till activities on the critical path (which may change
every time) are crashed.

Example

The table below depicts the activities, durations, and direct activity costs of a project.
The indirect cost is Rs. 2,000 per week. Calculate the minimum total cost of the project
and minimum duration of the project.

Normal and crash time and costs

Activity Time in weeks Cost Cost to

normal crash normal crash expedite
 per week
(slope)
1–2 8 4 3,000 6,000 750
1–3 5 3 4,000 8,000 2,000
2–4 9 6 4,000 5,500 500
3–5 7 5 2,000 3,200 600
2-5 5 1 8,000 12,000 1000
4–6 3 2.50 10,000 11,200 2400
5–6 6 2 4,000 6,800 700
6–7 10 7 6,000 8,700 900
5-7 9 5 4,200 9,000 1,200

G
11
3
5 4 14
7
2
36 / 5

7 8

4
0
/ 1

critical activity

(a) There are five paths:

 1 – 2 – 4 – 6 – 7 = 30 weeks
 1 – 2 – 5 – 6 – 7 = 29 weeks
 1 – 2 – 5 – 7 = 22 weeks
 1 – 3 – 5 – 6 – 7 = 28 weeks
 1 – 3 – 5 – 7 = 21 weeks

The critical path in all normal network is 1 – 2 – 4 – 6 – 7. The project duration is 30 weeks
and the total direct cost is Rs. 45,200 the total cost is Rs. 105,200 (direct + indirect cost).

Examining the time cost slope of activities on the critical path, we find that activity (2, 4) has
the lowest slope. In other words, the cost to expedite per week is lowest for activity (2, 4).
Hence, activity (2 – 4) is crashed. Figure below depicts the project network after crashing of
activity (2, 4)
 6
7
1 6
4
A
Cr du Normal
A Time
No c ra
Cra d

3 4

9
6

critical activity 10

crashed activity

as per figure the critical path is 1 – 2 – 5 – 6 – 7 with a length of 29 weeks, and the direct
cost is 46,700. The total cost is Rs. 46,700 + Rs. 58000 = Rs. 104,700

Examining the time cost slope of activities on the new critical path 1 – 2 – 5 – 6 - 7, we find
that activity (5 - 6) has the lowest slope. In other words, the cost to expedite per week is
lowest for activity (5 - 6). Hence, activity (5 – 6) is crashed. Figure below depicts the project
network after crashing of activity (5 - 6).

2
3 7 8
6
5
1 5 0/0
9
5
3 7

3 4

6
6
10

critical activity

crashed activity

as per figure the critical path is 1 – 2 – 4 – 6 – 7 with a length of 27 weeks, and the direct
cost is 49,500. The total cost is Rs. 49,500 + Rs. 54000 = Rs. 103,500.
Examining the time cost slope of non – crashed activities on the new critical path 1 – 2 – 4 –
6 - 7, we find that activity (1 - 2) has the lowest slope. In other words, the cost to expedite per
week is lowest for activity (1 - 2). Hence, activity (1 – 2) is crashed. Figure below depicts the
project network after crashing of activity (1 - 2).

2
2 7 8
6
5
1 5 0/0
9
5
3 7

3 6

6
4
10

critical activity

crashed activity

as per figure the critical path is 1 – 3 – 5 – 6 – 7 with a length of 24 weeks, and the direct
cost is 52,500. The total cost is Rs. 52,500 + Rs. 48000 = Rs. 100,500.

Examining the time cost slope of non – crashed activities on the new critical path 1 – 3 – 5 –
6 - 7, we find that activity (3 - 5) has the lowest slope. In other words, the cost to expedite per
week is lowest for activity (3 - 5). Hence, activity (3 – 5) is crashed. Figure below depicts the
project network after crashing of activity (3 - 5).

2
2 7 8
2
5
1 5 0/0
9
5
3 7

3 6

6
4
10
critical activity

crashed activity

as per figure the critical path is 1 – 2 – 4 – 6 – 7 with a length of 23 weeks, and the direct
cost is 53,700. The total cost is Rs. 53,700 + Rs. 46000 = Rs. 99,700.

Examining the time cost slope of non – crashed activities on the new critical path 1 – 2 – 4 –
6 - 7, we find that activity (6 - 7) has the lowest slope. In other words, the cost to expedite per
week is lowest for activity (6 - 7). Hence, activity (6 – 7) is crashed. Figure below depicts the
project network after crashing of activity (6 - 7).

2
24 7 4
/ 2
5
1 5 0/0
9
5
3 7

3 6

6
4
10

critical activity

crashed activity

as per figure the critical path is 1 – 2 – 4 – 6 – 7 with a length of 20 weeks, and the direct
cost is 56,400. The total cost is Rs. 56,400 + Rs. 40000 = Rs. 96,400.

Examining the non – crashed activities on the new critical path 1 – 2 – 4 – 6 - 7, we find that
only activity (4 - 6) is the non-crashed. Hence, activity (4 – 6) is crashed. Figure below
depicts the project network after crashing of activity (4 - 6).

2
23 7 4
/ 5
0 2

5 1
9
5
3 5

3 6
 6
4
7

critical activity

crashed activity

since all the activities on the critical path (1 – 2 – 4 – 6 – 7) have been crashed, there is no
possibility of time reduction. The minimum time duration is 19.5 weeks and the total direct
cost is Rs. 57,600. The total cost is Rs. 57,600 + Rs. 39000 = Rs. 96,600.

Thus, the minimum cost for the project will be Rs. 96,400 with project duration of 20
weeks.

The minimum duration of the project can be 19.5 weeks with total cost of Rs. 96,600.

Review questions.

Question 1. A project has 10 activities. The following table shows the information about the
activities:

Activity Preceding activity Duration in weeks

A - 6
B - 3
C A 5
D A 4
E A 3
F C 3
G D 5
H B, D, E 5
I H 2
J I, G, F 3

 Draw the network.

 Find the project duration.
 Identify the CPM.
 Prepare the schedule.

Ans: duration = 20 weeks;

Critical path: A – D – H – I – J

Question 2

A small project consisting of eight activities has the following characteristics:

 Time estimates in weeks

Activity Preceding Most optimistic Most likely Most

activity time (A) time (M) pessimistic time
(B)
A NONE 2 4 12
B NONE 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

 Draw the PERT network for the project.

 Determine the critical path.

 If a 30 weeks dead line is imposed, what is the probability that the project will be
finished within the time limit?

 If the project manager wants to be 99% sure that the project is completed on the
scheduled date, how many weeks before that date should he start the project work?

Ans; critical path: A – D – G – H

Probability to finish within 30 weeks = 65.91%

The project should start 34.7 weeks before to make sure 99%.

Question 3.

The following information is known for a project. Draw the network and find the
critical path. Capital letters denote activities and the numbers in the bracket denote
activity time.

This time must be completed before this can start

A (30) C

B(7) D

B G

B K

C (10) D

C G
 D (14) E

E (10) F

F (7) H

F I

F L

G (21) I

G L

H (7) J (15)

I (12) J

K (30) L (15)

Question 4.

Draw network diagram for the following activities whose predecessors are given
below:

job A B C D E F G H I J K
Immediate - A B C B E D, F E H G, I J
predecessors

Question 5.

A PERT network is having the following activities with their time estimates given
below:

Activity Optimistic Most likely Pessimistic

0-1 2 3.5 8
0-2 3 3.75 6
0-3 1 2.5 7
1-2 3 7.5 9
1-5 4 5.5 10
2-4 2 5 8
3-4 2 2.75 5
3-5 3 6 9
4-5 2 5 8
(a) Construct a network for this data and find the expected completion time of the
project.
(b) Find the probability of completing the project 3 days ahead of the expected
schedule.
Question 6.

The managing director of M / S Raj Kumari Manufacturing Company had an opportunity to

deal with a project. The project X is required to be completed within 8 months at a cost of $
1,00,000. The managing director called production manager Mr. LN Gupta and the finance
manager Mr. Srinivas Patil and asked them to determine the profitability of the project on 8
month basis. The time and cost under normal conditions and crashing conditions has been
given below:

Activity Normal Crashing

Time Cost Time Cost
1–2 2 8,000 1 13,000
1–3 3 7,000 1 19,000
1–4 6 11,000 5 13,500
2–3 4 6,000 3 10,000
2–4 2 9,000 1 10,000
2–5 7 8,500 6 11,500
3–5 4 10,500 3 16,000
4-5 3 5000 2 7,000

Main issues involved are:

 What suggestions should be given to managing director for undertaking the project?
 After having long discussions with the general manager, who is the second in
command in the organisation, the management somehow feels that during
implementation of the project, it will not be possible to crash the activity 2 – 4 from
its normal time. If so, how will the production manager and finance manager respond?
 CHAPTER – 13

GAME THEORY
Introduction

Many conflicting situations are found in everyday life, in economic, social, political, military
battles, advertising and marketing campaign by competing business firms. In these situations
two or more individuals have to take decisions that involve conflicting interests. A basic
feature in many of these situations is that the final result depends primarily on the
combination of strategies selected by the persons involved called adversaries or opponent
players. The approach to such competitive problems was developed by Von Neumann who
named it “game theory”. It is a mathematical theory that is useful in handling such problems.

Many research papers have appeared on two person zero-sum game. These games involve
only two competitors or players. They are called zero-sum because when one player wins, the
other one looses so that the sum of their net winning is zero.

An objective of the game theory is to develop rational criteria for selecting a strategy. This is
done under the assumption that both players are rational and that each will uncompromisingly
attempt to do as well as possible, relative to his opponent. Game theory assumes that both
players are actively trying to improve their own welfare in opposition to that of the opponent.

Competitive Situation

Competitive situation occur when two or more parties with conflicting interests operate.

 Marketing different brands of a commodity. Two or more brands of an item trying to

capture the market by adopting various methods.

 Campaigning for election. Two or more candidates who contest an election try to
capture more votes by adopting various methods.

 Fighting military battles. Two forces fighting a war try to gain supremacy over one
another by adopting various courses of action.

We consider each of the above situations to be a competitive game where the parties
(players) adopt a course of action (play the game).

Competitive game

Characteristics of a competitive game.

The following properties hold good for a competitive game.

  The number of players (competitors) is finite.

 Each of the players / competitors has a finite list of possible courses of action (moves)
known as strategy. The number of strategies need not be the same for each
competitor.

 The game is said to be played when each player adopts one of his course of action.
The play of the game results when each of the competitors chooses a single course of
action from the list of strategies available to him. The choices are assumed to be made
simultaneously so that no competitor knows his opponent’s choice until he is already
committed to his own.

 Each time the game is played, the corresponding combination of courses of action
leads to a transaction (payment) to each player. The payment is called pay-off (gain).
The pay-off may be monetary (money) or some such benefit as increased sales etc.

 The players do not communicate to each other.

 The players know the rules of the game before starting.

N-person game.

 A game in, which n players participate, is called n-person game.

 A game, in which two persons participate, is called two-person game.

Zero-sum game.

 If a game is such that whenever it is played the sum of the gains (pay-off) of the
players is zero, it is called zero-sum game.

 A zero-sum game which has two players is called two-person zero-sum game. It is
called “rectangular game”.

 In a two-person zero-sum game, the gain of the one player is equal to the loss of the
other.

Two-person zero-sum game (rectangular game)

A two-person zero-sum game is one in which:

 Two players participate.

 The gain of one player is the loss of other.

Consider a two-person zero-sum game with the players A and B. let A1, A2 ……Am be the m
courses of action for player A. let B1, B2 …..Bn be the n courses of action for player B. Let a ij
(I = 1, 2, ….m; j = 1, 2, ….n) be the pay off (gain) of player A when he plays the respective
course of action, Ai and player B plays the course of action Bj. then, the matrix depicted
below is the pay off (gain) matrix of player A.

Pay off matrix of player A

PLAYER B

B1 B2 B3 BN
2 20 / 20

PLAYER A

This is a m x n (read as m by n) game. Here a ij is A’s gain and B’s loss. Therefore (-a ij) is the
gain of B. to obtain the pay off matrix of B, write (-a ij) in place of aij in the above matrix and
then write the transpose of the matrix.

Courses of action:

 Player ‘A’ – A1, A2, A3, ------- AM

 Player ‘B’ – B1, B2, B3, ------- BN

 Pay-off of player ‘A’ – aiJ (i = 1,2----M; j= 1,2,-----N)

Strategy.

In a game, the strategy of a player is pre-determined. The player uses this strategy to select a
course of action during the game. The strategy of a player may be:

 Pure strategy. While playing a game, pure strategy of a player is his pre-
decision to adopt a specified course of action, irrespective of the strategy of
the opponent.

 Mixed strategy. While playing a game, mixed strategy is his pre-decision to

adopt his course of action according to certain pre-assigned probabilities.

Thus, if player A decides to adopt courses of action A1 and A2 with perspective probabilities
0.4 and 0.6, it is mixed strategy.

Example 1
(2-finger morra game)
Two persons A and B play a game. They should simultaneously raise their hand and exhibit
either one finger or two fingers. If both of them show one finger or both show two fingers, A
should pay Rupees 10 to B. On the other hand, if one player shows one finger and the other
player shows two fingers, B should pay Rupees 5 to A.

 Make the pay-off matrix

 Identify the strategies of player A.

Solution:

Here, pay-off matrix of player A is

PLAYER B

B1 B2

PLAYER A A1 -10 5

A2 5 -10

If the player A decides to show one finger (A1), his strategy is pure strategy. On the
other hand, if A decides to play A1 with probability of 0.5 and A2 with probability 0.5,
his strategy is mixed strategy. This means, if he is repeatedly playing, he should play
both A1 and A2 randomly almost equal number of times.

Maximin – minimax principle

Player A and player B are to play a game without knowing the other player’s strategy.
However, player A would like to maximize his profit and player B would like to minimize his
loss. Also each player expects his opponent to be calculative.

Suppose player A plays A1

Then, his gain would be a11, a12, ….. a1n accordingly B’s choice would be B1, B2 ….Bn. let α1
= min {a11, a12 ……a1n}.

Then α1 is the minimum gain of A when he plays A1 (α1 is the minimum pay off in the
first row.)

Similarly, if A plays A2, his minimum gain is α2, the least pay off in the second row.

You will find corresponding to A’s play A1, A2 …..Am, the minimum gains are the rows
minimums α1, α2 ….. αm.

Suppose A chooses the course of action where αi is maximum. Then

Maximin. Maximum of the row minimum in the pay-off matrix is called maximin.
Thus, α = maxi { minJ (aiJ)}

This is the maximum gain for player ‘A’ out of the minimum gains of the rows (αi )

Similarly, when player B plays, he would minimize his maximum losses. Thus,

Minimax. The minimum of the column maximums in the pay-off matrix is called minimax.

Thus, β = minJ {maxi (aiJ)}

Saddle point.

In a two person zero sum game, if the maximin and minimax are equal, the game has a saddle
point.

Saddle point is the position where the maximin and minimax coincide.

α=β=V

If the maximin occurs in the Rth row and if minimax occurs in Sth column, the position (R, S)
is the saddle point. Here, V = ars is called the value of the game.

The value of the game is the expected gain of player A when both the players adopt optimal
strategy.

 A game is said to be fair if its value is zero.

 If a game has saddle point, and if (R, S) is the saddle point, suggested solution to both
the players is pure strategy. For player A the suggested solution is AR. For player B
the suggested solution is BS.

 If the game does not has saddle point, the suggested solution is mixed strategy.

Solution to a game with saddle point

Consider a two person zero sum game with players A and B. Let A1, A2, A3, --------AM be the
courses of action for player A. Let B1, B2, ---------- BN be the courses of action for player B.
The algorithm to find saddle point is as follows:

 The minimum pay-off in each row of the pay-off matrix is encircled.

 The maximum pay-off in each column is written within a box.

 If any pay-off is circled as well as boxed, that pay-off is the value of the game.

 The corresponding position is the saddle point.

Let (R, S) be the saddle point. Then the suggested pure strategy for player A is AR. And for
player B is BS. The value of the game is ars..
Note: however, if none of the pay offs is circled or boxed, the game does not have a
saddle point. Hence, the suggested solution for the player is mixed strategy

Example 2
(2-finger Morra game)

Two persons A and B play a game. They should simultaneously raise their hand and exhibit
either one finger or two fingers. If both of them show one finger or both show two fingers, A
should pay Rupees 10 to B. On the other hand, if one player shows one finger and the other
player shows two fingers, B should pay Rupees 5 to A.

 Make the pay-off matrix

 Verify whether it has saddle point. If so, write down the solution for the game.

Solution:

The pay – off matrix of player A is as follows:

Player B

B1(one finger) B2 (two finger)

A1 (one finger) 7 4

Player A

A2 (two finger) 5 2

 The minimum pay off in each row is circled

 The maximum pay off in each column is boxed.
 Since none of the payoff is circled / boxed, the game does not have the saddle point.
Hence the solution for the game is mixed strategy for both the players.

Solved problem 2

Two persons A and B, without showing each other, place a coin each on the table. If the coin
happens to be of the same denomination, player A will take both of them. If they happen to
be of different denominations, player B will take both of them. Suppose player A has a few
one rupee coins and two rupee coins, and player B has one, two and five rupee coins:

 Make the pay-off matrix of A.

 Verify whether it has saddle point. If so, write down the solution for the game.
  What happens to the game if both the players play only with one rupee and two rupee
coins?

Solution:

(i) the pay off matrix of player A is as follows:

Player B

B1(one rupee) B2 (two rupees) B3 (5 rupees)

1 0
A1 (one rupee) 5
Player A
5
A2 (two rupee) 3 9

 The minimum pay off in each row is circled. (here, the minimums repeat)
 The maximum pay off in each column is boxed.
 The pay off ‘-1’ is circled as well as boxed. Therefore, the game has a saddle
point. It is the position (1, 3).
 The solution of the game is:
o Strategy for player A is A1
o Strategy for player B is B3
o Value of the game is v = -1 rupees.
(ii) if both players play only with one rupee and two rupee coins, the payoff
matrix of player A is

Player B

B1(one rupee) B2 (two rupees)

A1 (one rupee) 5
2.
Player A
5
A2 (two rupee) 6
6
Here, the game does not have a saddle point. Therefore, the suggested solution is
adoption of mixed strategy for the players.

Solved problem 3
A labour union of a firm is negotiating a new 5-year settlement regarding payments with the
management. The options the union has are A1: ‘Aggressive bargaining’, A2: ‘Bargaining
with reasoning’, A3: ‘Conciliatory approach’. The likely mode of response from the
management are: B1: ‘Aggressive bargaining’, B2: ‘Bargaining with reasoning’, B3:
‘Legalistic approach’ and B4: ‘Conciliatory approach’. The gains to the union in each case are
as follows:

Management

B1 B2 B3 B4

A1 20 15 12 35

Union A2 25 14 8 10

A3 -5 4 11 0

 What strategy would you suggest for the two sides?

 What is the value of the game?

Solution:

Management

B1 B2 B3 B4

A1 20 7 1 4

Union

A2 7 14 2 10

A3 4 4 11 0

 The minimum payoff in each row is circled.

 The maximum pay off in each column is boxed.
 The value 12 is circled as well as boxed. So the game has a saddle point. It is the
position (1, 3).

Therefore, the solution to the game is as follows:

 The strategy for the union is A1; “Aggressive bargaining”.

 The strategy for the management is B3; “legalistic approach”.
 The value of the game is v = 12.
Solved problem 4

Solve the game

Player B

B1 B2 B3

Player A A1 6 12 7

A2 7 9 8

Is the game fair?

Solution:

Player B

B1 B2 B3

A1 5 2 7

Player A

A2 0 9 1

 The minimum pay off in each row is circled.

 The maximum pay off in each column is boxed
 The value 7 is circled as well as boxed. So, the game has a saddle point. It is
position (2, 1)

Therefore, the solution to the game is:

 The strategy for the player A is A2 (second course of action).

 The strategy for the opponent player B is B1 (first course of action).
 The value of the game is v = 7 and
 Game is not fair.

Solved problem 5

A two-person zero-sum game, has the following pay-off matrix. Solve the game.

8 -1

3 0

0 -2

-7 -4

Is the game fair?

 Solution: player B

B1 B2

A1 9 5

Player A A2 3 5
A3 0 3

A4 5 -4

 The minimum pay off in each row is circled

 The maximum payoff in each column is boxed
 The value 0 is circled as well as boxed. So, the game has a saddle point at
position (2, 2).

Therefore, the solution to the game is:

 Strategy for the player A is A2 (second course of action)

 Strategy for opponent player B is B2 (second course of action).
 The value of the game is v = 0
 Since the value of the game is 0, the game is fair.

Dominance

In a rectangular game, the pay off matrix of player A is pay off in one specific row (r th row)
exceeding the corresponding pay off in another specific row (s th row). This means that
whatever course of action is adopted by player B, for A, the course of action A r yields greater
than the course of action As. therefore, Ar is a better strategy than As irrespective of B’s
strategy. Hence, you can say that Ar, dominates As.

Alternatively, if each pay-off in a specific column (pth column), is less than the corresponding
pay –off in another specific column (qth column) it means strategy Bp offers minor loss than
strategy Bq irrespective of A’s strategy. Hence, you can say that B p dominates Bq. therefore,
you can say that:

 In the pay-off matrix, if each pay-off in the R th row is greater than (or equal to) the
corresponding pay-off in the Sth row, AR dominates AS.

 In the pay off matrix, if each pay-off in the P th column is less than (or equal to) the
corresponding pay-off in the Qth column, BP dominates BQ.

Sometimes combination of two or more courses of action may dominate another course of
action. Whenever a course of action (say A S or BQ) is dominated by others, than that
course of action (As or Bq) can be deleted from the pay-off matrix. Such deletion will not
affect the choice of the solution. But it reduces the order of the pay off matrix. Successive
reduction of the order using dominance property helps in solving games more easily.
  Such deletion of course of action reduces the order of pay-off matrix.

Solving games using dominance

Consider a rectangular game with players A and B. Let A 1, A2, A3, --------AM be the courses
of action for player A. Let B1, B2, ---------- BN be the courses of action for player B.

Suppose the game has a saddle point. Use the dominance property in sequence to delete the
courses of action of A as well B till the pair comprising the saddle point remains alone. The
procedure to arrive at saddle point is as follows:

 In the pay-off matrix, if each pay-off in the R th row is greater than (or equal to) the
corresponding pay-off in the Sth row, AR dominates AS. And so AS is deleted.

 If each pay-off in the Pth column is less than (or equal to) the corresponding pay-off in
the Qth column, BP dominates BQ. And so BQ is deleted.

 The above steps are repeated in succession until the saddle point is reached. And
hence the solution is written down.

Note: sometimes, a convex combination of two or more courses of action may

dominate another course of action.

Solved problem 6

Solve the following game using the dominance property.

B1 B2 B3 B4

A1 20 15 12 35

A2 25 14 8 10

A3 -5 4 11 0

Solution:

In the pay off matrix, each pay off in the first row exceeds the corresponding pay off in
the third row. Therefore A1 dominates A3. So, A3 is deleted. Hence, the reduced matrix
is as follows:

B1 B2 B3 B4

A1 20 15 12 35

A2 25 14 8 10

Here, each pay off in the third column is less than the corresponding pay off in the first
column. Therefore, B3 dominates B1. Similarly, B3 dominates B2. Also B3 dominates B4.
Thus, the matrix reduces to
 B3

A1 12

A2 8

Here, since 12 > 8, A1 dominates A2. And so, finally the matrix reduces to:

B3

A1 12

Thus, (1, 3) is the saddle point and the solution to the game is as follows:

 Strategy for A is A1
 Strategy for B is B3
 Value of the game is v = 12

Solved problem 7

Solve the following zero-sum game and find its value.

Company Y

B1 B2 B3 B4

A 6 -2 4 1

Company X B 6 1 12 3

C -3 -2 -2 6

D 2 -3 7 7

Solution:

In the pay off matrix, each pay off in the second column is less than (or equal to) the
corresponding pay off in the third column. And so, the course of action B 2 dominates B3.
similarly B2 dominates B4. After deleting B3 and B4, the reduced matrix is as follows:

Company Y

B1 B2

A 6 -2

Company X B 6 1

C -3 -2
 D 2 -3

Here, pay off in the second row is greater than (or equal to) the corresponding pay off
in the first, third as well as fourth rows. Therefore B dominates A, C and D. after
deleting A, C and, the reduced matrix is as follows:

P Q

B 6 1

Here 1 < 6. Therefore Q dominates P. after deleting P; the reduced matrix is as follows:

B 1

Therefore solution to the game is:

 Strategy for company X is B.

 Strategy for company Y is Q.
 The value of the game is v = 1

Solved problem 8

In a two person zero sum game, the pay off matrix of A is as given below.

PLAYER B

B1 B2 B3

PLAYER A A1 4 7 0

A2 -1 3 6

Write down the pay-off matrix of player B.

Solution:

The pay off matrix of player B is as follows:

PLAYER A

A1 A2

B1 -4 1

PLAYER B B2 -7 -3

B3 0 -6
Review questions
Question 1

Solve the following rectangular game

-2 2 0 5 -8

3 2 1 2 2

-1 3 0 -4 2

0 -5 -4 3 -2

Answer:

The value of the game is v = 1. Both players to use pure strategy. A’s strategy is A 2
(second course of action). B’s strategy is B3 (third course of action).

Question 2

In a rectangular game, pay-off matrix of player A is

PLAYER B

B1 B2

PLAYER A A1 5 7

A2 4 0

 Solve the game

 Write down the pay-off matrix of B and then solve the game.

Answer:

Value of the game is v = 5. Both players to use pure strategy. A’s strategy is A 1 (first
course of action). While, B’s strategy is B1 (first course of action).

Question 3.

Solve the game with the pay off matrix for A as given in the table below.

Player B

B1 B2 B3 B4 B5

A1 3 -1 4 6 7
Player A A2 -1 8 2 4 12

A3 16 8 6 14 12

A4 1 11 -1 2 1

Question 4.

Solve the game with the pay off matrix for A as given in the table below.

Player B

B1 B2 B3

A1 -2 0 6

Player A A2 2 1 2

A3 4 -2 -4

Question 5.

Solve the game with the pay off matrix for A as given in the table below. How the game
should be played?

Player B

B1 B2

A1 -2 2

Player A A2 4 -3

Question 6.

Consider the game with the pay off matrix for A as given in the table below. How the
game should be played and what is the value of the game?

Player B

B1 B2

A1 -2 5

Player A A2 7 4

Question 7.

Solve the game by using dominance property with the pay off matrix for A as given in
the table below.

Player B
 B1 B2 B3

A1 2 -2 4

Player A A2 -1 4 2

A3 2 1 2

Question 8.

Solve the game by using dominance property with the pay off matrix for A as given in
the table below.

Player B

B1 B2 B3

A1 -2 4 2

Player A A2 2 -4 4

A3 6 8 -6

Question 9.

Solve the game with the pay off matrix for A as given in the table below.

Player Y

Y1 Y2 Y3 Y4 Y5

X1 6 3 -1 0 -3

Player X X2 3 2 -4 2 1

Question 10.

In a game of matching coins with two players, suppose:

 A wins one rupee when there are two heads

 Wins nothing when there are two tails
 Loses ½ rupee when there are one tail and one head.

Determine the pay off matrix, the best strategies for each player and the value of the
game.

Question 11.

In a small town, there are two discount stores ABC and XYZ. They are the only stores
that handle the festival goods. The total number of customers is equally divided between
the two because the price and quality of goods sold are equal. Both stores have good
reputations in the community, and they render equally good customer services. Assume
that a gain of customer by ABC is a loss to XYZ and vice versa.

Both stores plan to run annual pre Christmas sale during the first week of December.
Sales are advertised through the local news paper, radio and television media. With the
aid of advertising the pay off for ABC store is constructed and given below.

XYZ store

News paper radio Television

News paper 30 40 -80

ABC radio 0 15 -20

Television 90 20 50

Find optimal strategies for both stores and the value of the game.

Question 12.

In a children’s game each player says ‘stone’ or ‘scissor’ or ‘papers’. If one says ‘stone’
and the other ‘scissors’ then former wins a rupee. Similarly ‘scissors’ beats ‘paper’ and
‘paper’ beats ‘stone’ i.e., the player calling the former word wins a rupee. If the two
players name the same name, then there is a tie i.e. there is no pay off. Write down the
pay off matrix of either of the two players. Find the value of the game and hence write
down the optimal strategies of both players.
 CHAPTER 14

SEQUENCING
INTRODUCTION

A series, in which a few jobs or tasks are to be performed following an order, is called
sequencing. In such a situation, the effe3ctiveness measure (time, cost, distance etc.) is a
function of the order or sequence of performing a series of jobs. Problems of sequencing can
be classified into two major groups.

In the first type of problems, we have n jobs to perform each of which requires processing on
some or all ‘m’ different machines. If we analyze the number of sequences it runs to (n!) m
possible sequences and only a few of them are technologically feasible, i.e., those which
satisfy the constraints on the order in which each task has to be processed through ‘m’
machines.

In the second type of problem, we have a situation with a number of machines and a series of
jobs to perform. Once a job is finished, we have to take a decision on the next job to be
started.

Practically both types of problems seem to be intrinsically difficult and now we know
solutions only for some special cases of the first type of problem. For the second type of
problems, it appears that a few empirical rules have been obtained to arrive at the solution
and mathematical theory to be explored.

PROCESSING ‘N’ JOBS THROUGH TWO MACHINES

The sequencing problem with n jobs through two machines can be solve easily. Solution
procedure has been developed by SM Johnson. The problem can be stated as follows:

 Only two machines are involved. A and B

 Each job is processed in the order A – B
 The exact or expected processing time A1, A2 ……..An, B1, B2 ……..Bn are known.

A decision has to be arrived to find the minimum elapsed time fom the start of the first job to
the completion of last job. It has been established that the sequence that minimizes the
elapsed time are the same for both machines. The algorithms for solving the problem as per
SM Johnson are as follows:

 Select the smallest processing time occurring in the list A1, A2 ……An, B1, B2…..Bn. if
there is a tie, break the tie arbitrarily.
 If the minimum processing time is Ai, do the ith job first. If it is B j do the jth job last.
This decision is applicable to both machines.
  Having selected a job to be ordered, there are (n – 1) jobs left to be ordered. Apply the
steps 1 and 2 to the reduced set of processing times obtained by deleting the two
machines processing times corresponding to the job that is already assigned.
 Continue in this manner until all jobs have been ordered.

Solved problem 1

There are five jobs, each of which must go through the two machines A and B in the order A
– B. processing times in hours are as given below:

Job Processing time

Machine A Machine B
1 10 4
2 2 12
3 18 14
4 6 16
5 20 8

Determine a sequence for the five jobs that will minimize elapsed time.

Solution:

Applying the algorithm of Johnson, we observe that the smallest processing time is 2
hours for job 2 on machine A. thus, we select the job 2 for scheduling first.this is
displayed in the list.

The job 2 is eliminated from the list and the remaining set of process time is:

Job Processing time

Machine A Machine B
1 10 4
3 18 14
4 6 16
5 20 8

The smallest processing time is 4 hours for the job 1 in machine B. the job 1 is scheduled last
as it pertains to B

2 1
Proceeding in the same way, we have

Job Processing time

Machine A Machine B
3 18 14
4 6* 16
5 20 8

Yielding

2 4 1

The reduced set is

Job Processing time

Machine A Machine B
3 18 14
5 20 8*

yielding

2 4 5 1

And finally the optimal sequence is:

2 4 3 5 1

The calculation of the elapsed time corresponding to the optimal ordering is carried out using
individual processing times given in the question. The working is given in the following
table:

Machine A Machine B Idle time

Job Time in Time out Time in Time out Machine A Machine B
1 0 2 2 14 0 2
2 2 8 14 30 0 0
3 8 26 30 44 0 0
4 26 46 46 54 0 2
5 46 56 56 60 4 2

Thus, the minimum elapsed time is 60 hours. The idle time for machine A is 4 hours and for
machine B is 6 hours.

PROCESSING n JOBS THROUGH THREE MACHINES

There is no general solution available for the general sequencing problem of n jobs through
three machines A, B and C with prescribed order A – B – C for each job and no passing still a
few special cases can be attempted with the extension of the previous sections. Either or both
of the following conditions must be valid.

 The smallest processing time for machine A is at least as great as the largest
processing time for machine B. i.e., min Ai ≥ max Bj
 The smallest processing time for machine C is at least as great as the largest
processing time for machine B. i.e., min Ci ≥ max Bi

The procedure is to replace the problem with an equivalent problem of n jobs through two
machines. Let the two fictitious machines be x and y and we define the corresponding
processing times X1 and Y1 given by

X 1 = A 1 + B1

Y1 = B1 + C1
We follow the method of the previous section to the new problem with prescribed ordering x
and y. the resulting optimal sequence will also be optimal for the original problem.

Solved problem 2

We have five jobs and each of which has to go through the machines A, B and C in the order
A – B – C. processing times are as given below:

job Processing time (minutes)

A B C
1 40 50 80
2 90 60 100
3 80 20 60
4 60 30 70
5 50 40 110

Determine the sequence for the five jobs that will minimize the elapsed time in completion of
all the jobs.

First we check for the conditions to be satisfied as per Johnson’s rule before converting the
problem of three machines to an equivalent two machines.

 We see that the smallest time for A is 40 minutes and the largest time for B is 60
minutes. The condition that min Ai ≥ max Bj is not satisfied. But minimum of C is 60
and maximum of B is 60 and hence the condition min C i ≥ max Bi is satisfied. Hence
we are justified to convert to two machine problem. Then we have

job Processing time (minutes)

Machine X = (A + B) Machine Y = (B + C)
1 90 130
 2 150 160
3 100 80
4 90 100
5 90 150

Proceeding with the algorithm mentioned in the previous example, we get the solution.
Since there are three ties, there are several optimal orderings. They are:

1 4 5 2 3: 1 5 4 2 3: 4 5 1
2 3: 4 1 5 2 3: 5 1 4 2 3:
5 4 1 2 3

Any of these orderings may be use to sequence the jobs through machines A, B and C to find
the elapsed time T. we prepare the following table for any one of the sequencing.

job Machine A Machine B Machine C

Time in Time Idle time Time in Tim Idle time Time in Time Idle
out e out out time
1 0 40 - 40 90 40 – 0 = 90 170 90
40
4 40 100 - 100 130 100 – 90 = 170 240 0
10
5 100 150 - 150 190 150 - 130 240 350 0
= 20
2 150 240 - 240 300 240 – 190 350 450 0
= 50
3 240 320 - 320 340 320 – 300 450 510 0
= 20
510 – 320 = 510 – 340 0
190 = 170
Total idle time 190 Total idle time 310 Total idle time of 90
machine A = of machine B machine C

The elapsed time = 510 minutes

Solved problem 3:

Find an optimal sequence for the following sequencing problem of four jobs and five
machines when passing is not allowed. The processing time is given below:

job Processing time (minutes)

A B C D E
1 7 5 2 3 9
2 6 6 4 5 10
3 5 4 5 6 8
4 8 3 3 2 6
Solution:

The neessary condition to be satisfied is either or both of the following:

(i) min Ai ≥ max (Bi, Ci, Di)

(ii) min Ai ≥ max (Bi, Ci, Di)

Min Ai = 5 min Ei = 6 and max (Bi, Ci, Di) = 6

Since the second condition is satisfied, we can convert the problem of 5 machines to an
equivalent problem of 2 machines X and Y by combining the times for the first four machines
and combining the times for the last four machines respectively. Thus, we have

job Processing time (minutes)

Machine X = (A + B + C + D) Machine Y = (B + C + D + E)
1 17 19
2 21 25
3 20 23
4 16 14

Using the optimal sequence algorithm, an optimal sequence of 1- 3 – 2 – 4 can be obtained.

Now, to calculate the total time of optimal sequence, we take the individual processing time
of all jobs through all the machines as given in table below:

job Processing time (minutes)

A B C D E
1 0-7 7 - 12 12 - 24 14 – 17 17 – 26
2 7 - 12 12 - 16 16 – 21 21 – 27 27 - 35
3 12 - 18 18 - 24 24 – 28 28 - 33 35 - 45
4 18 - 26 26 -29 29 - 32 33 - 35 45 - 51

Thus, we have the minimum lapsed time as 51 hours and the idle time for machines A, B, C,
D, and E are 25, 33, 37, 35 and 18 hours respectively.

Processing two jobs through n machines

In this section we deal with the following situation

 There are K machines, denoted by A, B, C, …, K.

 Only two jobs are to be performed, job 1 and job 2.
 The technological ordering of each of two jobs through K machines is known in
advance. This technological sequence need not be the same for both jobs.
 The exact or expected processing times A1, B1,K1, A2, B2, …..K2 are known.
 The problem is to minimize the elapsed time for the start of the first job and to the
completion of the last job.
Solved Problem 4

Two jobs are to be processed through four machines A, B, C, D with the following
technological ordering:

Job 1: A B C D

Job 2: D B A C

Processing times are given in the following table:

job machines
A B C D
1 20 40 50 10
2 20 50 30 60

Find the optimum minimum elapsed time for both jobs and also the idle time for both jobs.

Solution

There are two ways of solving this problem

 Using Gantt chart

 Using Graphical method
(a) Gantt chart: in this method we draw the times taken for both jobs at different
machinas in the prescribed order. Care should be taken that no two machines are busy
simultaneously. The solution is given below:

l-----l-----l-----l-----l-----l-----l-----l-----l-----l-----l-----l-----l-----l-----l-----l-----l-----l

0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170

A B C D
Job 1

D B A C
Job 2

(b) Graphical method. The first step is to draw two axis, the horizontal axis
representing processing time for job 1 and the vertical axis, processing time for job 2.
The machine times for the two jobs marked to the same scale on the corresponding
axis in the technological order prescribed. The coordinates of any point on the grid
represent a possible state of completion of the two jobs.

Start from the origin and move through a series of segments which are horizontal,
or vertical or diagonal with a slope 1. Movement to the right indicates that the job 1
is proceeding while the job 2 is idle. Movement upward means that the job 2 is
proceeding while job 1 is idle. And movement diagonally means that both jobs are
proceeding simultaneously.

In moving, care should be taken that both jobs cannot be processed simultaneously
on the same machine, or the diagonal movement in the shaded area is forbidden.

Am optimal path or programme is one that minimizes idle time for job 1 indicated
by horizontal movement. Equivalently, an optimal path is one that minimizes the
idle time for job 2 indicated by vertical movement. So we have to choose a path with
as much travel on the diagonal as possible.

An optimal solution is obtained by drawing lines with arrows as shown. The elapsed
time is obtained by adding the idle time for either job to the processing time for that
job.

In the above problem, we see from the figure that the idle time for the chosen path is
4 for job 1 and zero for job 2. Elapsed time is equal to 12 + 4 = 16 + 0
 CHAPTER 15

FORECASTING
FORECASTING

Forecasting is predicting or estimating the future value of a variable. Quantitative

forecasting techniques such as regression methods and smoothing methods will be
discussed in this chapter.

REGRESSION METHODS

Regression methods deal with establishing a mathematical relationship between

independent and dependent variables. The variable that is to be estimated is called
the dependent variable while the variable that helps in estimation is called the
independent variable. Simple regression deals with a linear relationship between one
dependent and one independent variable. Multiple regression deals with one
dependent variable and two or more independent variables.

SIMPLE LINEAR REGRESSION

Consider the linear regression equation Y = a + bX, where Y is dependent variable,

X is the independent variable, a is the vertical axis intercept, and b is the slope of the
straight line. The values of constants a and b are obtained as follows:

a = (ΣX2 ΣY – ΣX ΣXY) / (nΣX2 – (ΣX)2); b = (nΣXY – ΣXΣY) / [nΣX2 – (ΣX)2]

where X is the independent variable data, Y is the dependent variable data, n is the
total number of observations. Using the above values of a and b in the linear
regression equation Y = a + bX, Y can be estimated for a future value of X. this
simple linear regression is also called as “casual linear regression”.

If the data are a time series, the independent variable is the time period and the
dependent variableis the variable to be forecasted, such as sales. Such a relationship,
known as time series linear regression or trend line, is expressed as Y = a +bT,
where T is the time period.

COFFICIENT OF CORRELATION AND DETERMINATION

The coefficient of correlation r is a relative measure of the relationship between the

dependent variable Y and the independent variable X. it is computed as follows:

r = [nΣXY – ΣXΣY] / [[nΣX2 – (ΣX)2][ΣY2 – (ΣY)2]]0.5

where -1 ≤ r ≤ +1. A perfect relationship exists when r = ± 1.

The coefficient of determination is the square of the coefficient of correlation. This is

a measure of the proportion of variance in the dependent variable Y that can be
explained by the independent variable X. Note that 0 ≤ r 2 ≤1. For example, if r2 = 0.9,
the independent variable explains 90% of the variation in the dependent variabl.

STANDARD ERROR OF ESTIMATE

The standard error (deviation) of estimate (forecast) is a measure of the dispersion

or scatter of the past data around the trend (regression) line.

It is given by the following equation:

SE =[(∑Y2 - a∑Y - b∑XY)/(n – 2)]0.5

SPECIAL CASE: LOGARITHMIC (EXPONENTIAL) MODELS

The logarithmic or exponential models are used for data depicting growth with no
sign of leveling off. Nonlinear function of the exponential models can be transformed
into linear finctions as follows:

(a) W = ABX
Taking common logarithms on both sides, Log W = Log A + (Log B)X
(b) W = ea + bX
Taking natural algorithm on both sides, ln W = (a + bX) ln e = a + bX
Substituting Y = ln W, Y = a +bX
After determining a and b in the usual manner, the parameters A and B can be
found through antilogarithms.

SMOOTHING METHODS
Consider the multiple regression equation Y = b 0 + b1X1 + ….. +bXn, where Y is the
dependent variable. X1, X2, ….Xn are the independent variables and b 0, b1, ….bn are
the coefficients.

SMOOTHING METHODS

One way of removing random variations in a time series is to smooth it. Two
commonly used smoothing methods are moving average and exponential smoothing.

MOVING AVERAGE

The moving average method finds the forecast for a particular time period by
averaging the data values of the most recent n periods in the time series.
Mathematically, the moving average is calculated as follows:

MA(n) = ∑(latest n data values) / n

For example, MA(3) would imply a three period moving average.

WEIGHTED MOVING AVERAGES

The weighted moving averages method uses different weights for the most recent n
data values.

EXPONENTIAL SMOOTHING
 Exponential smoothing finds the new forecast by taking the weighted average of the
immediately preceding actual and forecast values. The basic exponential model is as
follows:

Ft = αAt – 1 + (1 – α)Ft – 1 or Ft = Ft – 1 + α(At – 1 – Ft – 1)

Where, Ft = forecast value of period t
Ft – 1 = forecast value of period t – 1
At – 1 = actual value of period t – 1
Α = smoothing constant (0 ≤ α ≤ 1)
EXPONENTIAL SMOOTHING WITH TREND
(TREND – ADJUSTED EXPONENTIAL SMOOTHING)
Trend – adjusted exponential smoothing makes use of both an exponentially
smoothed component (St) and a trend component (Tt). the new forecast is obtained
through the following equations:

St = αAt + (1 – α)(St – 1 + Tt – 1)
Tt = β(St – St – 1) + (1 – β)Tt – 1
Ft +1 = St + Tt Ft +k = St + kTt
Where St is the smoothed forecast value in period t, T t is the trend estimate value in
period t, At is the actual value in period t, α is the exponential smoothing constant (0 ≤ α
≤ 1), β is the trend smoothing constant (0 ≤ β ≤ 1), F t + 1 is the forecast value in period t
+1, and Ft +k is the forecast value in period t +k.

Since the calculation of the first estimate of trend T2 needs the first two actual values,
the computational procedure starts with t = 2 (and not with t = 1) as follows:

S2 = A2
T2 = A2 – A1
S3 = αA3 + (1 – α)(S2 +T2)
T3 = β(S3 – S2) + (1 – β)T2

St = αAt + (1 – α)(St – 1 + Tt – 1)
Tt = β(St – St – 1) + (1 – β)Tt – 1

FORECAST ACCURACY

The three most commonly used measures of forecast accuracy are the mean absolute
deviation (MAD), the sum of squared errors (SSE), and the mean squared error (MSE).
These are defined as follows:

MAD = ∑| At - Ft| / n
SSE = ∑(At – Ft)2
MSE = ∑(At – Ft)2 / n
Where At is the actual value of time period t and F t is the forecast value at time t. the
difference between actual value and forecast value (At – Ft) is called forecast error.

FORECASTING TIME SERIES WITH MULTIPLICATIVE MODEL

The underlying data patterns of a time series consist of trend, cyclical, seasonal, and
random components.
A trend exhibits the long range upwards or downward direction of the time series data.
A cycle shows a long term wave like data pattern that repeats itself. Seasonality is a
short repetitive data pattern. The random component accounts for the irregular
changes due to many different factors that cannot be explained.

The multiplicative model is given below:

Yt = Tt x Ct x St x Rt

Where, Yt = actual time series value

Tt = trend component
Ct = cyclical component
St = seasonal component
Rt = random component

PROCEDURE TO DEVELOP TIME SERIES FORECAST FOR DATA WITH

SEASONALITY

1. Compute seasonal indices to measure the degree of differences among seasons as

follows:
(a) Isolate the trend cycle of the data by calculating moving averages, where
number of periods equals number of seasons. The resulting moving averages,
with no seasonal and almost no random variations, are as follows:

MAt = Tt x Ct

(b) Calculate the ratio of the time series to the moving average as given below:

Yt / MAt = Tt x Ct x St x Rt / (Tt x Ct) = Stx Rt

(c) Calculate the average of the above ratios for each season, which is a measure
of seasonal differences with almost no random variations.
(d) Adjust the above average ratios to obtain seasonal indices (SI), where the
average seasonal index is 1. For example, multiply each seasonal index by 4
(sum of the unadjusted seasonal indices).

Note: if the time series contains no discernible cyclical component, use regression
analysis instead of moving average in step (a).

The multiplicative model without the cyclical component is as follows:

Y t = Tt x S t x R t

Since linear regression gives the trend of the data, calculate the ratio of the time
series to the regression line as follows:

Yt / yt = Tt x St x Rt / Tt = St x Rt

Continue with step (c) and (d).

 2. Use the seasonal indices to deseasonalize the data. That is, remove the effect of
season from the time series, by dividing each time series value by the
corresponding seasonal index.
3. Obtain a linear regression equation of the form Tt = a + bt for the deseasonalized
data. This will represent the trend component of the multiplicative model.
4. Use the above trend equation to predict future trend values.
5. Multiply the above trend values by the corresponding seasonal indices to obtain
the forecasts as follows:

Ft = Tt x SIt

Note: in the multiplicative model, estimating trend and seasonality components pose no
problem. However, estimating the cyclical component is a judgment call, since it is
based on the economic or industry activity level.

CHAPTER 16

DECISION THEORY

INTRODUCTION

The one activity which distinguishes management from the other functions in an organization
is decision-making. We can study decision-making from many aspects; in many
circumstances decisions are taken for reasons which can only be studied by the use of
psychology or sociology, but we can also consider what is called "rational decision-making",
and this makes use of quantitative measures to a large degree.

The purpose of quantitative methods of presenting and analysing data is to provide the
decision-maker with a basis on which to make his or her decisions. This of course tends to
make decision-making more difficult. The reason is that, in the absence of good information
as to the possible consequences of a decision, decision-making becomes a matter of choosing
between those alternative courses of action which are before the decision-maker. A person
who makes decisions in a confident way may gain a good reputation as a decision maker;
most people tend to prefer someone who is resolute and confident to someone who hesitates
and tries to weigh up the alternatives. Having good leadership qualities in this manner does
not, unfortunately, mean that the leader always makes correct decisions. If there is an
extended interval between taking a decision and finding out what its ultimate consequences
are, then the leader may escape censure for a bad decision. It may take someone studying the
history of the events to make a clear judgment, by which time the person who made the
decision will have moved on to other things!

When good quantitative information is available and presented in a meaningful way, it

becomes a great deal easier to see what the consequences of different courses of action are.
They can then be compared and a decision taken on the judgement as to which course of
action is likely to have the best outcome. The more that is known about the possible
outcomes, the more we have to try to bring the measures of success to a common basis,
typically money. Attempts to do so when we are considering what is best for a community
rather than just a commercial organisation have led to the "cost-benefit analysis".
It is convenient to classify decision situations under three headings:

 Decisions under certainty;

 Decisions under risk; and
 Decisions under uncertainty.

We will look first at decisions under certainty.

A. DECISION-MAKING UNDER CERTAINTY

Here the outcome to a choice of course of action (a strategy) can be evaluated and the same
result will always occur for that choice of strategy. As the outcomes are determined, we call
this a deterministic situation. When all possible strategies can be listed and their outcomes
evaluated, the rational decision-maker simply chooses the strategy which gives the best value
to the outcome. The fact that there are many real-life situations in which people could take
decisions in this manner but do not, either because they do not assess all strategies, or simply
by perversity, shows the need to undertake behavioural studies.

There are three problems for the decision-maker and his or her quantitative advisors:

 To find an appropriate measure by which to evaluate outcomes

 To calculate the value of the measure for an outcome
 To identify the best outcome from a large number of outcomes.

Outcome Measures

In most organisational situations there are two popular measures – "How much?" and "How
long?". As most commercial decisions resolve to a matter of money or "the bottom line", the
first of these is the dominant measure. In non-commercial organisations other measures may
apply.

As the concept of organisational decision-making originated in wartime, we can look at the

choices facing a military commander when deciding what to do next. One such problem
concerned the supply of war materials from the USA to the UK and Europe across the
Atlantic Ocean during World War II. The possible measures were to:

 Maximise the total tonnage of goods delivered

 Minimise the time taken to deliver goods
 Minimise the loss of shipping.

They are conflicting objectives to some degree, and the "best" answer to the problem is not
easy to see. In the event the answers adopted were related to the shipping available. The very
fast ships, such as the liner Queen Mary, were used to minimise time, relying on their great
speed to evade attack. Most other ships were operated in convoys, to minimize shipping
losses, and to satisfy to some degree the total tonnage requirement.

Valuation of Measures

This is where the various mathematical techniques available come into their own. The
characteristic of such methods is that there is a "right" answer to the question posed.
Optimal Outcome

Some situations involve a very large number of choices. Even to list them may be a daunting
task, and to evaluate them all may be lengthy and costly. Fortunately some excellent methods
and algorithms have been developed, which can ensure that we consider all alternatives and
identify the "best" outcome. Critical path analysis is of this nature, providing a method of
systemising our knowledge so that the effect of choices of order of doing things can be seen.
Another very powerful method looks at the assignment of resources between a number of
competing uses. This is "linear programming". We do not study linear programming (LP) in
this course but you should know that it exists. It is used in areas such as:

 Investment decisions
 Product mix situations
 Product formulation
 Production scheduling
 Delivery scheduling.

B. DEFINITIONS

Before considering some criteria for making decisions under conditions of uncertainty and
risk, we need to establish the following definitions:

(a) Decision Alternatives

The term "decision alternatives" refers to the list of choices available to a decision maker. For
example, an investor may have to decide whether to purchase shares or bonds or both, which
would represent three decision alternatives.

(b) States of Nature

The future environmental conditions that influence the desirability of decision alternatives are
referred to as states of nature. For example, whether the stock market is rising, falling or
steady would be likely to influence the desirability of investing in shares or bonds, and so
would represent three possible states of nature facing an investor. For any given decision
problem, the states of nature should be exhaustive and mutually exclusive.

(c) Uncertainty

Under conditions of uncertainty, we assume that the possible states of nature are known to the
decision-maker, but their probabilities are unknown. Most decisions in business and
management are of this type, and probabilities often have to be estimated from past
experience, or the states of nature have to be regarded as equally likely.

(d) Risk

Under conditions of risk, we assume that the possible states of nature are known and that the
probabilities that they will occur are also known.

(e) Payoff Table

A payoff table has columns that represent the possible states of nature and rows that represent
the decision alternatives. Each cell of the table gives the payoff (usually a revenue, profit or
loss) associated with the given decision alternative and state of nature.

(f) Decision Tree

A decision tree is a diagrammatic representation of a decision problem, which shows the

decision alternatives, the possible states of nature, the state of nature probabilities (if known)
and the payoffs. It is an alternative way of presenting the information in a payoff table.

C. DECISION-MAKING UNDER UNCERTAINTY

We will investigate the problem of decision-making under uncertainty by examining a simple

business problem. Suppose a UK computer manufacturer produces laptop computers for the
Italian market in its UK factory, but recognises that future changes in the sterling-euro
exchange rate may make the laptops uncompetitive in the Italian market. To reduce this risk,
the company is considering the possibility of shifting production to a new factory in France.
The decision alternatives are:

1. UK production: continuing the production of the laptops for the Italian market in the UK.

2. French production: shifting production of the laptops for the Italian market to the new
factory in France.

The outcomes of the decisions are assumed to depend on what happens to the value of the
pound relative to the euro over the next five years. If the pound should weaken against the
euro, the profits from shifting production to France are likely to be higher, but if the pound
should strengthen against the euro, the profits from shifting production to France are likely to
be lower. So in this example, the states of nature are:

1. Weak £: over the next five years, the pound weakens against the euro.

2. Stable £: over the next five years, the pound is unchanged against the euro.

3. Strong £: over the next five years, the pound strengthens against the euro.

We assume for now that the probabilities of the pound being weak, stable or strong are
unknown. However, the likely payoffs under each state of nature are known and these are
shown in the payoff table (Table 16.1).

Table 16.1: Payoff table

Weak £ Stable £ Strong £

5
Decision UK production 7 12

Alternatives
French
production 10 9 8
Payoffs are profits in £ million

We consider four possible decision-making criteria: the Laplace, maximin, maximax

and minimax regret criteria.

(a) Laplace criterion: choose the decision alternative with the highest mean payoff
According to the Laplace criterion, the various states of nature are assumed to be equally
likely, and the decision-maker is required to choose the decision alternative with the highest
mean payoff. In our example: if UK production is maintained, the mean payout is £8 million;
but if production shifts to France, the mean payoff is £9 million. In this case, the Laplace
criterion requires the decision-maker to choose the "French production" decision alternative –
i.e. to shift production of the laptops for the Italian market to the new French factory.

(b) Maximin criterion: choose the decision alternative with the highest minimum payoff

The maximin criterion requires the decision-maker to choose the decision alternative with the
highest minimum payoff. In other words, the decision-maker is pessimistic and wishes to
avoid the lowest possible return. In our example, if production continues in the UK, the
minimum possible payoff is £5 million, but if production shifts to France, the minimum
payoff is £8 million. Therefore in this case the maximin criterion also requires the decision-
maker to choose the "French production" decision alternative – i.e. to shift production of the
laptops for the Italian market to the new French factory.

(c) Maximax criterion: choose the decision alternative with the highest maximum payoff

The maximax criterion requires the decision-maker to choose the decision alternative with the
highest maximum payoff. In other words, the decision-maker is optimistic and aims for the
highest possible return. In our example, if production continues in the UK, the maximum
possible payoff is £12 million, but if production shifts to France, the maximum payoff is £10
million. Therefore in this case the maximax criterion requires the decision-maker to choose
the "UK production" decision alternative – i.e. to maintain production of the laptops for the
Italian market in the UK factory.

(d) Minimax regret criterion: choose the decision alternative with the lowest maximum
regret

To apply the minimax regret criterion, we first have to convert the payoff table to a "regret
table". Each cell of the regret table contains a "regret" defined as the difference between the
payoff in that cell and the maximum payoff for that particular state of nature. The minimax
regret criterion then requires the decision-maker to choose the decision alternative with the
lowest maximum regret. For our example, the regret table is shown in Table 16.2.

From the regret table, we can see that if production continues in the UK, the maximum
possible regret is £5 million, but if production shifts to France, the maximum regret is £4
million. Therefore in this case the minimax regret criterion requires the decision maker to
choose the "French production" decision alternative – i.e. to shift production of the laptops
for the Italian market to the new French factory.
 Table 16.2: Regret Table

States of Nature

Weak £ Stable £ Strong £

Maximum payoff for each state of 10 9 12

nature

Decision UK production 5 2 0
Alternatives
French 0 0 4
production

Payoffs are profits in £ million

D. DECISION-MAKING UNDER RISK

Under conditions of risk, we assume that decisions are taken with full knowledge of the
probabilities associated with the possible states of nature. Let us then suppose that we
know that:

 The probability of a relatively weak pound over the next five years is 0.2
 The probability of a relatively stable pound over the next five years is 0.5
  the probability of a relatively strong pound over the next five years is 0.3.

As the three states of nature are exhaustive and mutually exclusive, these three
probabilities must sum to one (i.e. it is a certainty that the pound will be relatively weak,
stable or strong). These probabilities enable us to calculate the expected monetary
values of each decision alternative.
Expected Monetary Value
Given a set of n possible monetary values (say, x1, x2, ... xn) and n probabilities
associated with each monetary value (say, p1, p2, ... pn), the expected monetary
value (EMV) is calculated as the sum of the products of the monetary values and
their associated probabilities (i.e. p1x p2x2 ... pnxn). So we can write:

EMV p1x1 p2x2 ... pnxn pi xi for i 1, 2, ... n

Where xi represents the ith monetary value and pi represents the probability
associated with the ith monetary value.
With regard to decision-making problems, the x's are the payoffs resulting from each
decision alternative for a given state of nature and the p's are the state of nature
probabilities. Returning to our example, the payoff table with the probabilities
included and the calculated EMVs is shown in Table 16.3.

Table 16.3: Payoff table with probabilities and EMVs

States of Nature

Weak £ Stable £ Strong £ EMVs

Probabilities 0.1 0.5 0.4

Decision UK 5 7 12 8.8
Alternatives production

French 10 9 8 8.7
production

Payoffs are profits in £ million

The EMV of the decision to maintain production in the UK (denoted EMV UK) is
calculated as:

EMVUK (0.1 5) (0.5 7) (0.4 12) 8.8 (i.e. £8.8 million)

While the EMV of the decision to shift production to France (denoted EMV F) is
calculated as:
EMVF (0.1 10) (0.5 9) (0.4 8) 8.7 (i.e. £8.7 million).
So in this case, where a strong pound has a higher probability (0.4) than a weak
pound (0.1), the decision alternative with the higher EMV is "UK production" – i.e.
continue producing the laptops for the Italian market in the UK.

However, note that if the probabilities were changed so that a strong pound was less probable
than a weak pound, the EMVs would also change. Suppose that:

 The probability of a relatively weak pound over the next five years is 0.4
 The probability of a relatively stable pound over the next five years is 0.5
 The probability of a relatively strong pound over the next five years is 0.1.

Now the EMV of the decision to maintain production in the UK (denoted EMVUK) would be
calculated as:
EMVUK (0.4 5) (0.5 7) (0.1 12) 6.7 (i.e. £6.7 million)

and the EMV of the decision to shift production to France (denoted EMV F) would be
calculated as:
EMVF (0.4 10) (0.5 9) (0.1 8) 9.3 (i.e. £9.3 million).

In this case, the decision alternative with the higher EMV would be "French production" –
i.e. shift production of the laptops for the Italian market to France.

Decision Trees

A decision problem illustrated in a payoff table (such as that in Table 16.3) may also be
illustrated by means of a decision tree, which shows diagrammatically the decision
alternatives, the states of nature, the payoffs and the payoff probabilities. The decision tree
for the decision problem facing our UK computer manufacturer is shown in Fig. 16.1.

Figure 15.1: Decision tree for the UK computer manufacturer

In a decision tree, the decisions and states of nature are depicted by branches and the
points at which decisions are taken or at which states of nature occur are depicted by
nodes. There are three types of nodes:
 Decision nodes that depict decision points (represented by square boxes)
 Chance nodes that depict points at which states of nature occur (represented by
circles)
 Payoff nodes that depict the final payoffs (represented by vertical lines).
The computer manufacturer has to decide between UK production and French production
of the laptops for the Italian market. The point at which this decision is taken is depicted
by decision node 1 (i.e. the square box labelled 1 on the left-hand side of the decision
tree). If UK production is chosen, the payoff depends on the state of nature that occurs.
This is depicted by the chance node 2 (i.e. the circle labelled 2 in the diagram). Similarly,
if French production is chosen, the payoff also depends on the state of nature that occurs.
This is depicted by the chance node 3 (i.e. the circle labelled 3 in the diagram). Finally,
the vertical lines at the right-hand side of the decision tree depict the final payoffs
associated with each decision and each state of nature. The state of nature probabilities
are also shown on the appropriate branches of the decision tree. The decision tree
provides the same information as the payoff table and can be used in the same way to
compute the EMVs for each decision alternative.

E. COMPLEX DECISIONS
Some decisions are taken as part of a set, and if there is a mix of decisions and ranges of
outcomes, it can become difficult to see what is happening. The problem of tendering for
contracts is of this type.
As an example, consider the case of a small builder who is invited to tender for a contract.
The builder knows that there is a second tender coming up, more valuable than the first,
but that unless he puts in a tender for the first he will not be invited to tender for the
second. If he wins the first contract he must complete it and cannot bid for the second. He
knows that his costs on the contracts will depend on the weather; also that his chance of
winning the contract will depend on his bid.
He makes estimates as follows:
Contract A
________________________________________________________________________
Bid level high: 0.2 chance of winning the contract
Bid level medium: 0.6 chance of winning the contract
Bid level low: 0.9 chance of winning the contract
Bid level high: Profit: £60,000
Bid level medium: Profit: £40,000
Bid level low: Profit: £25,000
Effect of weather:
Very bad weather Reduce profits by £15,000
Poor weather Reduce profits by £5,000
Probability of very bad weather = 0.2
Poor weather = 0.4
________________________________________________________________________
 Contract B
________________________________________________________________________
Bid level high: 0.3 chance of winning the contract
Bid level medium: 0.7 chance of winning the contract
Bid level low: 0.9 chance of winning the contract
Bid level high: Profit: £100,000
Bid level medium: Profit: £80,000
Bid level low: Profit: £60,000
Effect of weather:
Very bad weather Reduce profits by £20,000
Poor weather Reduce profits by £10,000
Probability of very bad weather = 0.2
Poor weather = 0.4
It costs £5,000 to prepare a bid, which has been allowed for in all the profit figures. What
should he do? We can draw a decision tree to review the choices and possible outcomes.
A partial decision tree for this problem is shown in Figure 16.2.
Figure 16.2: Completed tree
Figure 16.3: Completed tree showing values of outcomes
We can now work through the tree from right to left and, at each chance node, find the
expectation for that node and, for each decision node, choose the strategy which gives the
highest expectation.

Check the tree carefully so that you can see where all the figures come from. Note that
decisions are shown by putting a cross stroke on the branches chosen.

We find that the best decision is to make a high bid for the first contract; if this is not
successful, make a medium bid for the second contract. The expectation of profit with this
choice of strategies is £48,920.

You may notice that the second part of the tree, concerning the second contract, repeated
itself three times. We could have evaluated this as a separate decision tree and entered the
result, which was medium bid each time, in a simpler tree for the first bid. Sometimes
successive decisions interact; the profit on the second contract might have been linked to
what happened in the first, in which case the tree must be analysed as a unit.

Interpreting Expectations
It is important to realise what an expectation, such as the answer to the last problem,
actually means.
When the builder has completed a contract, the profit will be one of the figures at the ends
of the twigs in the tree. The exact figure will depend on the builder's decisions, which
contract the builder gets and the weather. The figure of £48,920 we found is the average
profit the builder would make if it were possible to repeat the contract many times,
following the rule we derived. Of course in real life this is not possible, so why bother?
The answer is that if a company is in a situation where similar decisions have to be taken
frequently, then it is reasonable to take an average view. The company can afford to lose
money on some contracts as long as on average, taken over many contracts, it makes a
profit. The analysis is then valid. In other words, for a large construction company which
bids for many contracts the analysis is suitable. Compare this to the situation of our small
builder. If he is short of cash he may be worried that if he does not get the contract, he
will be unable even to pay the cost of preparing the bid.
Finite model table
 Standard Normal Distribution Table
Z 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09
0.0 0.0000 0.0040 0.0080 0.0120 0.0160 0.0199 0.0239 0.0279 0.0319 0.0359
0.1 0.0398 0.0438 0.0478 0.0517 0.0557 0.0596 0.0636 0.0675 0.0714 0.0753
0.2 0.0793 0.0832 0.0871 0.0910 0.0948 0.0987 0.1026 0.1064 0.1103 0.1141
0.3 0.1179 0.1217 0.1255 0.1293 0.1331 0.1368 0.1406 0.1443 0.1480 0.1517
0.4 0.1554 0.1591 0.1628 0.1664 0.1700 0.1736 0.1772 0.1808 0.1844 0.1879
0.5 0.1915 0.1950 0.1985 0.2019 0.2054 0.2088 0.2123 0.2157 0.2190 0.2224
0.6 0.2257 0.2291 0.2324 0.2357 0.2389 0.2422 0.2454 0.2486 0.2517 0.2549
0.7 0.2580 0.2611 0.2642 0.2673 0.2704 0.2734 0.2764 0.2794 0.2823 0.2852
0.8 0.2881 0.2910 0.2939 0.2967 0.2995 0.3023 0.3051 0.3078 0.3106 0.3133
0.9 0.3159 0.3186 0.3212 0.3238 0.3264 0.3289 0.3315 0.3340 0.3365 0.3389
1.0 0.3413 0.3438 0.3461 0.3485 0.3508 0.3531 0.3554 0.3577 0.3599 0.3621
1.1 0.3643 0.3665 0.3686 0.3708 0.3729 0.3749 0.3770 0.3790 0.3810 0.3830
1.2 0.3849 0.3869 0.3888 0.3907 0.3925 0.3944 0.3962 0.3980 0.3997 0.4015
1.3 0.4032 0.4049 0.4066 0.4082 0.4099 0.4115 0.4131 0.4147 0.4162 0.4177
1.4 0.4192 0.4207 0.4222 0.4236 0.4251 0.4265 0.4279 0.4292 0.4306 0.4319
1.5 0.4332 0.4345 0.4357 0.4370 0.4382 0.4394 0.4406 0.4418 0.4429 0.4441
1.6 0.4452 0.4463 0.4474 0.4484 0.4495 0.4505 0.4515 0.4525 0.4535 0.4545
1.7 0.4554 0.4564 0.4573 0.4582 0.4591 0.4599 0.4608 0.4616 0.4625 0.4633
1.8 0.4641 0.4649 0.4656 0.4664 0.4671 0.4678 0.4686 0.4693 0.4699 0.4706
1.9 0.4713 0.4719 0.4726 0.4732 0.4738 0.4744 0.4750 0.4756 0.4761 0.4767
2.0 0.4772 0.4778 0.4783 0.4788 0.4793 0.4798 0.4803 0.4808 0.4812 0.4817
2.1 0.4821 0.4826 0.4830 0.4834 0.4838 0.4842 0.4846 0.4850 0.4854 0.4857
2.2 0.4861 0.4864 0.4868 0.4871 0.4875 0.4878 0.4881 0.4884 0.4887 0.4890
2.3 0.4893 0.4896 0.4898 0.4901 0.4904 0.4906 0.4909 0.4911 0.4913 0.4916
2.4 0.4918 0.4920 0.4922 0.4925 0.4927 0.4929 0.4931 0.4932 0.4934 0.4936
2.5 0.4938 0.4940 0.4941 0.4943 0.4945 0.4946 0.4948 0.4949 0.4951 0.4952
2.6 0.4953 0.4955 0.4956 0.4957 0.4959 0.4960 0.4961 0.4962 0.4963 0.4964
2.7 0.4965 0.4966 0.4967 0.4968 0.4969 0.4970 0.4971 0.4972 0.4973 0.4974
2.8 0.4974 0.4975 0.4976 0.4977 0.4977 0.4978 0.4979 0.4979 0.4980 0.4981
2.9 0.4981 0.4982 0.4982 0.4983 0.4984 0.4984 0.4985 0.4985 0.4986 0.4986
3.0 0.4987 0.4987 0.4987 0.4988 0.4988 0.4989 0.4989 0.4989 0.4990 0.4990