ADDIS ABABA UNIVERSITY

COLLEGE OF BUSINESS AND ECONOMICS

SCHOOL OF COMMERCE

OPERATIONS RESEARCH
COURSE MODULE

ADDIS ABABA, OCTOBER 2019

Ashenafi A. (PhD Research Fellow) Page 1

 Table of Contents
CHAPTER ONE ............................................................................................................................5
INTRODUCTION TO OPERATIONS RESEARCH ................................................................5
1.0.Introduction ................................................................................................................................5
1.1.Introduction To Operations Research ........................................................................................5
1.2.History Of Operations Research ................................................................................................7
1.3.Stages Of Development Of Operations Research ......................................................................9
1.4.Phases Of An Operations Research Project ............................................................................. 11
1.5.Operations Research Tools And Techniques ........................................................................... 15
1.6.Limitations Of Operations Research ........................................................................................ 16
CHAPTER TWO ......................................................................................................................... 19
LINEAR PROGRAMMING MODELS (LPM) ........................................................................ 19
2.0. Introduction ............................................................................................................................. 19
2.1.Introduction To Linear Programming Models ......................................................................... 19
2.2.Linear Programming Models (LPM) ....................................................................................... 20
2.2.1. Components Of Lp Models ..............................................................................................20
2.2.2. Assumptions Of Lp Models .............................................................................................22
2.3.Formulating LP Models ........................................................................................................... 23
2.4.Linear Programming Applications ........................................................................................... 23
2.5.Solving Linear Programming Models ...................................................................................... 30
2.5.1. Graphical Linear Programming Methods .........................................................................30
2.5.2.The Simplex Method Of Solving Lp Problems.................................................................43
2.6.The Post Optimality Analysis .................................................................................................. 69
2.6.1. Sensativity Analysis .....................................................................................................69
2.6.2.Duality ...........................................................................................................................77
CHAPTER THREE ..................................................................................................................... 87
TRANSPORTATION AND ASSIGNMENT PROBLEMS ..................................................... 87
3.0. Introduction ............................................................................................................................. 87
3.1. Transportation Problems ......................................................................................................... 87
3.1.1. Formulating The Transportation Models .........................................................................88
3.1.2. Finding An Initial Feasible Solution ................................................................................89
3.1.3. Evaluating/Testing A Solution For Optimality ..............................................................101
3.1.4 Special Cases In Transportation Problem .......................................................................112
3.2. Assignment Problem ............................................................................................................. 117
3.2.1. Introduction ....................................................................................................................117
3.2.2. Mathematical Statement Of Problem .........................................................................117
3.2.3. Application Areas Of Assignment Problem ...............................................................117
3.3.4. Solution Methods........................................................................................................118
3.2.5 Special Cases In Assignment Problem ....................................................................... 126
CHAPTER FOUR..................................................................................................................... 137
DECISION THEORY/ANALYSIS ......................................................................................... 137
4.0. Introduction .......................................................................................................................... 137
4.1. Characteristics Of Decision Theory ..................................................................................... 138
4.2. The Payoff Table.................................................................................................................. 139
4.3. Decision Making Under Certainty ....................................................................................... 139
4.4. Decision Making Under Complete Uncertainty .................................................................. 140

Ashenafi A. (PhD Research Fellow) Page 2

 4.4.1. Maximax Criteria ....................................................................................................... 140
4.4.2. Maximin Criteria ....................................................................................................... 140
4.4.3. Minimax Regret ......................................................................................................... 141
4.4.4. Principle Of Insufficient Reason/ Equal Likelihood/ Laplace Crteria ...................... 142
4.4.5.The Hurwitz Criterion ................................................................................................ 143
4.5. Decision Making Under Risk (With Probabilities) .............................................................. 144
4.5.1. Expected Monetary Value (Emv) .............................................................................. 144
4.5.2. Expected Opportunity Loss (Eol) .............................................................................. 145
4.5.3. Expected Value Of Perfect Information (Evpi) ......................................................... 145
4.6. Decision Theory And Decision Trees .................................................................................. 147
CHAPTER FIVE ...................................................................................................................... 158
NETWORK ANALYSIS/MODEL .......................................................................................... 158
5.0. Introduction .......................................................................................................................... 158
5.1 Pert And Cpm Network ........................................................................................................ 158
5.2. Project Crashing ................................................................................................................... 167
5.3. Probabilistic Time Estimates ............................................................................................... 172
UNIT SIX .......................................................................................................... Error! Bookmark not defined.
INTRODUCTION TO GAME THEORY .............................................................................. 178
6.0.Introduction ........................................................................................................................... 178
6.1 Overview Of Game Theory .................................................................................................. 178
6.2.Definition Of A Game And Game Theory ........................................................................... 178
6.3.The Basic Elements Of A Game ........................................................................................... 179
6.4. Two-Person Zero-Sum Games ............................................................................................. 181
References ................................................................................................................................... 187
Answer Key To Self Assessment Questions ............................................................................... 188

Ashenafi A. (PhD Research Fellow) Page 3

 MODULE INTRODUCTION
Dear Students! Welcome to the course operations research, commonly abbreviated as “OR.” It
is one of the major courses in business fields of studies such as Management, Leadership,
Accounting, Finance, Procurement, etc. This course is essentially an application of quantitative
techniques/models in business decision makings. The terms ‗operations research‘ and
‗management science‘ are often used interchangeably.

Decision-making in today‘s social and business environment has become a complex task. The
uncertainty of the future and the nature of competition and social interaction greatly increase the
difficulty of managerial decision-making. Knowledge and technology are changing rapidly, new
problems with little or no precedents continually arise. Hence, in order to effectively address this
problem and provide leadership in the advancing global age, decision-makers cannot afford to
make decision by simply applying their prior experiences, guess work or intuition. To this effect,
operations research as one of the quantitative decision making tool offers the decision-maker a
method of evaluating every possible alternative by using various techniques to know the
potential outcomes.

This course covers the basic quantitative models that can be used to describe and solve certain
kinds of problems such as product mix, investment, transportation, work assignment, capacity
planning, Project Management and other business related problems that confront managers.

The course has six units. To help you in studying the course material, this course is classified in
to two modules. The first module covers linear programming and its applications. It has two
units: Unit one deals with Introduction to operations research and unit two discusses about the
linear programming (LP).

The second module deals with applications of operations research and has four units:
Transportation and Assignment problems, Decision theory, Project Management, and
Introduction to Game Theory. Each unit is further subdivided into sections including summaries
and self-check exercises.

The time allotted for this course is equivalent to a course with four credit hours per week in a
conventional class room study. On average, one contact hour of conventional classroom is
considered to be equivalent to three study hours in distance learning. Therefore, you are expected
to devote at least 192 hours to successfully accomplish your work on this course.

After successful completion of this course, learners should be able to:

explain the historical background of operations research;
explain the basic concepts in operations research;
apply linear programming models in business problem solving;
be acquainted with the formulation of operations research models;
solve operations research models;
internalize the scientific approach to decision making;
be equipped with the techniques of project management;
explain introductory concepts of game theory; and
apply introductory concepts of game theory to decision making.

Ashenafi A. (PhD Research Fellow) Page 4

 CHAPTER ONE
INTRODUCTION TO OPERATIONS RESEARCH

1.0.INTRODUCTION
After Studying this chapter, you should be able to:
Understand the meaning, purpose, and tools of Operations Research
Describe the history of Operations Research
Describe the Stages of Operations Research
Explain the Applications of Operation Research
Describe the Limitations of Operation Research
Understand the Operations Research specialist and Manager relationship

1.1. INTRODUCTION TO OPERATIONS RESEARCH

The British/Europeans refer to "operational research", the Americans to "operations research" -
but both are often shortened to just "OR" - which is the term we will use. Another term which is
used for this field is "management science" ("MS"). The Americans sometimes combine the
terms Operations Research and MS together and say "OR/MS" or "ORMS". Yet other terms
sometimes used are "industrial engineering" ("IE") and "Decision Science" ("DS"). In recent
years there has been a move towards a standardization upon a single term for the field, namely
the term "OR".

Operation Research is a relatively new discipline. The contents and the boundaries of the OR are
not yet fixed. Therefore, to give a formal definition of the term Operations Research is a difficult
task. The Operations Research starts when mathematical and quantitative techniques are used to
substantiate the decision being taken. The main activity of a manager is the decision making. In
our daily life we make the decisions even without noticing them. The decisions are taken simply
by common sense, judgment and expertise without using any mathematical or any other model in
simple situations. But the decision we are concerned here with are complex and heavily
responsible. Examples are public transportation network planning in a city having its own layout
of factories, residential blocks or finding the appropriate product mix when there exists a large
number of products with different profit contributions and production requirement etc.

Operations Research tools are not from any one discipline. Operations Research takes tools from
different discipline such as mathematics, statistics, economics, psychology, engineering etc. and
combines these tools to make a new set of knowledge for decision making. Today, Operations
Research became a professional discipline which deals with the application of scientific methods
for making decision, and especially to the allocation of scarce resources. The main purpose of
Operations Research is to provide a rational basis for decisions making in the absence of
complete information, because the systems composed of human, machine, and procedures may
do not have complete information.

Operations Research can also be treated as science in the sense it describing, understanding and
predicting the systems behavior, especially man-machine system. Thus Operations Research
specialists are involved in three classical aspect of science, they are as follows:
Determining the systems behavior
Analyzing the systems behavior by developing appropriate models

Ashenafi A. (PhD Research Fellow) Page 5

 Predict the future behavior using these models

The emphasis on analysis of operations as a whole distinguishes the Operations Research from
other research and engineering. O.R. is an interdisciplinary discipline which provided solutions
to problems of military operations during World War II, and also successful in other operations.
Today business applications are primarily concerned with Operations Research analysis for the
possible alternative actions. The business and industry befitted from Operations Research in the
areas of inventory, reorder policies, optimum location and size of warehouses, advertising
policies, etc. As stated earlier defining Operations Research is a difficult task. The definitions
stressed by various experts and Societies on the subject together enable us to know what
Operations Research is, and what it does. They are as follows:

According to the Operational Research Society of Great Britain (OPERATIONAL

RESEARCH QUARTERLY, l3(3):282, l962), Operational Research is the attack of
modern science on complex problems arising in the direction and management of large
systems of men, machines, materials and money in industry, business, government and
defense. Its distinctive approach is to develop a scientific model of the system,
incorporating measurements of factors such as change and risk, with which to predict and
compare the outcomes of alternative decisions, strategies or controls. The purpose is to
help management determine its policy and actions scientifically.

Randy Robinson stresses that Operations Research is the application of scientific

methods to improve the effectiveness of operations, decisions and management. By
means such as analyzing data, creating mathematical models and proposing innovative
approaches, Operations Research professionals develop scientifically based information
that gives insight and guides decision- making. They also develop related software,
systems, services and products.

Morse and Kimball have stressed Operations Research is a quantitative approach and
described it as ― a scientific method of providing executive departments with a
quantitative basis for decisions regarding the operations under their control‖.

Saaty considers Operations Research as tool of improving quality of answers. He says,

―Operations Research is the art of giving bad answers to problems which otherwise
have worse answers‖.

Miller and Starr state, ―Operations Research is applied decision theory, which uses
any scientific, mathematical or logical means to attempt to cope with the problems that
confront the executive, when he tries to achieve a thorough-going rationality in dealing
with his decision problem‖.

Pocock stresses that Operations Research is an applied Science. He states ―Operations

Research is scientific methodology (analytical, mathematical, and quantitative) which by
assessing the overall implication of various alternative courses of action in a management
system provides an improved basis for management decisions‖.

Ashenafi A. (PhD Research Fellow) Page 6

Activity1: How do you define operations research?
______________________________________________________________________________
______________________________________________________________________________
_
1.2.HISTORY OF OPERATIONS RESEARCH
Operation Research is a relatively new discipline. Whereas 70 years ago it would have been
possible to study mathematics, physics or engineering (for example) at university it would not
have been possible to study Operation Research, indeed the term Operations Research did not
exist then. It was really only in the late 1930's that operational research began in a systematic
fashion, and it started in the UK. As such it would be interesting to give a short history of
Operations Research.

1936
Early in 1936 the British Air Ministry established Bawdsey Research Station, on the east coast,
near Felixstowe, Suffolk, as the centre where all pre-war radar experiments for both the Air
Force and the Army would be carried out. Experimental radar equipment was brought up to a
high state of reliability and ranges of over 100 miles on aircraft were obtained.

It was also in 1936 that Royal Air Force (RAF) Fighter Command, charged specifically with the
air defense of Britain, was first created. It lacked however any effective fighter aircraft - no
Hurricanes or Spitfires had come into service - and no radar data was yet fed into its very
elementary warning and control system.

It had become clear that radar would create a whole new series of problems in fighter direction
and control so in late 1936 some experiments started at Biggin Hill in Kent into the effective use
of such data. This early work, attempting to integrate radar data with ground based observer data
for fighter interception, was the start of Operations Research.

1937
The first of three major pre-war air-defense exercises was carried out in the summer of 1937.
The experimental radar station at Bawdsey Research Station was brought into operation and the
information derived from it was fed into the general air-defense warning and control system.
From the early warning point of view this exercise was encouraging, but the tracking information
obtained from radar, after filtering and transmission through the control and display network,
was not very satisfactory.

1938
In July 1938 a second major air-defense exercise was carried out. Four additional radar stations
had been installed along the coast and it was hoped that Britain now had an aircraft location and
control system greatly improved both in coverage and effectiveness. Not so! The exercise
revealed, rather, that a new and serious problem had arisen. This was the need to coordinate and
correlate the additional, and often conflicting, information received from the additional radar
stations. With the outbreak of war apparently imminent, it was obvious that something new -
drastic if necessary - had to be attempted. Some new approach was needed.

Ashenafi A. (PhD Research Fellow) Page 7

Accordingly, on the termination of the exercise, the Superintendent of Bawdsey Research
Station, A.P. Rowe, announced that although the exercise had again demonstrated the technical
feasibility of the radar system for detecting aircraft, its operational achievements still fell far
short of requirements. He therefore proposed that a crash program of research into the
operational - as opposed to the technical - aspects of the system should begin immediately. The
term "operational research" (RESEARCH into military OPERATIONS) was coined as a suitable
description of this new branch of applied science. The first team was selected from amongst the
scientists of the radar research group the same day.

1939
In the summer of 1939 Britain held what was to be its last pre-war air defense exercise. It
involved some 33,000 men, 1,300 aircraft, 110 antiaircraft guns, 700 searchlights, and 100
barrage balloons. This exercise showed a great improvement in the operation of the air defense
warning and control system. The contribution made by the Operations Research team was so
apparent that the Air Officer Commander-in-Chief RAF Fighter Command (Air Chief Marshal
Sir Hugh Dowding) requested that, on the outbreak of war, they should be attached to his
headquarters at Stanmore in north London.

Initially, they were designated the "Stanmore Research Section". In 1941 they were redesignated
the "Operational Research Section" when the term was formalized and officially accepted, and
similar sections set up at other RAF commands.

1940
On May 15th 1940, with German forces advancing rapidly in France, Stanmore Research Section
was asked to analyses a French request for ten additional fighter squadrons (12 aircraft a
squadron - so 120 aircraft in all) when losses were running at some three squadrons every two
days (i.e. 36 aircraft every 2 days). They prepared graphs for Winston Churchill (the British
Prime Minister of the time), based upon a study of current daily losses and replacement rates,
indicating how rapidly such a move would deplete fighter strength. No aircraft were sent and
most of those currently in France were recalled.

This is held by some to be the most strategic contribution to the course of the war made by
Operations Research (as the aircraft and pilots saved were consequently available for the
successful air defense of Britain, the Battle of Britain).

1941 onward
In 1941, an Operational Research Section (ORS) was established in Coastal Command which
was to carry out some of the most well-known Operations Research work in World War II.

The responsibility of Coastal Command was, to a large extent, the flying of long-range sorties by
single aircraft with the object of sighting and attacking surfaced U-boats (German submarines).
The technology of the time meant that (unlike modern day submarines) surfacing was necessary
to recharge batteries, vent the boat of fumes and recharge air tanks. Moreover U-boats were
much faster on the surface than underwater as well as being less easily detected by sonar.

Ashenafi A. (PhD Research Fellow) Page 8

Thus the Operation Research started just before World War II in Britain with the establishment
of teams of scientists to study the strategic and tactical problems involved in military operations.
The objective was to find the most effective utilization of limited military resources by the use of
quantitative techniques. Following the end of the war OR spread, although it spread in different
ways in the UK and USA.

In 1951 a committee on Operations Research formed by the National Research Council of USA,
and the first book on ―Methods of Operations Research‖, by Morse and Kimball, was published.
In 1952 the Operations Research Society of America came into being.

Success of Operations Research in army attracted the attention of the industrial mangers who
were seeking solutions to their complex business problems. Now a days, almost every
organization in all countries has staff applying operations research, and the use of operations
research in government has spread from military to wide variety of departments at all levels. The
growth of operations research has not limited to the U.S.A. and U.K., it has reached many
countries of the world.

Activity2: When do you think operations research came into an existence as a discipline (field of
study)?
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________

1.3.STAGES OF DEVELOPMENT OF OPERATIONS RESEARCH

The stages of development of Operations Research are also known as phases and process of
Operations Research, which has six important steps. These six steps are arranged in the
following order:

Step1: Observe the problem environment

The first step in the process of Operations Research development is the problem environment
observation. This step includes different activities; they are conferences, site visit, research,
observations etc. These activities provide sufficient information to the Operations Research
specialists to formulate the problem.

Step 2: Analyze and define the problem

This step is analyzing and defining the problem. In this step in addition to the problem definition
the objectives, uses and limitations of Operations Research. study of the problem also defined.
The outputs of this step are clear grasp of need for a solution and its nature understanding.

Step 3: Develop a model

This step develops a model; a model is a representation of some abstract or real situation. The
models are basically mathematical models, which describes systems, processes in the form of
equations, formula/relationships. The different activities in this step are variables definition,
formulating equations etc. The model is tested in the field under different environmental

Ashenafi A. (PhD Research Fellow) Page 9

constraints and modified in order to work. Sometimes the model is modified to satisfy the
management with the results.

Step 4: Select appropriate data input

A model works appropriately when there is appropriate data input. Hence, selecting appropriate
input data is important step in the Operations Research development stage or process. The
activities in this step include internal/external data analysis, fact analysis, and collection of
opinions and use of computer data banks. The objective of this step is to provide sufficient data
input to operate and test the model developed in Step 3.

Step 5: Provide a solution and test its reasonableness

This step is to get a solution with the help of model and input data. This solution is not
implemented immediately, instead the solution is used to test the model and to find there is any
limitations. Suppose if the solution is not reasonable or the behavior of the model is not proper,
the model is updated and modified at this stage. The output of this stage is the solution(s) that
supports the current organizational objectives.

At this step the solution obtained from the previous step is implemented. The implementation of
the solution involves mo many behavioral issues. Therefore, before implementation the
implementation authority has to resolve the issues. A properly implemented solution results in
quality of work and gains the support from the management.

Applications of Operations Research

Today, almost all fields of business and government utilizing the benefits of Operations
Research. There are voluminous of applications of Operations Research. Although it is not
feasible to cover all applications of operations research in brief. The following are the
abbreviated set of typical operations research applications to show how widely these techniques
are used today:

Accounting:
Assigning audit teams effectively
Credit policy analysis
Cash flow planning Developing standard costs Establishing costs for byproducts
Planning of delinquent account strategy

Construction:
Project scheduling, monitoring and control Determination of proper work force Deployment of work
force
Allocation of resources to projects

Facilities Planning:
Factory location and size decision
Estimation of number of facilities required
Hospital planning
International logistic system design Transportation loading and unloading Warehouse location decision

10

Ashenafi A. (PhD Research Fellow) Page 10

Finance:
Building cash management models
Allocating capital among various alternatives
Building financial planning models
Investment analysis
Portfolio analysis
Dividend policy making

Manufacturing:
Inventory control
Marketing balance projection Production scheduling Production smoothing

Marketing:
Advertising budget allocation
Product introduction timing
Selection of Product mix
Deciding most effective packaging alternative

Organizational Behavior / Human Resources:

Personnel planning
Recruitment of employees
Skill balancing
Training program scheduling
Designing organizational structure more effectively

Purchasing:
Optimal buying
Optimal reordering
Materials transfer

Research and Development:

R & D Projects control
R & D Budget allocation
Planning of Product introduction

Activity 3: What is the nature and significance of operations research?

______________________________________________________________________________
______________________________________________________________________________

1.4.PHASES OF AN OPERATIONS RESEARCH PROJECT

Drawing on our experience with the Two Mines problem we can identify the phases that a (real-
world) OR project might go through:

Problem identification
Diagnosis of the problem from its symptoms if not obvious (i.e. what is the problem?)

11

Ashenafi A. (PhD Research Fellow) Page 11

 Delineation of the sub-problem to be studied. Often we have to ignore parts of the entire problem.
Establishment of objectives, limitations and requirements.

Formulation as a mathematical model

Model is a representation of real objects/situations. A model is a selective abstraction of reality.
It is a simplified and often idealized representation of real world problems. A good model should
capture the important details of reality without including minor details that would be difficult to
understand rather than illuminate. In this sense, a model can be taken as a selective abstraction
because only those details that are considered as important for the problem at hand are included
in the model. Thus, it is important to carefully decide which aspects of reality to include in a
model.

Activity 4: What are the different types of models?

______________________________________________________________________________
______________________________________________________________________________

Types of Models
Types of models are classified in a number of ways taking different basis as a ground like the
level of abstraction, form and the way they are used. Based on the level of abstraction, models
can be classified as:
Physical (iconic) models
Analog models
Symbolic (mathematical) models

Physical (icon) models: These are the representation of the situation, problem or a real object.
In modeling terminology, physical replicas are referred to as iconic models. For example, a scale
of an airplane is a representation of a real air plane. Similarly, a child‘s toy truck is a model of a
real truck. The model airplane and toy truck are examples of models that are physical replicas of
the real objects.

Analog models: These are abstract models mostly showing inter and intra relationships
between two or more parameters. It is two dimensional. For example, it may show the
relationship between an independent variable with that of a dependent variable. For instance;
histogram, frequency table, cause-effect diagram, flow charts, Gantt charts, price-demand graph,
world map and others.

Symbolic (mathematical) models: The third classification of models-the type we will

primarily be studying in this course- includes representations of a problem by a system of
symbols and mathematical relationships or expressions. Such models are critical part of any
quantitative approach to decision making.

Examples:
(a + b) 2=a2+2ab+b2
Total Cost=10Q+1000 (where Q is quantity produced)

12

Ashenafi A. (PhD Research Fellow) Page 12

 Total Revenue=30Q (where Q is quantity sold)
Total profit=20Q-1000 (where Q is quantity sold). The assumption here is quantity sold is
equal to quantity produced.

Types of mathematical models

The attention of this course is geared towards this category of models. Depending on the degree
of uncertainty we have about a problem, we can classify these models as deterministic and
probabilistic/predictive/stochastic/.

Deterministic models: represent problems of decision making under conditions of certainty. That
is to say, we use deterministic models when we know all the numerical values in the model
certainly. For example, if we are going to decide on the size of a warehouse to be constructed,
our model will be deterministic if we know the exact amount of demand (quantity of commodity
to be stored in the warehouse).

Probabilistic/stochastic models: deal with problems of decision making under uncertainty or risk
and the probabilities of the alternative states of nature are known.

It may be that a problem can be modeled in differing ways, and the choice of the appropriate
model may be crucial to the success of the OR project. In addition to algorithmic considerations
for solving the model (i.e. can we solve our model numerically?) we must also consider the
availability and accuracy of the real-world data that is required as input to the model.

Note that the "data barrier" ("we don't have the data!!!") can appear here, particularly if people
are trying to block the project. Often data can be collected/ estimated, particularly if the potential
benefits from the project are large enough.

You will also find, if you do much OR in the real-world, that some environments are naturally
data-poor, that is the data is of poor quality or nonexistent and some environments are naturally
data-rich. As examples of this church location study (a data-poor environment) and an airport
terminal check- in desk allocation study (a data-rich environment). This issue of the data
environment can affect the model that you build. If you believe that certain data can never
(realistically) be obtained there is perhaps little point in building a model that uses such data.

3. Model validation (or algorithm validation)

Model validation involves running the algorithm for the model on the computer in order to
ensure:
the input data is free from errors
the computer program is bug-free (or at least there are no outstanding bugs)
the computer program correctly represents the model we are attempting to validate
the results from the algorithm seem reasonable (or if they are surprising we can at least
understand why they are surprising). Sometimes we feed the algorithm historical input
data (if it is available and is relevant) and compare the output with the historical result.

13

Ashenafi A. (PhD Research Fellow) Page 13

4. Solution of the model
Standard computer packages, or specially developed algorithms, can be used to solve the model
(as mentioned above). In practice, a "solution" often involves very many solutions under varying
assumptions to establish sensitivity. For example, what if we vary the input data (which will be
inaccurate anyway), then how will this affect the values of the decision variables? Questions of
this type are commonly known as "what if" questions nowadays.

Note here that the factors which allow such questions to be asked and answered are:
the speed of processing (turn-around time) available by using pc's; and
the interactive/user-friendly nature of many pc software packages.

Implementation
This phase may involve the implementation of the results of the study or the implementation of
the algorithm for solving the model as an operational tool (usually in a computer package). In the
first instance detailed instructions on what has to be done (including time schedules) to
implement the results must be issued. In the second instance operating manuals and training
schemes will have to be produced for the effective use of the algorithm as an operational tool.

It is believed that many of the OR projects which successfully pass through the first four phases
given above fail at the implementation stage (i.e. the work that has been done does not have a
lasting effect). As a result one topic that has received attention in terms of bringing an OR
project to a successful conclusion (in terms of implementation) is the issue of client involvement.
By this is meant keeping the client (the sponsor/originator of the project) informed and consulted
during the course of the project so that they come to identify with the project and want it to
succeed.
.

Figure 1.1. A graphical description of Operations research decision making process

Activity 5: What are the different types of models?
______________________________________________________________________________
______________________________________________________________________________

14

Ashenafi A. (PhD Research Fellow) Page 14

1.5. OPERATIONS RESEARCH TOOLS AND TECHNIQUES
Operations Research uses any suitable tools or techniques available. The common frequently
used tools/techniques are mathematical procedures, cost analysis, electronic computation.
However, operations researchers given special importance to the development and the use of
techniques like linear programming, game theory, decision theory, queuing theory, inventory
models and simulation. In addition to the above techniques, some other common tools are non-
linear programming, integer programming, dynamic programming, sequencing theory, Markov
process, network scheduling (PERT/CPM), symbolic Model, information theory, and value
theory. There is many other Operations Research tools/techniques also exists. The brief
explanations of some of the above techniques/tools are as follows:

Linear Programming: This is a constrained optimization technique, which optimize some

criterion within some constraints. In Linear programming the objective function (profit, loss or
return on investment) and constraints are linear. There are different methods available to solve
linear programming.

Game Theory: This is used for making decisions under conflicting situations where there are
one or more players/opponents. In this the motive of the players are dichotomized. The success
of one player tends to be at the cost of other players and hence they are in conflict.

Decision Theory: Decision theory is concerned with making decisions under conditions of
complete certainty about the future outcomes and under conditions such that we can make some
probability about what will happen in future.

Queuing Theory: This is used in situations where the queue is formed (for example customers
waiting for service, aircrafts waiting for landing, jobs waiting for processing in the computer
system, etc). The objective here is minimizing the cost of waiting without increasing the cost of
servicing.

Inventory Models: Inventory model make a decisions that minimize total inventory cost. This
model successfully reduces the total cost of purchasing, carrying, and out of stock inventory.

Simulation: Simulation is a procedure that studies a problem by creating a model of the process
involved in the problem and then through a series of organized trials and error solutions attempt
to determine the best solution. Sometimes this is a difficult/time consuming procedure.
Simulation is used when actual experimentation is not feasible or solution of model is not
possible.

Non-linear Programming: This is used when the objective function and the constraints are not
linear in nature. Linear relationships may be applied to approximate non-linear constraints but
limited to some range, because approximation becomes poorer as the range is extended. Thus,
the non-linear programming is used to determine the approximation in which a solution lies and
then the solution is obtained using linear methods.

15

Ashenafi A. (PhD Research Fellow) Page 15

Dynamic Programming: Dynamic programming is a method of analyzing multistage decision
processes. In this each elementary decision depends on those preceding decisions and as well as
external factors.

Integer Programming: If one or more variables of the problem take integral values only then
dynamic programming method is used. For example number or motor in an organization, number
of passenger in an aircraft, number of generators in a power generating plant, etc.

Markov Process: Markov process permits to predict changes over time information about the
behavior of a system is known. This is used in decision making in situations where the various
states are defined. The probability from one state to another state is known and depends on the
current state and is independent of how we have arrived at that particular state.

Network Scheduling: This technique is used extensively to plan, schedule, and monitor large
projects (for example computer system installation, R & D design, construction, maintenance,
etc.). The aim of this technique is minimize trouble spots (such as delays, interruption,
production bottlenecks, etc.) by identifying the critical factors. The different activities and their
relationships of the entire project are represented diagrammatically with the help of networks and
arrows, which is used for identifying critical activities and path. There are two main types of
technique in network scheduling, they are:

Program Evaluation and Review Technique (PERT) – is used when activities time is not
known accurately/ only probabilistic estimate of time is available.

Critical Path Method (CPM) – is used when activities time is known accurately.
Information Theory: This analytical process is transferred from the electrical communication
field to operations research field. The objective of this theory is to evaluate the effectiveness of
flow of information with a given system. This is used mainly in communication networks but
also has indirect influence in simulating the examination of business organizational structure
with a view of enhancing flow of information.

Activity 6: List and explain types of operations research techniques.

______________________________________________________________________________
______________________________________________________________________________
1.6. LIMITATIONS OF OPERATIONS RESEARCH
Operations Research has number of applications; similarly it also has certain limitations. These
limitations are mostly related to the model building and money and time factors problems
involved in its application. Some of them are as given below:

Distance between O.R. specialist and Manager: Operations Researchers job needs a
mathematician or statistician, who might not be aware of the business problems. Similarly,
a manager is unable to understand the complex nature of Operations Research. Thus there
is a big gap between the two personnel.
Magnitude of Calculation: The aim of the O.R. is to find out optimal solution taking into
consideration all the factors. In this modern world these factors are enormous and

16

Ashenafi A. (PhD Research Fellow) Page 16

 expressing them in quantitative model and establishing relationships among these require
voluminous calculations, which can be handled only by machines.
Money and Time Costs: The basic data are subjected to frequent changes, incorporating
these changes into the operations research models is very expensive. However, a fairly
good solution at present may be more desirable than a perfect operations research solution
available in future or after some time.
Non-quantifiable Factors: When all the factors related to a problem can be quantifiable
only then operations research provides solution otherwise not. The non-quantifiable
factors are not incorporated in O.R. models. Importantly O.R. models do not take into
account emotional factors or qualitative factors.
Implementation: Once the decision has been taken it should be implemented. The
implementation of decisions is a delicate task. This task must take into account the
complexities of human relations and behavior and in some times only the psychological
factors.

1.7. SUMMARY
Operations Research is relatively a new discipline, which originated in World War II, and
became very popular throughout the world. India is one of the few first countries in the world
who started using operations research. Operations Research is used successfully not only in
military/army operations but also in business, government and industry. Now a day‘s operations
research is almost used in all the fields.

Proposing a definition to the operations research is a difficult one, because its boundary and
content are not fixed. The tools for operations search is provided from the subject‘s viz.
economics, engineering, mathematics, statistics, psychology, etc., which helps to choose possible
alternative courses of action. The operations research tool/techniques include linear
programming, non-linear programming, dynamic programming, integer programming, Markov
process, queuing theory, etc.

Operations Research has a number of applications. Similarly it has a number of limitations,

which is basically related to the time, money, and the problem involves in the model building.
Day-by- day, operations research gaining acceptance, because it improve decision making
effectiveness of the managers. Almost all the areas of business use the operations research for
decision making.

1.8. Self Assessment Questions-1

Part I. Multiple choice questions
1.For analyzing a problem, decision makers should normally study
its qualitative aspects
its qualitative aspects
both (a) and (b)
neither (a) nor (b)
A model is:
an essence of reality
an approximation
an idealization

17

Ashenafi A. (PhD Research Fellow) Page 17

3.all of the above Managerial decisions are based on
an evaluation of quantitative data
the use of qualitative factors
numbers produced by formal models
all of the above
4.Every mathematical model
must be deterministic
requires computer aid for its solution
represent data in numerical form
all of the above
A physical model is an example of
an iconic model
an analog model
a verbal model
a mathematical model
6.A qualitative approach to decision making is used when______________
the problem is simple
the problem is new
the problem is familiar
all except b
One of the following is not an area of application of OR
forecasting
capital budgeting
project management
none of the above

Part II. Short Answer Questions

8.The first step in decision making is _________________________.
9.Operations research is multidisciplinary. Discuss
Define decision making.
Differentiate between deterministic and probabilistic models.

18

Ashenafi A. (PhD Research Fellow) Page 18

 CHAPTER TWO
LINEAR PROGRAMMING MODELS (LPM)

2.0. INTRODUCTION
In the previous chapter, you studied about introduction to operations research. In this chapter,
you will study about Linear programming (LP) which is one of the most popular tools of
operations research. LP models enable users to find optimal (best possible) solutions to certain
problems in which the solution must satisfy a given set of requirements, or constraints. The unit
has four main sections. Section one discusses with introduction to linear programming. It
provides general introduction to linear programming. It introduces terminologies in LP, discusses
the use of linear programming models and illustrates formulation of models. Section two
describes the graphical solutions to linear programming problems. Section three covers the
simplex method, a general purpose algebraic technique for solving linear programming
problems. Section four discusses post optimality analysis of optimal solutions. In each section,
there are in-text questions and learning activities to initiate your study and to check yourself
whether you understood the contents of each section well. Moreover, at the end of the unit, there
are checklists, summaries and self check exercise questions that will facilitate your study and
ensure your understanding of the concepts in the unit. Therefore, please work hard to understand
this unit very well and to appreciate the application of linear programming to business decision
makings.

After studying this lesson, you should be able to:

Define the basic concepts in LP;
Formulate Linear Programming Problem
Identify the characteristics of linear programming problem
Make a graphical analysis of the linear programming problem
Solve the problem graphically
Understand the basics of simplex method
Explain the simplex calculations
Describe various solutions of Simplex Method
Understand Big-M method
Realize the special cases of LP model;
Perform sensitivity analysis of an LP model
Understand the Dual Linear programming Problem
Formulate and solve the Dual Problem
Understand the purposes of a Dual Problem

2.1.INTRODUCTION TO LINEAR PROGRAMMING MODELS

In 1947, George Danzig developed the use of algebra for determining solutions to problems that
involved the optimal allocation of scarce resources. In spite of numerous potential applications in
business, response to this new technique was low due to substantial computational burden, which
is now removed with subsequent advances in computer technology and related software during
the last three decades.

19

Ashenafi A. (PhD Research Fellow) Page 19

The term linear implies that all the mathematical relations used in the problem are linear or
straight-line relations, while the term programming refers to the method of determining a
particular program or plan of action, i.e., the use of algorithms that is a well defined sequence of
steps that will lead to an optimal solution. Taken as a whole, the term linear programming refers
to a family of mathematical techniques for determining the optimum allocation of resources and
obtaining a particular objective when there are alternative uses of the limited or constrained
resources.

The technique of linear programming is applicable to problems in which the total effectiveness
can be expressed as linear function of individual allocations and the limitations on resources give
rise to linear equation or inequalities of the individual allocations. The usefulness of this
technique is enhanced by the availability of several user-friendly softwares such as STORM,
TORA, QSB+, LINDO, etc. However, there is no general package for building an LP model.
Model building is an art of practice.

2.2.LINEAR PROGRAMMING MODELS (LPM)

Linear programming models are mathematical representations of LP problems. Linear
programming models have certain characteristics in common. Knowledge of these characteristics
enables us to recognize problems that are amenable to a solution using LP models, and to be able
to correctly formulate an LP model. These characteristics can be grouped as components and
assumptions. The components relate to the structure of a model, where as the assumptions reveal
the conditions under which the model is valid.

2.2.1. COMPONENTS OF LP MODELS

There are four major components of LP models including: Objective function, decision variables,
constraints and parameters.
a) Objective and Objective Function
The objective in problem solving is the criterion by which all decisions are evaluated. It provides
the focus for problem solving. In linear programming models, a single, quantifiable objective
must be specified by the decision maker. Because we are dealing with optimization, the objective
will be either maximization or minimization. Hence, every LP problem will be either
maximization or a minimization problem. Once the objective is specified, it becomes the
measure of effectiveness against which alternate solutions are judged. An LP model consists of a
mathematical statement of the objective called the objective function.
b) Decision variables
They represent unknown quantities to be solved for. The decision maker can control the value of
the objective, which is achieved through choices in the levels of decision variables. For example,
how much of each product should be produced in order to obtain the greatest profit?

c) Constraints
However, the ability of a decision maker to select values of the decision variables in an LP
problem is subject to certain restrictions or limits coming from a variety of sources. The
restrictions may reflect availabilities of resources (e.g., raw materials, labor time, etc.), legal or
contractual requirements (e.g., product standards, work standards, etc.), technological
requirements (e.g., necessary compressive strength or tensile strength) or they may reflect other
limits based on forecasts, customer orders, company policies, and so on. In LP model, the

20

Ashenafi A. (PhD Research Fellow) Page 20

restrictions are referred to as constraints. Only solutions that satisfy all constraints in a model are
acceptable and are referred to as feasible solutions. The optimal solution will be the one that
provides the best value for the objective function.

Constraints can be arranged into three groups:

System constraints – involve more than one decision variable,
Individual constraints – involve only one variable, and
Non-negativity constraints – specify that no variable will be allowed to take on a negative
value. The non-negativity constraints typically apply in an LP model, whether they are
explicitly stated or not.

Parameters
The objective function and the constraints consist of symbols that represent the decision
variables (e.g., X1, X2, etc.) and numerical values called parameters. The parameters are fixed
values that specify the impact that one unit of each decision variable will have on the objective
and on any constraint it pertains to as well as the numerical value of each constraint. The
following simple example illustrates the components of LP models:

Activity 1: Identifying components of an LP model

Given the following LP model, answer the questions that follow:
Maximize Z = 4x1 + 7x2 + 5x3 (profit)
Subject to:
2x1 + 3x2+ 6x3
5x1 +  300 labor hours
3x1 + 5x2+ 4x3  200kg raw material A
2x3 x1  360kg raw material B =30
x2 units
x1, x2 ,x3   40
0 …………...non-negativity constraint
units

21

Ashenafi A. (PhD Research Fellow) Page 21

 Identify the decision variables.
What is the objective of the LP models?
Identify the objective function of the model.
What are the unit profit of product 1, product 2, and product 3 respectively?
If x1 =30, x2 =50, x3=0, what is the value of the objective function?

Comments: The unknown variables in the model are decision variables. The objective of the
model is mentioned in the objective function. The unit profit of product 1, product 2, and product
3 are the coefficients of x1, x2 ,x3 in the objective function respectively. To find the values of the
objective function given the values of the decision variables, you can easily substitute the values
of the decision variables in the objective function.

2.2.2.ASSUMPTIONS OF LP MODELS
Linearity (proportionality)
The linearity requirement is that each decision variable has a linear impact on the objective
function and in each constraint in which it appears. In terms of a mathematical model, a function
or equation is linear when the variables included are all to the power 1 (not squared, cubed,
square root, etc.) and no products (e.g., x1x2) appear. On the other hand, the amount of each
resource used (supplied) and its contribution to the profit (or cost) in the objective function must
be proportional to the value of each decision variable. For example, if production of one unit
requires 5 hours of a particular resource, then making 3 units of that product requires 15 hours
(3x5) of that resource.

b) Divisibility (Continuity)
The divisibility requirement pertains to potential values of decision variables. It is assumed that
non-integer values are acceptable. However, if the problem concerns, for example, the optimal
number of houses to construct, 3.5 do not appear to be acceptable. Instead, that type of problem
would seem to require strictly integer solutions. In such cases, integer-programming methods
should be used. It should be noted, however, that some obvious integer type situations could be
handled under the assumption of divisibility. For instance, suppose 3.5 to be the optimal number
of television sets to produce per hour, which is unacceptable, but it would result in 7 sets per two
hours, which would then be acceptable.

c) Certainty
This requirement involves two aspects of LP models. One aspect relates to the model parameters,
i.e., the numerical values. It is assumed that these values are known and constant. In practice,
production times and other parameters may not be truly constant. Therefore, the model builder
must make an assessment as to the degree to which the certainty requirement is met. Large
departures almost surely will have a significant effect on the model. The other aspect is the
assumption that all relevant constraints have been identified and represented in the model.

d) Additivity
The value of the objective function and the total amount of each resource used (or supplied),
must be equal to the sum of the respective individual contributions (profit or cost) by decision
variables. For example, the total profit earned from the sale of two products A and B must be

22

Ashenafi A. (PhD Research Fellow) Page 22

equal to the sum of the profits earned separately from A and B. Similarly, the amount of a
resource consumed for producing A and B must be equal to the sum of resources used for A and
B respectively.

e) Non-negativity
It assumes that negative values of variables are unrealistic and, therefore, will not be considered
in any potential solutions. Only positive values and zero will be allowed and the non-negativity
assumption is inherent in LP models.

Activity 1: Attempt the following questions

Define what linear programming models are and explain its purpose.
______________________________________________________________________________
______________________________________________________________________________
What are the assumptions and components of linear programming models?
______________________________________________________________________________
_____________________________________________________________________________

2.3.FORMULATING LP MODELS
Just as it is to define a problem, careful formulation of the model that will be used to solve the
problem is important. Linear programming algorithms (solution techniques) are widely used and
understood and computer packages are readily available for solving LP problems. Consequently,
obtaining solutions is not the real issue, what is very important to note is failure to check that all
constraints have been accounted for and have been correctly formulated results in ill-structuring
of the model that can easily lead to poor decisions.

Steps in formulating LP models:

Identify the decision variables.
Determine the objective function.
Identify the constraints.
Build and validate the model.

In many cases, the decision variables are obvious; in others it might require brief discussion with
the appropriate manager. However, identifying the constraints and determining appropriate
values for the parameters can require considerable time and effort. Potential sources of
information include historical records, interviews with managers and staff, and data collection.
Validating the model will involve a critical review of the output, perhaps under a variety of
inputs, in order to decide if the results are reasonable.

2.4.LINEAR PROGRAMMING APPLICATIONS

There is a wide range of problems that lend themselves to solution by linear programming
techniques. This discussion is only meant to give an indication of the LP techniques for
managerial decision making and the apparent diversity of situations to which linear programming
can be applied. Some of these include production management (product mix, blending problems,
production planning, Assembly line balancing…), Marketing management (media selection,
traveling sales man problem, physical distribution), Financial management (portfolio selection,

23

Ashenafi A. (PhD Research Fellow) Page 23

profit planning), agricultural application, military applications, personnel management (staffing
problem, Determination of equitable salary and etc.

Example1: ABC private limited company is engaged in the production of power and traction
transformers. Both of these categories of transformers pass through three basic processes: core
preparation, core to coil assembly, and vapor phase drying. A power transformer yields a
contribution of Birr 50,000 and traction transformer contributes Birr 10,000. The time required in
the production of these two products in terms of hours for each of the processes is as follows.

Power transformer Traction Transformer

Core preparation 75 15
Core to Coil Assembly 160 30
Vapor Phase Drying 45 10

If the capacities available are 1000, 1500, and 750 machine hours in each processes respectively,
formulate the problem as LP?

Solution
Step1. Identify decision variables
Since the products to be produced are power and traction transformers using the available
resource to attain the objective set, we consider them as decision variables. This is because the
organization‘s problem here is how many of each product to produce in order to attain the
objective, which requires the management decision.

Let X1 = the no of power transformers to be produced.

X2= the no of traction transformer to be produced

Step2. Determine Objective Function

From the problem above, we understand that the problem is maximization problem.
Hence,
Maximize Z = 50,000X1+ 10,000X2
This is because, each unit of X1 contributes Birr 50,000 and X2 contributes Birr 10,000 to
objective function.

Step 3. Identify constraints and parameters

Here we have two constraints: structural and non-negativity constraints. The structural
constraint is the amount of machine hours available in each process.

Parameters are already identified in the table of the problem.

Note: Step 3 and 4 can be performed simultaneously as:
75X1 + 15X2 < 1000 hrs- Core preparation process
160X1 + 30X2 < 1500 hrs- Core to Coil Assembly Machine hour constraint
45X1 + 10X2 < 750 hrs- Vapor Phase Drying
X1, X2 > 0 Non-negativity constraints

24

Ashenafi A. (PhD Research Fellow) Page 24

Step 4. Building and validating the model
Max Z = 50,000X1+ 10,000X2
Subject to:
75X1 + 15X2 < 1000 hrs
160X1 + 30X2 < 1500 hrs
45X1 + 10X2 < 750 hrs
X1,X2>0

Example 2: An individual investor has Birr 70,000 to divide among several investments. The
alternative investments are municipal bonds with an 8.5% return, certificates of deposits with a
10% return, Treasury bill with a 6.5% return, and income bonds with a 13% return. The amount
of time until maturity is the same for each alternative. However, each investment alternative has
a different perceived risk to the investor; thus it is advisable to diversify. The investor wants to
know how much to invest in each alternative in order to maximize the return. The following
guidelines have been established for diversifying the investment and lessening the risk perceived
by the investor.

No more than 20% of the total investment should be in an income bonds.

The amount invested in certificates of deposit should not exceed the amount invested in other
three alternatives.
At least 30% of the investment should be in treasury bills and certificates of deposits.
The ratio of the amount invested in municipal bonds to the amount invested in treasury bills
should not exceed one to three.
The investor wants to invest the entire Birr 70,000.

Required: Formulate a LP model for the problem.

Solution
Step1. Identify decision variables
Four decision variables represent the monetary amount invested in each investment alternative.
Let X1, X2, X3 and X4 represent investment on municipal bonds, certificates of deposits,
Treasury bill, and income bonds respectively.

Step2. Determine Objective Function

The problem is maximization because from the word problem we know that the objective of the
investor is to maximization return from the investment in the four alternatives. Therefore, the
objective function is expressed as:

Max Z= 0.085 X1+ 0.1X2+ 0.065X3+0.13X4

Where,
Z = total return from all investment
0.085 X1= represents return from the investment in municipal bonds.
100 X2= represents return from the investment in certificates in
deposit 0.065 X3= represents return from the investment in treasury bills
0.130 X4= represents return from the investment in income bonds

25

Ashenafi A. (PhD Research Fellow) Page 25

Step 3. Identify constraints and parameters
In this problem the constraints are the guide lines established for diversifying the total
investment. Each guideline is transformed in to mathematical constraint separately.

Guideline 1 above is transformed as:

X4 ≤20 %( 70,000)  X4 <14,000

Guideline 2
X2≤X1 +X3+ X4  X2 - X1 -X3- X4≤0

Guideline 3
X2+ X3 ≥ 30 %( 70,000)  X2 + X3 ≥ 21,000

Guideline 4
X1/ X3≤1/3 3X1<X33X1-X3≤ 0

Guideline 5
X1+ X2+ X3+X4 = 70,000, because the investor‘s money equals to the sum of money invested in
all alternatives.

This last constraint differs from ≤ and ≥ inequalities previously developed in that there is a
specific requirement to invest an exact amount. Therefore, the possibility to invest more or less
than Birr 70,000 is not considered.

This problem contains all three types of constraints possible in LP problems: ≤ , ≥ and =. As this
problem demonstrates there is no restriction on mixing these types of constraints.

Step 4. Building and validating the model

The complete LPM for this problem can be summarized as:
Maximize Z = 0.085 X1+ 0.1X2+ 0.065X3+0.13X4
Subjected to:
X4 ≤ 14,000
X2- X1 -X3 - X4≤ 0
X2+X3 ≥21,000
3X1-X3 ≤ 0
X1 + X2 + X3 + X4 = 70,000,
X1+ X2+ X3+X4 0

Example 3: A chemical corporation produces a chemical mixture for the customer in 1000-lb
batches. The mixture contains three ingredients- Zinc, mercury and potassium. The mixture must
conform to formula specifications (i.e., a recipe) supplied by a customer. The company wants to
know the amount of each ingredient to put in the mixture that will meet all the requirements of
the mix and minimize total cost.

The customer has supplied the following formula specifications for each batch of mixture.
1. The mixture must contain at least 200 lb of Mercury

26

Ashenafi A. (PhD Research Fellow) Page 26

 The mixture must contain at least 300 lb of Zinc.
The mixture must contain at least 100 lb of Potassium
The cost per lb of mixture is Birr4 of Mercury, Birr 8 of Zinc and of Potassium of Birr 9.

Required: Formulate LPM for the problem

Solution
Decision variables
The model of this problem consists of three decision variables representing the amount of each
ingredient in the mixture.
X1 = number of lb of mercury in a batch
X2 = number of lb of zinc in a batch
X3 = number of lb of potassium in a batch

Objective function
Minimize Z = 4X1 + 8X2 + 9X3
Constraints
X1  200 lb… specification 1
X2  300 lb… specification 2
X3  100 lb… specification 3
Finally, the sum of all ingredients must equal 1000 pounds.
X1 + X2 + X3= 1000

The complete LPM of the problem is:

Minimize Z = 4X1 + 8X2 + 9X3
Subject to:
X1  200
X2  300
X3  100
X1 + X2 + X3= 1000
X1 , X2 ,X30

Activity 2: Attempt the following problems

Supermarket store chain has hired an advertising firm to determine the types and amount of
advertising it should have for its stores. The three types of advertising available are radio and
television commercials, and news papers advertisements. The retail chain desires to know the
number of each type of advertisement it should purchase in order to maximize exposure. It is
estimated that each ad or commercial will reach the following potential audience and cost the
following amount.
Type of Advertisement Exposure(people per ad or commercial) Cost
Television commercial 20,000 Birr 15,000
Radio commercial 12,000 6,000
News paper advertisement 9,000 4,000

The company must consider the following resource constraints.

a) The budget limit for advertising is Birr 100,000

27

Ashenafi A. (PhD Research Fellow) Page 27

 The television station has time available for 4 commercials.
The radio station has time available for 10 commercials.
The news paper has space available for 7 ads.
The advertising agency has time and staff available for producing no more than a total of
15 commercials and/or ads.
Required: formulate the LPM of the problem

Solution:
Maximize Z= 20,000X1 + 12,000X2 + 9,000X3
Subject to:
15,000X1 + 6,000X2 + 4,000X3  Birr 100,000
X1 4
X2 10
X3  7
X1 + X2 +X3 15
X1, X2, X3  0
The manager of a department store in Addis Ababa is attempting to decide on the types and
amounts of advertising the store should use. He has invited representatives from the local
radio station, television station, and newspaper to make presentations in which they describe
their audiences.
a. The television station representative indicates that a TV commercial, which costs Birr 15,000,
would reach 25,000 potential customers. The breakdown of the audience is as follows.

Male Female
Old 5,000 5, 000
Young 5,000 10,000

b. The news paper representative claims to be able to provide an audience of 10,000 potential
customers at a cost of Birr 4000 per ad. The breakdown of the audience is as follows.

Male Female

Old 4,000 3,000

Young 2,000 1,000

c. The radio station representative says that the audience for one of the station‘s commercials,
which costs Birr 6000, is 15,000 customers. The breakdown of the audience is as follows.

Male Female

Old 1,500 1,500

Young 4,500 7,500

28

Ashenafi A. (PhD Research Fellow) Page 28

The store has the following advertising policy:
Use at least twice as many radio commercials as news paper ads.
Reach at least 100,000 customers
Reach at least twice as many young people as old people
Make sure that at least 30% of the audience is women.

Available space limits the number newspaper ads to 7. The store wants to know the optimal
number of each type of advertising to purchase to minimize total cost.

Required: Formulate appropriate linear programming model.

Solution:
Step 1.Identify the decision variables
Let X1=represents advertisement on Television
Let X2=represents advertisement on Radio
Let X3=represents advertisement on Newspaper

Step 2.Determine the objective functions

Minimize Z = 15000x1 + 4000x2 + 6000x3

Step 3.Identify the constraints and the parameters

Reach twice as many radio commercials as newspaper ads
X3 − 2x2 ≥ 0
Reach at least 100,000
25000x1 + 10000x2 + 15000x3 ≥ 100000
Reach at least twice as many young people as old people
− 5000x1 − 11000x2 + 6000x3 ≥ 0
Make sure that at least 30% of the audience is women
7500x1 + 1000x2 + 4500x3 ≥ 0
Available space limits number on newspaper ad
X2≤7

Step 4. Build and validate the model

Minimize Z = 15000x1 + 4000x2 + 6000x3
subject to:
x3 − 2x2 ≥ 0 (a)
25000x1 + 10000x2 + 15000x3 ≥ 100000(b)
5000x1 − 11000x2 + 6000x3 ≥ 0 (c)
7500x1 + 1000x2 + 4500x3 ≥ 0(d)
x2 ≤ 7(e)
x1, x2, x3 ≥ 0

An electronic company produces three types of parts for automatic machines. It purchases
casting of the parts from the local foundry and then finishes parts on drilling, shaping and
polishing machines.

29

Ashenafi A. (PhD Research Fellow) Page 29

 The selling prices of parts A, B, and C respectively are Birr8, 10, 14. All parts made can be sold.
Casting for parts A, B and C respectively costs Birr 5, 6 and 10. The shop possesses only one of
each type of machine. Costs per hour to run each of the three machines are Birr 20 for drilling,
30 for shaping, and 30 for polishing. The capacities (parts per hour) for each part on each
machine are shown in the following table:

Machine Capacity per hour

Part A Part B Part C

Drilling 25 40 25
Shaping 25 20 20
Polishing 40 30 40

The management of the shop wants to know how many parts of each type it should produce per
hour in order to maximize profit for an hour‘s run. Formulate this problem as an LP model so as
to maximize total profit to the company.

Answer: Try yourself

2.5.SOLVING LINEAR PROGRAMMING MODELS

Following the formulation of a mathematical model, the next stage in the application of LP to
decision making problem is to find the solution of the model. An optimal, as well as feasible
solution to an LP problem is obtained by choosing from several values of decision variables
x1,x2…xn , the one set of values that satisfy the given set of constraints simultaneously and also
provide the optimal ( maximum or minimum) values of the given objective function. The most
common solution approaches are to solve graphically and algebraically the set of mathematical
relationships that form the model, thus determining the values of decision variables.

2.5.1. GRAPHICAL LINEAR PROGRAMMING METHODS

Graphical linear programming is a relatively straightforward for determining the optimal solution
to certain linear programming problems involving only two decision variables. Although graphic
method is limited as a solution approach, it is very useful in the presentation of LP, in that it
gives a ―picture‖ of how a solution is derived thus a better understanding of the solution.
Moreover, graphical methods provide a visual portrayal of many important concepts.

In this method, the two decision variables are considered as ordered pairs (X1, X2), which
represent a point in a plane, i.e, X1 is represented on X-axis and X2 on Y-axis.

Graphical method has the following advantages:

It is simple
It is easy to understand, and
It saves time.

30

Ashenafi A. (PhD Research Fellow) Page 30

Example 1: ABC Company is to start a production of two spare parts. The production manager
would like to produce so many parts to earn per unit profit of Birr 60 from the sale of type I and
Birr 50 from the sale of type II parts. However, the total available time to assemble the items, to
inspect them and the space for their storage is limited to 100 hours, 22 hours and 39 square feet
respectively. Four hours of assembly time, two hours of inspection time and three square feet of
storage space is required to produce each unit of type I spare part. On the other hand 10 hours of
assembly time, 1 hour of inspection time and 3 square feet of storage space is needed to produce
each unit of type II spare parts. How many of each item should the Company produce and sale to
maximize its profit? Use the graphical approach.

Solution
Graphical solution method involves the following steps

Step 1: Formulate the LPM of the problem

The model is then;
Maximize 60X1 + 50X2
Subject to: Assembly 4 X1 +10 X2≤ 100 hrs
Inspection 2 X1 +1X2 ≤ 22 hrs
Storage 3 X1 +3X2 ≤ 39 cubic feet
X1, X2≥ 0
Step 2. Plot each of the constraints on the graph
The step begins by plotting the non-negativity constraint, which restricts our graph only to the
first quadrant. We then can deal with the task of graphing the rest of the constraints in two parts.
For example, let us take the first constraint 4X1 + 10X2 < 100 hrs. First we treat the constraint as
equality: 4X1 + 10X2 = 100. Then identify the easiest two points where the line intersects each
axis by alternatively equating each decision variable to zero and solving for the other: when
X1=0, X2 becomes 10 and when X2=0, X1 will be 25. We can now plot the straight line that is
boundary of the feasible region as we have got two points (0, 10) and (25,0).

Step 3. Identify the feasible region

The feasible region is the solution space that satisfies all the constraints simultaneously. It is the
intersection of the entire region represented by all constraints of the problem. We shade in the
feasible region depending on the inequality sign. In our example above, for all the constraints

31

Ashenafi A. (PhD Research Fellow) Page 31

except the non-negativity constraint, the inequality sign is ‗less than or equal to‘ and it
represents region of the plane below the plotted lines.

Step 4. Locate the optimal solution

The feasible region contains an infinite number of points that would satisfy all the constraints of
the problem. Point that will make the objective function optimum will be our optimal solution.
This point is always found among the corner points of the solution space. Once the constraints
are plotted and feasible region is determined, we use either of the two graphic methods of

Graphic approach to find a solution for LP model consisting of only two decision variables:
The extreme point enumeration method
The objective (Iso-profit or cost) function line approach

a) The Objective Function Approach

This approach involves plotting an objective function line on the graph and then using that line to
determine where in the feasible solution space the optimal point is. We use the same logic as
plotting a constraint line except that it is not an equation until we equate it to some right hand
side quantity. Any quantity will do till we find a line that would last touch the feasible solution
space.

The optimal solution to an LP problem will always occur at a corner point because as the
objective function line is moved in the direction that will improve its value (e.g., away from the
origin in our profit maximization problem), it will last touch one of these intersections of
constraints. Then we determine which two constraints intersect there (in our case inspection and
storage constraints) and solve the equations simultaneously to obtain the mix of the two decision
variables that gives the value of the objective function at the optimum. Simultaneously solving
inspection and storage equations, we find the quantity of type 1 microcomputer to be produced
(X1) = 9 and that of type 2 (X2) = 4 giving the maximum profit of 60(9) + 50(4) = Birr740

32

Ashenafi A. (PhD Research Fellow) Page 32

Note: The maximization problems the movement of the iso-lines is outward from the origin;
while for minimization the movement is inward to the origin.

b) The Extreme(Corner) Point Approach

Corner or extreme point graphic method states that for problems that have optimal solutions, a
solution will occur at the corner point in the case of unique solution, while in the case of multiple
solutions, at least one will occur at a corner point as these multiple solutions will be
combinations of those points between two corner points. The necessary steps for this approach is
after graphing the problem, we determine the values of the decision variables at each corner
point of the feasible region either by inspection or using simultaneous equations. We then
substitute the values at each corner point into the objective function to obtain its value at each
corner point and select the one with the highest value of the objective function (for a maximum
problem) or lowest value (for a minimum problem) as the optimal solution.

Extreme Method of Solving Maximization Problems with ≤ constraint

Maximization of objective function involves finding the point where the combination of products
results in maximum value of objective function. The constraints are connected with < sign. The
solution space lies below the slant line and is bounded by the line segments. The origin and other
points below the slant lines are in the solution space (i.e., feasible region).

How to find corner points for the above LPM

Corner Points Coordinates Values

A (0, 10)
D (11, 0)______

33

Ashenafi A. (PhD Research Fellow) Page 33

Corner points b& c should be found by applying the simultaneous equations for the lines
crossing each other at points b & c as follows:

At corner point b, the two lines crossing each other are 4X1+10X2<100 (Assembly time) and
3X1+3X2 <39 (Storage space), hence by applying simultaneous equation, the corner point
values: 4x1 + 10x2 = 100
3x1 + 3x2 = 39

4X1+10X2═100 (3)
3X1+3X2 ═39 (-4)

12X1+30X2═300
-12X1-12X2═ -144
18X2═ 156 18x2=156
18=18

X2=8

Then, Substitute this value into either of the two of constraints to find X1 value as follows:
4X1+10X2═100
4X1+10(8)═100
4X1+80═100
4X1 ═100-80
4X1 ═20
4X1/4 ═20/4
X1═5
Therefore, at corner point b the coordinate values are (5, 8)

At corner point b, the two lines crossing each other are 2X1+X2<22 (Assembly time) and
3X1+3X2 <39 (Storage space), hence by applying simultaneous equation, the corner point
values:
2x1 + x2 = 22
3x1 + 3x2 = 39
2x1 + x2 = 22(−3)
3x1 + 3x2 = 39(1)

-6X1-3X2=-66
3X1+3X2 =39
-3X1 =-27

-3X1 =-27
-3 =-3
X1=9

34

Ashenafi A. (PhD Research Fellow) Page 34

 Then, substitute this value into either of the two of constraints to find X2 value as follows:
2X1+X2=22
2(9)+X2=22
18 + X2=22
X2=22-18
X2=4
Therefore, at corner point b the coordinate values are (9, 4)

The next step is to multiply all coordinates by the objective function and select the maximum
value as an optimal solution.

Corner Point Maximize Z=60X1+50X2 Value

Coordinates
A (0,10) 60(0)+50(10) =$500
B (5,8) 60(5)+50(8) =$700
C (9,4) 60(9)+50(4) =$740*
D (11,0) 60(11)+50(0) =$660
Therefore, Maximum Z=$740
X1=9
X2=4

Using this method for our example, simultaneously solving for corner points a, b, c, and d, we
find corresponding profit values of $500, $700, $740, and $660, respectively giving us the same
solution as the above one at C. Therefore, the optimal solution is x1= 9 units and x2 = 4 units
while the optimal value of objective function is 740.

2.5.1.1.SLACK VERSUS SURPLUS

Slack is the amount of a scarce resource that is unused by a given solution. Slack can potentially
exist in a < constraint. Slack variables are considered in the objective function by using a
coefficient of zero for each of them. When all the constraints are written as equalities after
adding a slack variable to each of them, the linear program is said to be in standard form. For
example, in the Assembly constraint 4X1 +10X2 < 100 hrs, the slack value is 100 – [4(9)
+10(4)] = 24.

Surplus on the other hand is the amount by which the optimal solution causes a > constraint to
exceed the required minimum amount. It can be determined in the same way that slack can, i.e.,
substitute the optimal values of the decision variables into the left side of the constraint and
solve. The difference between the resulting value and the original right hand side amount is the
amount of surplus. Surplus should also be accounted for in the objective function by using
coefficients of zero likewise.

Continuing for the above problem, we should have compute the slack values as the inequality
signs of all the constraints are ≤. Slack can potentially exist in a ≤ constraint. Slack variables are
considered in the objective function by using a coefficient of zero for each of them. When all the

35

Ashenafi A. (PhD Research Fellow) Page 35

constraints are written as equalities after adding a slack variable to each of them, the linear
program is said to be in standard form.

Assembly Time
4X1+10X2═100
4(9)+10(4)═100
36+40═100
76<100
Therefore, there are 24 units of slack value. In this case, the firm has a total of 100 hours of
assembly time, however, due to the above decision of producing of 9 units of X1 and 4 units of
X2, the firm use only 76 hours of available time but 24 hours of unused time in assembly
department.

Inspection Time
2X1+X2=22
2(9) +(4)═22
18+4═22
22═22

Therefore, there is no slack value. In this case the firm has a total of 22 hours to inspect items in
inspection department and the firm consumes or uses all the available time effectively due to the
above decision of producing 9 units of X1 and 4 units of X2.

Storage space
3X1+3X2=39
3(9)+3(4)═39
27+12═39
39═39

Therefore, there is no slack value. In this case the firm has a total of 39 cubic feet of space to
store the two microcomputers and the firm uses all the available space effectively due to the
above decision of producing 9 units of X1 and 4 units of X2.

Interpretation:
For a firm to maximize its profit (740), it should produce 9 units of the Model I microcomputer
and 4 units of model II.

Extreme Method of Solving Minimization Problems with ≥ constraints

Solving minimization problems involve where the objective functions (like cost function) will be
minimum. The constraints are connected to RHS values with > sign. But, in real world problems,
mixes of constraint is also possible. The value of RHS involves the lowest value of the
constraint. The solution space lies above the slant lines and it is not enclosed. It extends
indefinitely above the lines in the first quadrant. This means simply that cost increases without
limit as more and more units are produced. The minimum cost will occur at a point along the
inner boundary of the solution space. The origin and other points below the lines are not in the
solution space.

36

Ashenafi A. (PhD Research Fellow) Page 36

 Therefore, Minimum Z=$0.38
X1=25/11
X2=24/11

Example2 : Solve the following LP model using the graphic approach:

Minimize cost: 0.1x+0.07y
Subject to:
6x + 2y  18
8x+ 10y 40
1
x, y  0
Solution
Plot the constraints using this procedure: Treat each constraint as equality.
For the first two constraints, find the x intercept by setting y equal to 0 and solving for x, and
then find y intercept by setting x equal to 0 and solving for y. For the third constraint, y is simply
equal to 1.
Plot these constraints and shade in the feasible solution space as shown below:

Figure 2.2. Feasible region of constraints

The extreme points can be determined either by inspection or from simultaneous equations. The
results are summarized as follows:

37

Ashenafi A. (PhD Research Fellow) Page 37

 Extreme points How determined? Value of the objective function
X Y
0 9 Observation 0.10(0)+0.07(9)=Birr 0.63
5 0 Observation 0.10(5)+0.07(0)=Birr 0.50
2.27 2.19 Simultaneous 0.10(2.27)+0.07(2.19)=Birr0.38 (smallest)
equation
The minimum value of the objective function is Birr 0.38, which occurs when x=2.27 and
y=2.19.

Surplus
If there is a difference between the minimum required amount and the amount used, the
difference is called surplus. Surplus is the amount by which the optimal solution causes a 
constraint to exceed the required minimum amount. It can be determined in the same way that
slack is calculated. Substitute the optimal values of the decision variables into the constraint and
solve. The difference between the resulting value and the original right-hand side amount is the
amount of surplus. Surplus can potentially occur in a  constraint.

Computing the amount of surplus

Constraints Minimum Amount used Surplus=Amount used minus
requirement x = 2.27, y = 2.19 minimum requirement
1 18 6*2.27+2*2.19=18 18-18=0
2 40 8*2.27+10*2.19=40 40-40=0
3 1 2.19 2.19-1=1.19

Interpretation: The value of x is 2.27 units and the value of y is 2.19 units to attain a minimum
cost of Birr 0.38. There is a surplus of 1.19 units in the third constraint.

Activity 3: Solving an LP model using the graphic approach

Given the LP model:
Maximize profit: 10x1 + 16x2
Subject to:
10x1 + 20x2  110
25x1 + 20x2  200
x1, x2  0
Find the optimal solution using the graphic method.
Determine the amount of slack for each constraint.
What is the value of the objective function at the optimal solution?
What is the maximum available resource for each constraint?

Comments: Plot each of the constraints and determine the region or area that contains all of the
points that satisfy the entire set of constraints. Then you can determine the optimal solution using
the corner point method. The optimal solution should contain the values of decision variables,
slacks and maximum profit.

38

Ashenafi A. (PhD Research Fellow) Page 38

 Consider the following LPM.
Minimize Z = 0.1x+0.07y
Subject to:
6x+2y ≥ 18
8x+10y ≥ 40
y≥1
x,y > 0
Required: Find the values of x and y which makes the objective function minimum? Use
graphical approach.

Answer:
Corner point Coordinates Minimize Z=0.1x+0.07y
X Y
A 0 9 0.63
B 25/11 24/11 0.38*
C 15/4 1 0.445
Therefore, minimum Z=0.38 where X=25/11 and Y=24/11

2.5.1.2.SPECIAL CASES IN GRAPHICAL

SOLUTIONS a) Unboundedness
Unboundedness occurs when the decision variable increased indefinitely without violating any of
the constraints. The reason for it may be concluded to be wrong formulation of the problem such
as incorrectly maximizing instead of minimizing and/or errors in the given problem. Checking
equalities or rethinking the problem statement will resolve the problem.

Example: Maximize Z = 10X1 + 20X2

Subject to 2X1 + 4X2 > 16
X1+5X2>15
X1, X2 >

Following the above listed steps of graphical solution method, we find the following graph for
the above model:

Solution

The shaded area represents the set of all feasible solutions and as can be seen from the graph, the
solution is unbounded.

39

Ashenafi A. (PhD Research Fellow) Page 39

 b) Redundant Constraints
In some cases, a constraint does not form a unique boundary of the feasible solution space. Such
a constraint is called a redundant constraint. A constraint is redundant if its removal would not
alter the feasible solution space. Redundancy of any constraint does not cause any difficulty in
solving LP problems graphically. Constraints appear redundant when it may be more binding
(restrictive) than others.

Example: Maximize Z=2x1+3x2

Subject to:
2x1+x2≤20
2x1+3x2≤60
X2≤ 25

X1, x2 ≥0

Solution

c) Infeasibility
In some cases after plotting all the constraints on the graph, feasible area (common region) that
represents all the constraint of the problem cannot be obtained. In other words, infeasibility is a
condition that arises when no value of the variables satisfy all the constraints simultaneously.
Such a problem arises due to wrong model formulation with conflicting constraints. For
example,

Maximize Z = 3X1+2X2
Subject to: 2X1 + X2 ≤ 2
3X1 + 4X2 ≥12
X1, X2≥0

Solution

40

Ashenafi A. (PhD Research Fellow) Page 40

 d) Multiple optimal solutions
Recall the optimum solution is that extreme point for which the objective function has the largest
value. It is of course possible that in a given problem there may be more than one optimal
solution.

There are two conditions that should be satisfied for an alternative optimal solution to exist:
The given objective function is parallel to a constraint that forms the boundary of the feasible
region. In other words, the slope of an objective function is the same as that of the constraint
forming the boundary of the feasible region; and

The constraint should form a boundary on the feasible region in the direction of optimal
movement of the objective function. In other words, the constraint should be an active constraint.
Note: The constraint is said to be an active or binding or tight, if at optimality the left hand side
equals the right hand side. In other words, an equality constraint is always active. An inequality
sign may or may not be active.

For example
Maximize Z = 8X1+16X2
Subject to: X1 + X2 ≤200 …….C1
3X1 + 6X2 ≤ 900 ……C2
X2 ≤125 ……. C3
X1,X2≥0

Solution

In the problem above, using extreme point method and solving for values of corner points
simultaneously, the objective function assumes its maximum value of 2,400 at two corner points
B (50,125) and C (100,100). Therefore, the optimal solution is found on the line segment
connecting the two corner points. One benefit of having multiple optimal solutions is that for
other (perhaps qualitative) reasons, a manager may prefer one of them to the others, even though
each would achieve the same value of the objective function. In practical terms, one of the two
corner points is usually chosen because of ease in identifying its values.

Other corner points are computed by identifying the two corner points and keeping the value of
the objective function constant, where 50  X1  100 and 100  X1  125. By selecting one
value for one of the decision variables from the domain, we determine the value for another
decision variable keeping the objective function‘s value. Let X1 = 80.

41

Ashenafi A. (PhD Research Fellow) Page 41

Then, Z = 8X1+16X2  2400 = 8(80) +16X2
16X2 = 1760
X2 = 110
Activity 4: Identifying special issues in graphic solution of an LP model
Does the following LP involve infeasibility (no-feasibility solution), unbounded, and/or
alternative optimal solutions? Explain.
Maximize profit: 4x1 + 8x2
Subject to:
2x1 + 2x210
-x1 + x2 8
x1 , x2 0
Maximize: 2x1 + x2
Subject to:
x1 ≥ 2
x2  5
x1 ,x2 ≥0

Comments: You can draw the graph of the constraints and check if the solution is infeasible,
unbounded and alternative. You can refer 2.2.7 of this module to refresh about special issues of
the graphic approach of solving an LP model.

2.5.1.3.Mix of constraints
It is only in some cases that the maximization problems consist of constraints connected to the
RHS value with only ≤. By the same token, constraints of minimization problems are not always
connected to their RHS with ≥. This is to mean that both maximization and minimization
problems may consist of constraints connected to RHS with a mix of algebraic signs ( <, >, =).

Example
Minimize Z=1500x+2400y
Subjected to:
4x+Y>24 …I
2x+3y>42 …II
X+4y>36 …III
X<14 …IV
y<14 …V
x, y>0

42

Ashenafi A. (PhD Research Fellow) Page 42

 Solution:

Constraints Coordinates Objective functions

Corner intersected X Y MinZ=1500x+2400y
points
A I&V 2.5 14 $37,350
B IV&V 14 14 $54,600
C III&IV 14 5.5 $34,200
D II&III 12 6 $32,400
E I&II 3 12 $33,300

The minimum value of the objective function is 32,400. Therefore, the optimum solution is
found at the corner point D, where constraint II&III are intersected. Hence, optimal solution: x
=12, y = 6.

2.5.2.THE SIMPLEX METHOD OF SOLVING LP PROBLEMS

The previous section illustrated how two-variable LP problems could be solved graphically.
However, in practice, graphical solutions do not represent a realistic alternative to most LP
problems because the majority of these problems involve more than two decision variables,
whereas the graphical method is limited to problems with two decision variables. This section

43

Ashenafi A. (PhD Research Fellow) Page 43

introduces the simplex procedure, which is a general purpose approach for solving LP problems
regardless of the number of decision variables. The simplex procedure will be demonstrated for a
maximization and minimization problems.

2.5.2.1.Overview of the Simplex Method

In the previous discussion, we discussed how the graphical solution procedure can be used to
solve linear programming problems involving two decisions variables. However, most linear
programming problems are too large to be solved graphically, and an algebraic solution
procedure must be employed. The most widely used algebraic procedure for solving linear
programming problems is called the simplex method. The simplex procedure is a general
purpose approach that can be used in any LP model regardless of the number of decision
variables. This procedure is discussed in different conditions: Maximization problems having
only  constraints, minimizations and maximizations with mixed constraints (having some
constraints that are not  type).

The simplex method is an iterative technique that begins with a feasible solution that is not
optimal, but serves as a starting point. Through algebraic manipulation, the solution is improved
until no further improvement is possible (i.e., until the optimal solution has been identified).

We learned in the previous section that the optimal solution to a LP model will occur at an
extreme point of the feasible solution space. This is true even if a model involves more than two
variables; optimal solutions occur at these points. Extreme points represent intersections of
constraints. Of course, not every intersection will result in an extreme point of the feasible
solution space; some will be outside of the feasible solution space. Hence, not every solution will
be a feasible solution. Solutions which represent intersections of constraints are called basic
solutions; those which also satisfy all the constraints, including the non-negativity constraints,
are called basic feasible solutions. The simplex method is an algebraic procedure for
systematically examining basic feasible solutions. If an optimal solution exists, the simplex
method will identify it.

2.5.2.2.Steps in the simplex procedure for a maximization LP problem

The simplex procedure for a maximization problem with all  constraints consists of the
following steps.
Write the LP model in a standard form: When all of the constraints are written as
equalities, the linear program is said to be in standard form. We convert a maximization LP
model into a standard form by applying the slack variables, s, which carries a subscript that
denotes which constraint it applies to. For example, s1 refers to the amount of slack in the first
constraint, s2 to the amount of slack in the second constraint, and so on. When slack variables are
introduced to the constraints, they are no longer inequalities because the slack variable accounts
for any difference between the left and right-hand sides of an expression at optimal solution.
Hence, once slack variables are added to the constraints, they become equalities. Furthermore,
every variable in a constraint must be represented in the objective function and thus, slacks have
to be represented in the objective function. However, since slack does not provide any real
contribution to the objective, each slack variable is assigned a coefficient of zero in the objective
function.

44

Ashenafi A. (PhD Research Fellow) Page 44

Slack variables are variables we introduce to the left side of  constraints to make them an
equality. These variables represent the amount of unused scarce resources or capacity. Taking
the aforementioned microcomputer problem, the LP model is:
Maximize Z: 60x1 + 50x2
Subject to:
4x1 + 10x2  100
2x1 + x2  22
3x1 + 3x2  39
x1, x2 0
Its standard form is as follows:
Maximize Z= 60x1 + 50x2 + 0s1 + 0s2 + 0s3
Subject to:
4x1 + 10x2 + s1 = 100
2x1 + x2 + s2 = 22
3x1 + 3x2 + s3 = 39
x1, x2, s1, s2, s3  0

The constraints form a system of simultaneous linear equations. As a rule, linear programming
models have fewer equations than variables. For example, in this case there are three equations
and five variables (two decision variables and three slack variables). When a system of equations
has more variables than equations, unique solutions are not possible. Instead the solution
possibilities are infinite. Thus, unless the number of equations equals the number of variables, a
unique solution cannot be found.

In order to make the number of variables equal to the number of equations, enough variables can
be set equal to zero, so that the remaining number of nonzero variables equals the number of
equations. In the case of the above LP model, if two of the variables were set equal to zero, the
result would be three equations and three variables. In general, if we have n variables and m
equations, with n>m, we must set n-m of the variables equal to zero in order to obtain a system
of equations in which the number of variables equal the number of equations. And if we then
solve the resulting system of equations for the values of the remaining variables, we will obtain a
solution that represents an intersection of constraints. In other words, by setting n-m variables
equal to zero, we are able to determine a solution that occurs at one of the constraint
intersections. The result is a basic solution. Moreover, the variables that have been set equal to
zero are referred to as non-basic variables, whereas the remaining variables are referred to as
basic variables.

Activity 6: Writing an LP model in standard form

Consider the following linear program:
Maximize profit: x1 + 2x2
Subject to :
x1 + 5x2  10
2x1 + 6x2  16
x1 , x2≥0

45

Ashenafi A. (PhD Research Fellow) Page 45

Required:
Write the LP model in standard form.
In reality, how do you define slack (idle resources)? Give examples.
Why should you add slacks?
What is the coefficient of slacks in the objective function? Why?

Comments: Add slack variables in the constraint function and the inequality will become an
equation. Moreover, the slack variables should be represented in the objective function with zero
coefficients as they contribute nothing to the objective of a firm. Slacks are also assumed to be
non-negative like decision variables.

2.Finding An Initial Feasible Solution

A basic feasible solution is a basic solution that satisfies all of the constraints in a model,
including the non-negativity constraints. As noted in the preceding section, we can obtain a basic
solution by setting n-m of the variables equal to zero and solving for the remaining variables. For
example, in the microcomputer problem (five variables, three constraints), we must set two
variables equal to zero. Suppose we set the two decision variables equal to zero. They would
drop out of the equations, and we would be left with the following very simple equations:

s1 = 100
s2 = 22
s3 = 39
Obviously, it is very easy to see what the values of the remaining variables are. Moreover, we
can verify that this is a basic feasible solution because the equations are satisfied and no variable
has a negative value. Hence, an easy way for us to obtain a basic feasible solution in this case
was to set the decision variables equal to zero. In fact, in a maximization problem in which all
constraints are of the  variety, an initial basic feasible solution can be obtained by setting all of
the decision variables equal to zero. Therefore, the initial solution will consist entirely of slack
variables.

The variables that are solved for are the basic variables, and the other variables are non-basic
variables. The complete solution consists of the basic and the non-basic variables. Therefore, for
the microcomputer problem, the complete initial solution is:
X1= 0
X2= 0
S1=100
S2 =22
S3= 39

3.Develop The Initial Simplex Tableau: The mathematical characteristics of each extreme
point are summarized in a table which is referred to as a tableau. The simplex procedure starts
checking optimum solution from the origin, where the value of all the decision variables is zero.
To develop the initial simplex tableau;
List the variables across the top of the tableau and write the objective function coefficient of
each variable just above it.

46

Ashenafi A. (PhD Research Fellow) Page 46

 Provide one row for each constraint in the body of the tableau. List each basic variable in the
basis column, one per row.
In the Cj column, enter the coefficient of the objective function (zero for each slack variable),
where Cj is the coefficient of variable j in the objective function.
Compute values for row Zj: For each column, the Zj value is obtained by multiplying each of
the numbers in the column by their respective row coefficients in column Cj. The row Zj
values are determined column by column. For example, the Zj value for x1 column is
computed in this way:
0(4) +0(2) +0(3) =0
Similarly, the Zj value for x2 column is:
0(10) +0(1) +0(3) =0
For s1 column, the result is:
0(1) +0(0) +0(0) =0

Because each of the row coefficients is 0 means that all the row values will equal zero for the
initial tableau. There is also a Zj value for the quantity (right hand side value) column. It is
computed in similar fashion:
0(100) +0(22) +0(39) =0
Computer values for Cj – Zj. For each column, the value in row Zj is subtracted from the Cj
value in the top row. Thus, for the x1 column we have 60-0=60, for the x2 column we have 50-
0=50, and for each of the slack columns we have 0-0=0.

The simplex tableau in the simplex procedure is equivalent to the solution space and the
checking process is done for each corner points. The corner points represent the intersection of
two constraints (i.e., the coordinate value for decision variables at the intersection point). The
variables in the basis of a simplex tableau are like variables at the intersection point. The value
for these variables at this point is found from the quantity column of the tableau. All constraints
must also be represented in the simplex tableau if the simplex procedure is to yield a solution
that takes all the constraints into account.The simplex procedure starts checking from the origin,
where the value of all the decision variables is zero. If we start from the origin by assuming a
value of zero for all the decision variables, and if all the constraints are to be represented in the
tableau, how can we develop the initial simplex tableau?

The initial simplex tableau represents all the constraints in the model without violating the
assumption of starting from the origin through the help of other variable(s). These variables are
slack variables. Look at the basis in the following simplex tableau.

Initial simplex tableau

Cj 60 50 0 0 0 RHS
Basis x1 x2 s1 s2 s3 Quantity
s1 0 4 10 1 0 0 100
s2 0 2 1 0 1 0 22
s3 0 3 3 0 0 1 39
Zj 0 0 0 0 0 0 (profit)
Cj-Zj 60 50 0 0 0 0

47

Ashenafi A. (PhD Research Fellow) Page 47

Interpreting the Solution
The meaning of the above simplex tableau is that the solution to the model is s1 = 100, s2 =22,
s3 =39, x1 =0, x2=0 and profit =0.

Each tableau represents a basic feasible solution to the problem. Let us examine this particular
tableau to see some of the kinds of information it contains. As previously noted, this solution is
comprised entirely of the slack variables. We can read the variables that are in the basis, or ―in
solution‖ down the left side of the tableau, and their values down the right side. This tells us that
s1 = 100, s2 =22, s 3 =39. If the variable is not included in the basis, its value is equal to zero.
Therefore, because x1 and x2 are not listed, their values are zero. Hence, this solution
corresponds to the origin.

A feature of variables that are in solution is that a zero (0) appears in the bottom row of the
tableau in that variable‘s column. Thus, all three slack variables have a zero in their respective
columns in the bottom row. Conversely, the two variables not in solution both have non-zero
values in the bottom row. In addition, note the values in the column of every variable that is in
solution. In each case, the column is made up of a single 1 and the rest 0‘s. For example, in s 1
the values are
1


0 
0

This is called a unit vector. Variables that are in solution will have unit vector columns (i.e., one
and the rest 0). Moreover, the 1 will appear in the same row that the variable appears in.
Thus, s1 is the first variable listed in the basis. Hence, a 1 appears at the intersection of the first
row and s1 column.

The last value in the Zj row (i.e., in the quantity column) indicates that the value of the objective
function is 0 for this solution.

The values in the bottom row indicate any potential for improvement. For example, the 60 in
column x1 reveals that if one unit of x1 could be brought into the solution (i.e., if x1=1), the
value of the objective function would increase by Birr 60, and that additional units of x1 also
would increase the value of the objective function by Birr 60 per unit. The other values in the
bottom row have similar interpretations.

Then our next task is to check if the solution is optimal to the model. This will be proved by
looking at the net evaluation row (Cj-Zj) row in the tableau. The rule is:

A Simplex solution in maximization problem is optimal if the Cj-Zj row consists entirely of
zero and negative numbers (i.e., there are no positive values in the bottom row). That is,
Cj-Zj≥0

In our example, there are two positive numbers (60 and 50). This indicates that the initial
simplex tableau doesn‘t give us the optimal solution to the model.

48

Ashenafi A. (PhD Research Fellow) Page 48

4. Develop subsequent tableaus
To develop subsequent tableaus, we go through the following steps:
Identify the variable that has the largest positive value in row Cj-Zj. This is the next
variable to come into the solution basis. The largest positive value in the above example is 60
because it has the largest profit potential. The column containing this largest positive value is
called the pivot column; and the variable in the pivot column is the entering variable to bring
about the improvement in the level of the objective function. The x1 column is designated as the
pivot column, and the numbers in that column in the body of the table will be key values for
developing the second tableau. 4
 
2 
3


Let us see why the case is. The numbers in the body of the first tableau are the coefficients of the
constraints. Thus, in the x1 column, the numbers are 4, 2, and 3, which are the x1 coefficients in
the three constraints. Each number indicates how much of the basic variable for that row must be
given up, or reduced, in order to get one unit of x1. For instance, the 4 in the first row tells us that
4 units of slack, s1, will be required to obtain one unit of x1. Similarly, the 2 in the second row
indicates that 2 units of slack, s2, must be used up to obtain the same one unit of x1 and the 3 in
the third row tells us that 3 units of s3 must be used up to get the unit of x1. The numbers in the
body of a simplex tableau are, therefore, referred to as substitution rates because they indicate
how much each basic variable will be used in order to obtain 1 unit of the entering variable.

Because each unit of x1 that can be brought into solution will increase profit by Birr 60, we want
to make as much x1 as possible. The amount of x1 that can be made will depend on the
substitution rates and the amount of slack variables, which is shown in the quantity column.

Identify the leaving variable by applying the minimum ratio test.

Pick out each coefficient in the pivot column that is positive. Divide the right hand side
values for these coefficients in the same row. But, do not divide by a zero or negative
value.
Identify the row that has the smallest of these ratios and the basic variable for that row is
the leaving variable.
Replace the leaving variable by the entering basic variable in the solution basis column of
the next tableau.
In our example, we will take ratios: 100 , 22 and 33 which is 25, 11 and 13.
4 2 3
First Tableau
Basis Cj 60 50 0 0 0
x1 x2 s1 s 2 s3 Quantity Minimum ratio test
s1 0 4 10 1 0 0 100 100/4 = 25
*
s2 0 2 1 0 1 0 22 22/2 = 11 pivot row
s3 0 3 3 0 0 1 39 39/3 = 13
Zj 0 0 0 0 0 0 Profit=0
Cj-Zj 60 50 0 0 0 0
*Pivot element

49

Ashenafi A. (PhD Research Fellow) Page 49

The second constraint is the most restrictive because it reveals that there is only enough of the
second resource to make 11 units of x1. Hence, the smallest of the ratios governs how much x1
we can make. Further, in making the 11 units of x1, the second resource will be used up (i.e., s2
will be reduced to zero). Thus, s2 will be drop out of the solution mix (basis) and its place will be
taken by x1. That is, x1 is the entering variable and s2 is the leaving variable.

It sometimes happens that some of the substitution rates are for the variable we want to bring
into solution are zero or negative. A zero would imply that none of the corresponding variables
in the basis is required to obtain a unit of the entering variable. A negative substitution rate
implies that bringing a variable into solution will increase the amount of a basic variable. In
effect, the entering variable frees up additional resources. The point is the negative and zero
substitution rates will not limit the amount of the entering variable that can be made. We,
therefore, need only to concern ourselves with positive substitution rates; there is no need to
divide the quantity column values by a negative or a zero substitution rate. Consequently, in
determining the variable that will leave the basis, our rule is:

Select the leaving variable as the one which has the smallest non-negative ratio of quantity
divided by substitution rate.
Up to this point we have determined that variable x1 will enter the basis, and that variable s2 will
leave the solution. The row of the variable that will leave the solution (in this case s2) is called
the pivot row. Its values will be used to obtain values in the same row of the next tableau.
Dear learner, having this, what do you think is the next step?
The next step is to find unit vectors for the new basic variable using row operations on the pivot
column. The row operation includes:

Interchanging any two rows.

Adding or subtracting a multiple of a non-zero constant of one row to another
Multiplying any row of by a non-zero constant.

Let us begin our computations with the revised values for new row x1. We want to force the 2
(called the pivot element); which is the intersection of x1 column and s2 row of the initial
tableau to 1. We can force the value to 1 by dividing the 2 by itself.

That is, given the second constraint as: 2x1 + x2 + 0s1 +1s2 +0s3 = 22
This equation can be transformed into an equivalent equation with a 1 as the x1 coefficient by
multiplying each element by ½.
Multiply by ½: x1 + 1/2x2 + 0s1 +1/2s2 +0s3 = 11
The revised coefficients can now be entered in the second tableau as shown below:

Basis Cj 60 50 0 0 0
x1 x2 s1 s2 s3 Quantity
s1 0
x1 60 1 1/2 0 1/2 0 11
s3 0
Zj
Cj-Zj

50

Ashenafi A. (PhD Research Fellow) Page 50

The new pivot row will be used to transform the values of the other row equations into
equivalent equations that can be used to easily identify the next solution by elementary row
operations. We can easily see how the first constraint can be revised so that the x1 coefficient is
0 by writing out the first constraint equation and the new pivot row equations:
First constraint: 4x1 + 10x2 + 1s1 +0s2 +0s3 = 100
Pivot row: x1 + 1/2x2 + 0s1 +1/2s2 +0s3 = 11

If we multiply the pivot row by 4 and subtract the result, the x1 will drop out of the equation and
become zero, which is what we want. The multiple of the pivot row is:
4x1 + 2x2 + 0s1 +2s2 +0s2 = 44

Subtracting this from the fist constraint gives us our new equation for the first constraint:
First constraint: 4x1 + 10x2 + 1s1 +0s2 +0s3 = 100
Pivot multiple: - (4x1 + 2x2 + 0s1 +2s2 +0s3 = 44)
0x1 + 8x2 + 1s1 -2s2 +0s3 = 56

We can now enter the coefficients of this transformed constraint equation in the first row of the
second tableau. A similar operation can be performed on the third constraint equation and a
partially completed second tableau is given as follows:
Basis Cj 60 50 0 0 0 RHS
x1 x2 s1 s2 s3 Quantity
s1 0 0 8 1 -2 0 56
x1 60 1 1/2 0 1/2 0 11
s3 0 0 3/2 0 -3/2 1 6
Zj
Cj-Zj

Now we are ready to compute the values for row Zj. These are found by multiplying the values
in each column by the corresponding coefficients in the Cj column and adding them. The
computations are as follows:

x1 x2 s1 s2 s3 Quantity
s1 0(0)=0 0(8)=0 0(1)=0 0(-2)=0 0(0)=0 0(56)=0
x1 60(1)=60 60(1/2)=30 60(0)=0 60(1/2)=0 60(0)=0 60(11)=660
s3 0(0)=0 0(3/2)=0 0(0)=0 0(-3/2)=0 0(1)=0 0(6)=0
Zj 60 30 0 30 0 660

Now the Cj-Zj values can be computed. Thus,

x1 x2 s1 s2 s3
Cj 60 50 0 0 0
Zj 60 30 0 30 0
Cj-Zj 0 20 0 0 0

51

Ashenafi A. (PhD Research Fellow) Page 51

 Completed second tableau
Basis Cj 60 50 0 0 0
x1 x2 s1 s2 s3 Quantity
s1 0 0 8 1 -2 0 56
x1 60 1 1/2 0 1/2 0 11
s3 0 0 3/2 0 -3/2 1 6
Zj 60 30 0 30 0 660
Cj-Zj 0 20 0 -30 0

Interpreting the Second Tableau

At this point, variables s1, x1 and s3 are in solution. Not only are they listed in the basis, they
also have a 0 n row Cj-Zj. The solution at this point is:
s1=56
x1 =11
s3=6
Note, too, that x2 and s2 are not in the solution. Hence, they are each equal to zero. The profit at
this point is Birr 660, which is read in the quantity column in row Zj. Also, note that each
variable in solution has a unit vector in its column.

Checking the bottom row of the tableau, we note that there is a positive number (i.e., 20 in the x2
column). This indicates that the solution can be improved. The implication is that for every unit
of x2 that we can bring into the solution, profit will increase by Birr 20. Therefore, we must
develop a third tableau.

Dear learner, please develop the third tableau by your own. Use the procedures used above.

Developing the Third Tableau

So far we have determined that variable x2 will enter the solution. Now we must identify the
variable that will leave the solution. To do so this, we divide quantity column values to the
corresponding constraint coefficients in the x2 column. These computations are shown below:

Basis Cj 60 50 0 0 0
x1 x2 s1 s2 s3 Quantity
s1 0 0 8 1 -2 0 56 56/8=7
x1 60 1 1/2 0 1/2 0 11 11/0.5=22
s3 0 0 3/2 0 -3/2 1 6 6/1.5=4 smallest non-
negative ratio
Zj 60 30 0 30 0 660
Cj-Zj 0 20 0 -30 0
The smallest non-negative ratio is in the third row. This tells us that variable s3 must be taken out
of the solution (i.e., set equal to zero), and replaced with incoming variable x2. Because x2 will
come into the solution, its coefficient in the x2 column must be transformed to a 1. As before, if
we note that the current value at that location (i.e., the intersection of the entering variable‘s
column and the entering variable‘s row) is 3/2, we can accomplish this by multiplying every

52

Ashenafi A. (PhD Research Fellow) Page 52

element in the third row by 2/3. The resulting values become the pivot row for the third tableau.
Omitting the variables from the list for convenience, we have:

2/3(0 3/2 0 -3/2 1 6)=0 1 0-1 2/3 4

These values now can be entered into the third tableau as shown below:
Basis Cj 60 50 0 0 0
x1 x2 s1 s2 s3 Quantity
s1 0
x1 60
x2 50 0 1 0 -1 2/3 4
Zj
Cj-Zj

The values in the remaining two constraint rows now must be transformed to equivalent
equations in which the elements in the x2 column are zero. This can be accomplished in the
following way:

Multiply the values in the pivot row by the value in the entering variable‘s column for the row
being transformed. Thus, to transform values in row 1, multiply the pivot values of the third
tableau by 8. Then subtract the results from the values that appeared in the preceding tableau.
Similarly, to obtain new row 2 values, multiply the pivot values in the third tableau by ½ and
subtract the result from the row 2 values of the preceding tableau.

For the first row, multiplying the pivot values by 8, we obtain these values:
8(0 1 0 -1 2/3 4)=0 8 0-8 16/3 32
Subtracting these from the previous row 1 values, we obtain

Previous values 0 8 1 -2 0 56
-(0 8 0 -8 16/3 32)
= 0 0 1 6 -16/3 24

For row 2, multiplying the pivot values by 1/2, we obtain these values:
1/2(0 1 0 -1 2/3 4)=0 1/2 0-1/2 1/3 2

Subtracting these from the previous row 2 values, we obtain

Previous values 1 1/2 0 1/2 0 11

- (0 1/2 0 -1/2 1/3 2)
= 1 0 0 1 -1/3 9

53

Ashenafi A. (PhD Research Fellow) Page 53

Partially completed third tableau
Basis Cj 60 50 0 0 0
x1 x2 s1 s2 s3 Quantity
s1 0 0 0 1 6 -16/3 24
x1 60 1 0 0 1 -1/3 9
x2 50 0 1 0 -1 2/3 4
Zj
Cj-Zj

Dear learner, please calculate the values of Zj and Cj-Zj by your own.
These values can be calculated as follows:
60 50 0 0 0
x1 x2 s1 s2 s3 Quantity
s1 0(0) =0 0(0) =0 0(1) =0 0(6) =0 0(-16/3)=0 0(24)=0
x1 60(1)=60 60(0)=30 60(0)=0 60(1)=60 60(-1/3)=-60/3 60(9)=540
X2 50(0) =0 50(1)=50 50(0)=0 50(-1)=-50 50(2/3)=100/3 50(4)=200
Zj 60 50 0 10 40/3 740
Cj-Zj 0 0 0 -10 -40/3

Completed third tableau

Basis Cj 60 50 0 0 0
x1 x2 s1 s2 s3 Quantity
S1 0 0 0 1 6 -16/3 24
X1 60 1 0 0 1 -1/3 9
X2 50 0 1 0 -1 2/3 4
Zj 60 50 0 10 40/3 740
Cj-Zj 0 0 0 -10 -40/3

Interpreting the Third Tableau

In this tableau, all of the values in the bottom row are either negative or zero, indicating that no
additional potential for improvement exists. Hence, this tableau contains the optimal solution,
which is: x1 = 9 units
x2 = 4 units
s1 = 24 hours
Note, too, that because they do not appear in the solution, variables s2 and s3 have values of zero.
Thus, in order to achieve the maximum daily profit, the company should produce 9 units of
model 1 and 4 units of model 2 microcomputers. This will leave no slack in either inspection or
storage constraints. However, there will be 24 hours of assembly time that is unused. The
maximum profit will be Birr 740.
Example 2: Solve the following LPM using simplex
method Maximize Z = 10x1+6x2+7x3
Subject to
Material 4x1+8x2+5x3  860
Labor 2x1+3x2+1x3  400
x1, x2, x 3  0

54

Ashenafi A. (PhD Research Fellow) Page 54

Solution
The standard form will be:
Maximize Z =10x1+6x2+7x3 +0S1 +0 S2
Subject to:
Material 4x1+8x2+5x3 + S1 = 860
Labor 2x1+3x2+1x3 + S2 = 400
x1, x2, x 3, S1, S2  0

Develop the initial Simplex tableau

The standard form for the model is as developed above. Therefore we can just go to the
representation of the standard form using the simplex tableau.

Initial tableau
C 10 6 7 0 0 RHS
Basis X1 X2 X3 S1 S2 values
S1 0 4 8 5 1 0 860
S2 0 2 3 1 0 1 400
Z 0 0 0 0 0 0
C-Z 10 6 7 0 0

C 10 6 7 0 0 RHS
Basis X1 X2 X3 S1 S2 values
S1 0 4 8 5 1 0 860 860 4 = 215
S2 0 2 3 1 0 1 400 400 2
= 200
Leaving
Z 0 0 0 0 0 0 variable
C-Z 10 6 7 0 0

Entering
variable
C 10 6 7 0 0 RHS
Basis X1 X2 X3 S1 S2 values
S1
X1 10 1 3/2 1/2 0 1/2 200
Z
C-Z

C 10 6 7 0 0 RHS
Basis X1 X2 X3 S1 S2 values
S1 0
X1 10 1 3/2 1/2 0 1/2 200
Z
C-Z

55

Ashenafi A. (PhD Research Fellow) Page 55

 2nd Simplex tableau

C 10 6 7 0 0 RHS
Basis X1 X2 X3 S1 S2 values
S1 0 0 2 3 1 -2 60 -4R2 +R1
X1 10 1 3/2 1/2 0 1/2 200
Z 10 15 5 0 10 2000 = Profit
C-Z 0 -9 2 0 -10

C 10 6 7 0 0 RHS
Basis X1 X2 X3 S1 S2 values
S1 0 0 2 3 1 -2 60
X1 10 1 3/2 ½ 0 1/2 200
Z 10 15 5 0 10 2000
C-Z 0 -9 2 0 -10

3rd Tableau

C 10 6 7 0 0 RHS
Basis X1 X2 X3 S1 S2 values
X3 7 0 2/3 1 1/3 -2/3 20
X1 10 1 7/6 0 -1/6 5/6 190 -1/2(R1) + R2
Z 10 70/6 7 4/9 11/3 2040 = Optimal profit
C-Z 0 -5.67 0 -4/9 -11/3

The fourth tableau is the optimal tableau because there is no any positive value in the net
evaluation row. This indicates that the optimal level of profit is achieved when: X1 = 20, X2 = 0,
X3 =190, S1 = 0, S2 =0 and S3 = 0. The maximum profit is 2,040.

Activity 7: Formulating and solving an LP model using the simplex method

A firm produces two lines of products, hand tools and home appliances. The unit contribution for
each product line has been computed by the company‘s accounting department as Birr 50 per
unit and Birr 60 per unit respectively. The time requirement of the products and total time
available in each assembly line is provided as follows:
Departments Hand tools Home appliances Available hours
1 2 hours 3 hours 1500 hours
2 3 hours 2 hours 1500 hours

In addition, the demand for the products restricts the production to the maximum of 400 units of
each of the two items. How much of each item must be produced if the company is to maximize
its profit? Use the simplex approach.

56

Ashenafi A. (PhD Research Fellow) Page 56

Solution: x1 = 300 units, x2 = 300, s1 = 0, s2 = 0, s3 = 100, and s4 = 100 and maximum profit is
Birr 33,000

2.5.2.3.MINIMIZATION LINEAR PROGRAMMING PROBLEMS (Big M-method )

The Big-M Method is a technique, which is used in removing artificial variables from the basis
.In this method; we assign coefficients to artificial variables, undesirable from the objective
function point of view. If objective function Z is to be minimized, then a very large positive price
(called penalty) is assigned to each artificial variable. Similarly, if Z is to be maximized, then a
very large negative cost (also called penalty) is assigned to each of these variables. Following are
the characteristics of Big-M Method:
High penalty cost (or profit) is assumed as M
M as a coefficient is assigned to artificial variable A in the objective function Z.
Big-M method can be applied to minimization as well as maximization problems with the
following distinctions:
i. Minimization problems
-Assign +M as coefficient of artificial variable A in the objective function Z
of the minimization problem.
ii. Maximization problems:
-Here –M is assigned as coefficient of artificial variable A in the objective
function Z of the maximization problem.
Coefficient of S (slack/surplus) takes zero values in the objective function Z
For minimization problem, the incoming variable corresponds to the highest negative value

The solution is optimal when there is no negative value of Cj-Zj.(For minimization LPP case)
The various steps involved in using simplex method for minimization problems are similar to the
maximization method except how to determine the Entering Variable and the optimal solution, as
shown below:
Step 1. Formulate the linear programming model, and express the mathematical model of
L.P. Step 2. Develop the standard format of the LPM
Step 3.Develop an initial solution is set up and check if it is optimal. The minimization tableau is
optimal if all Cj - Zj row values are ≥ 0.
Step 4.Further iterate towards an optimal solution if the current tableau is not optimal by
identifying the entering variable, leaving variable and the pivot element.
Determine the variable to enter the basic solution. To do this, we identify the column
with the largest negative value in the Cj - Zj row of the table.
Next we determine the departing variable from the basic solution. If an artificial
variable goes out of solution, then we discard it totally and even this variable may not
form part of further iterations. Same procedure, as in maximization case, is employed
to determine the departing variable.
Convert the pivot element into 1 and the remaining elements in the pivot column into
0.

57

Ashenafi A. (PhD Research Fellow) Page 57

Step 5. Develop the second tableau. We evaluate the entries for next simplex table in exactly the
same manner as was discussed earlier in the maximization case.
Step 6. Step (3—5) are repeated until an optimum solution is obtained.

So the following are the essential things to observe in solving for minimization problems:
The entering variable is the one with the largest negative value in the Cj-Zj row while the
leaving variable is the one with the smallest non-negative ratio.
The optimal solution is obtained when the Cj-Zj row contains entirely zeros and positive
values.

Example1: Assume the following minimization problem.

Minimize Z = 7X1+9X2
Subject to 3X1+6X2 ≥36
8X1+4X2 ≥64
X1, X2 ≥0

We introduce both surplus and artificial variables into both constraints as

follows. Minimize Z = 7X1+9X2+0S1+0S2+MA1+MA2
Subject to 3X1+6X2 –S1+A1 = 36
8X1+4X2-S2+A2 = 64
X1, X2 ≥0
So the subsequent tableaus for this problem are shown below. To remind in these tableaus is in
transforming from one tableau to another, we perform elementary row operations to obtain the
unit vector in the pivot column for the entering variable into the solution.

Initial Simplex Tableau

Cj 7 9 0 0 M M RHS
Basic V. X1 X2 S1 S2 A1 A2
A1 M 3 6 -1 0 1 0 36
A2 M 1 4 0 -1 0 1 64
Zj 11M 10M -M -M M M 50M
Cj-Zj 7-11M 9-10M M M 0 0

Second Simplex Tableau

Cj 7 9 0 0 M RHS
Basic V. X1 X2 S1 S2 A1
A1 M 0 9/2 -1 3/8 1 12
X1 7 1 ½ 0 -1/8 0 8
Zj 7 7/2+9/2M -M 3/8M-7/8 M 56+12M
Cj-Zj 0 11/2-9/2M M 7/8-3/8M 0

58

Ashenafi A. (PhD Research Fellow) Page 58

 Third Simplex Tableau
Cj 7 9 0 0 RHS
Basic V. X1 X2 S1 S2
X2 9 0 1 -2/9 1/12 8/3
X1 7 1 0 1/9 -1/6 20/3
Zj 7 9 -11/9 -5/12 212/3
Cj-Zj 0 0 11/9 5/12

The third tableau represents a final tableau since it is the optimal solution with entirely zeros and
non-negative values in the Cj-Zj row.

Therefore, the optimal solution is: X1 = 20/3 and X2 = 8/3 and value of objective function is
212/3.

Example 2: Solove the following LPM using simplex method

Minimize Z=25x1 +30x2
Subject to:
20x1+15x2 > 100
2x1+ 3x2 > 15
x1, x2 >0
Solution
Standardize the problem
Minimize Z=25x1 +30x2 +0s1+0s2 +MA1+MA2
Subject to:
20x1+15x2- s1+A1 = 100
2x1+ 3x2 –s2+A2 = 15
x1, x2 , s1, s2 ,A1 ,A2 > 0

Initial simplex tableau

The initial basic feasible solution is obtained by setting x1= x2= s1= s2=0 No
production, x1= x2= s1=0==>20(0) +15(0) - 0+A1 = 100 ==> A1 = 100
x1= x2= s2=0==>0(0)+3(0) - 0+A2 =15==> A2 = 15

Initial (first) Tableau

Cj 25 30 0 0 M M
BV X1 X2 S1 S2 A1 A2 RHS(Q)
A1 M 20 15 -1 0 1 0 100
A2 M 2 3 0 -1 0 1 15
Zj 22M 18M -M -M M M 115M
Cj - Zj 25-22M 30-18M M M 0 0

59

Ashenafi A. (PhD Research Fellow) Page 59

 Second Tableau
Cj 25 30 0 0 M
BV X1 X2 S1 S2 A2 RHS(Q)
X2 25 1 3/4 -1/20 0 0 5
A2 M 0 3/2 1/10 -1 1 5
Zj 25 75/4+3/2 -5/4+1/10M -M M 125+5M
M
Cj - Zj 0 45/4-3/2M 5/4-1/10M M 0

Third Tableau
Cj 25 30 0 0
BV X1 X2 S1 S2 RHS(Q)
X2 25 1 0 -1/10 ½ 5/2
X2 30 0 1 1/15 -2/3 10/3
Zj 25 30 -1/2 - 162.5
15/2
Cj - Zj 0 0 1/2 15/2

Since all Cj - Zj row values are > 0, optimal solution is

reached Minimum Z=$162.5
X1=5/2
X2=10/3

2.5.2.4.SOME SPECIAL ISSUES IN SIMPLEX METHOD

During the course of solving linear programming problems, various conditions can occur that
make the problem unusual and, in some instances, impossible to solve. In this sub section, three
of these special issues are examined; unbounded solutions, degeneracy, and multiple optimal
solutions.

Unbound solutions: A solution is unbound if the objective function can be improved without
limit. The condition is relatively easy to recognize in a Simplex solution. Recall that the way to
determine which variable will leave an interim solution is to compute the ratios of values in the
quantity column and values in the pivot column. The variable with the smallest positive ratio will
leave the solution. The solution is unbound if there are no positive ratios. A negative ratio means
that increasing x1 would increase resources. A zero ratio means that increasing x1 would not use
any resources. This condition generally arises because the problem is incorrectly formulated. If
the constraints do not properly bind the solution, or the objective function is stated as
maximization when it should be a minimization, unbound solutions will occur.

Example: Solve the following linear programming model using the simplex procedure.
Maximize profit: 2x1 + 3x2
Subject to:
-x1 + x2  1
-1/2x1 + x2  3
x1, x2  0

60

Ashenafi A. (PhD Research Fellow) Page 60

The final tableau for the problem is:
Basis Cj 10 6 7 0 Quantity
x1 s1 s1 s2 values
x2 0 0 1 -1 2 5
x1 10 1 0 -2 2 4
Zj 2 3 -7 10 23
Cj-Zj 0 0 7 -10

Even though there is a positive value in the net evaluation row indicating that the solution is not
optimal, we cannot determine the variable to be replaced because there is no positive ratio in the
pivot column. This special condition is referred to as unbounded.

Degeneracy: In the process of developing the next simplex tableau for a tableau that is not
optimal, the leaving variable must be identified. This is normally done by computing the ratios of
values in the quantity column and the corresponding row values in the entering variable column,
and selecting the variable whose row has the smallest non-negative ratio. In some cases, there
will be a tie for the lowest non-negative ratio. Such an occurrence is referred to as degeneracy,
because it is theoretically possible for subsequent solutions to cycle (i.e., to return to previous
solution). There are ways of dealing with ties in a specific fashion; however, it will be usually
enough to simply select one row (variable) arbitrarily and proceed with the computations.
However, there will be no improvement in the solution (level of the objective function); it will
rather be the same as the previous solutions.

Multiple Optimal Solutions: This is the condition when the same maximum value of the
objective function might be possible with a number of different combinations of values of the
decision variables. It occurs when the objective function is parallel to any of the binding
constraint function. Two functions are said to be parallel if the ratio of coefficients of variables is
the same.

When using the simplex approach, we can recognize alternate optimal solution if the Cj-Zj row
contains a zero value for one or more of the non-basic variables (variables not in the solution) in
the final simplex tableau.

Activity 8: Solve the following linear programming model using the simplex procedure.
Maximize profit: 60x1 + 30x2
Subject to:
4x1 + 10x2  100
2x1 + 1x2  22
3x1 + 3x2  39
x1, x2  0
First recognize that the above model is the case of multiple optimal solution because the
objective function 60x1 + 30x2 is parallel to the second constraint 2x1 + 1x2  22. That is, the
ratio of coefficients of variables is the same (i.e. coefficient of x1 to coefficient of x2 both in the
objective function and the second constraint  60 = 2 = 2 :1).
30 1

61

Ashenafi A. (PhD Research Fellow) Page 61

Final tableau for modified microcomputer problem with alternate optimal solution
Basis Cj 60 30 0 0 0 RHS(Quantity)
x1 x2 s1 s2 s3
s1 0 0 8 1 -2 0 56
x1 60 1 1/2 0 1/2 0 11
s3 0 0 3/2 0 -3/2 1 6
Zj 60 30 0 30 0 660
Cj-Zj 0 0 0 -30 0

There is a zero in row Cj-Zj in the x2 column and that indicates the existence of an alternate
optimal solution. The presence of this zero tells us that variable x2 can be brought into solution
without increasing or decreasing the value of the objective function.

2.5.2.5.SIMPLEX:MAXIMIZATION WITH MIXED CONSTRAINTS AND MINIMIZATION

In the previous sub section the simplex approach to the solution of LP problems was introduced.
The technique was demonstrated by using maximization problem in which all of the constraints
are a  variety. In this subsection the question of how to deal with problems that do not have a 
constraint is addressed. Such conditions can occur in the context of both maximization and
minimization problems. LP models with all the =,  , and  signs is called mixed constraints.
Constraints with = and  signs are handled a bit differently. Here, we begin with a discussion of
how to modify= constraints and  constraints so that they are in a form that is compatible with a
simplex solution. Next, a simplex solution of a maximization problem that has a mix of three
constraint types will be illustrated followed by a simplex solution of a minimization problem.

Equality Constraints and Greater or Equal to Constraints

As we have discussed earlier, the simplex technique requires that the LP model should be in its
standard form. Constraints that are  can be put in standard form by adding a slack variable to
the constraint.

Dear learner, how do you convert an = constraint to standard form? In the case of equality
constraints slacks are not acceptable; such constraints require a precise amount. However, as we
saw earlier, the origin serves as a convenient starting point for an initial solution. Thus, there is a
conflict using the origin as starting point, with all decision variables set equal to zero, and
equality constraints, which do not permit any slack. This disparity is resolved in simplex by the
introduction of artificial variables. Artificial variables have no physical interpretations, they
merely serve as a device to enable us use the simplex procedure. During the simplex procedure,
artificial variables are quickly eliminated from the solution. In fact, the optimal solution should
never contain an artificial variable with non-zero value. Artificial variables have the same non-
negativity restriction that decision variables and slack variables have.

Example: Let us assume that we do have the following constraint in an LP model under
consideration: 3x1 + 5x2 = 40,

62

Ashenafi A. (PhD Research Fellow) Page 62

Its standard form will be: 3x1 + 5x2 + A1 = 40. A1 does mean an artificial variable for the first
constraint.

Dear learner, note that an artificial variable is a non-negative variable introduced to facilitate
the computation of an initial basic feasible solution.

Note that if a problem has five constraints, and the fifth constraint listed is an equality
constraint, its artificial variable would be designed as A5.
Now consider 8x1 + 3x2  50

constraint, a  Unlike a  constraint is never allowed to have a

value that is less than the
minimum amount designated by the constraint. However, it can have a value that exceeds the
designated amount (50 in this case). In some situations, a solution will result in values of the
decision variables that exceed the designated minimum level of a constraint. That excess amount
is referred to as surplus. Surplus variables are resources beyond the minimum requirements.
This surplus amount is subtracted from the left side of the constraint function to put it in standard
form.
Example3: 8x1 + 3x2  50 will look like:
8x1 + 3x2 – s1 = 50, where s1 = the surplus amount for the first constraint.
By subtracting the surplus, the constraint becomes equality. As before, to allow for initial
solutions that will be less than that amount, an artificial variable also must be added. Thus, the
standard form of the above constraint is: 8x1 + 3x2 – s1 + A1 = 50

Dear learner, how are we going to represent surplus and artificial variables in the objective
function?

Surplus variables are handled in exactly the same way as slack variables are. Namely, they are
assigned coefficients of zero in the objective function.

Assignment of coefficients for artificial variables depends on whether the problem is a

maximization or minimization problem. In either case, we are using artificial variables to
facilitate solution and we do not want them to appear in the final solution.

In maximization problems, assigning a large negative contribution will ensure that an artificial
variable will not appear in the optimal solution. Conversely, in a minimization problem,
assigning a large positive cost will have a similar effect. Although it would be acceptable to
arbitrarily select a large negative or positive value for the artificial variable coefficients, it is
more common to simply designate the coefficient with large M.

Assigning large negative number to be the coefficient of the artificial variable in the objective
function for a maximization problem will facilitate the exclusion of the artificial variable from
the basis as quickly as possible.

Assigning large positive number to be the coefficient of the artificial variable in the objective
function for a minimization problem will facilitate the exclusion of the artificial variable from
the basis as quickly as possible.

63

Ashenafi A. (PhD Research Fellow) Page 63

Thus, in a maximization problem, the objective function might be:
Maximize profit: 10x1 + 15x2 +0s1 +0s2-MA2

Note that from the objective function, we can tell that there must be two constraints: the first of
which must be a  constraint with s1 as its slack variable and the second of which must be a 
constraint with s2 as the surplus variable and A2 as the artificial variable. If there had not been an
s2 in the objective function, we would have concluded that the second constraint was equality.

Conversely, in a minimization problem, the objective function might be:

Minimize cost: 6x1 + 3x2 +0s1 +0s2+MA2
Again we can conclude that there are two constraints, the first a  and the second a  .

Example 4: Solve the following problem using the simplex approach:

Maximize profit: 6x1 + 8x2
Subject to:
Contraint 1: x2  4
Contraint 2: x1 + x2 = 9
Constraint 3: 6x1 + 2x2  24
x1, x2  0
As usual, the simplex procedure starts solving LP models first by putting them into their standard
form.

The standard form to the above problem is:

Maximize profit: 6x1 + 8x2 + 0s1 + 0s3 – MA2 – MA3
Subject to:
x2 + s1 =4
x1 + x2 + A2 = 9
6x1 + 2x2 – s3 + A3 = 24
x1, x2, s1, s3, A1, A3  0

Then we will develop the initial simplex tableau. In order to develop the initial tableau, follow
these guidelines: List variables across the top of the tableau in order, beginning with the decision
variables, then the slack/surplus variables, followed by any artificial variables. For the initial
basis, use artificial variables for constraints with them. Otherwise, use a constraint‘s slack
variable. Hence, surplus variables will not appear in an initial solution. Using these guidelines,
the tableau, shown below is developed.

The initial simplex tableau

Basis Cj 6 8 0 0 -M -M Quantity
x1 x2 s1 s3 A2 A3
s1 0 0 1 1 0 0 0 4
A2 -M 1 1 0 0 1 0 9
A3 -M 6 2 0 -1 0 1 24
Zj -7M -3M 0 M -M -M -33M
Cj-Zj 6+7M 8+3M 0 -M 0 0

64

Ashenafi A. (PhD Research Fellow) Page 64

The initial solution can be interpreted in this way. The first constraint has a slack of 4 units;
artificial variable A2 has a value of 9 and artificial variable A3 has a value of 24. The ―profit‖ is
- 33M Birr, where M is simply a very large number.

As before, the entering variable will be the one with the largest positive value in row Cj-Zj. The
two positives are 6 + 7M for x1 and 8 + 3M for x2. Because M is so large relative to the 6 and 8,
the latter can simply be ignored. Thus, we can easily see that +7M is larger than +3M; therefore,
we select x1 variable to enter to the solution. Next, dividing the values in the x1 column into their
respective quantity column values, we obtain these ratios:

Row Ratio
s1 4/0=undefined
A2 9/1=9
A3 24/6=4

The leaving variable is determined in the same way as we did in our previous discussions. The
one with the least positive ratio will be the leaving variable. Because A3 yields the smallest
positive ratio, it will be the leaving variable.
Now do not forget to exclude the third artificial variable,A3, from the simplex tableau when you
develop the second tableau. Once an artificial variable leaves the basis, it is no longer needed.
The second tableau will look like the following:
Second Tableau
Basis Cj 6 8 0 0 -M Quantity
x1 x2 s1 s3 A2
s1 0 0 1 1 0 0 4
A2 -M 0 23 0 16 1 5
x1 6 1 13 0 -1 6 0 4
Zj 6 2-23M 0 -1 -1 6M -M 24–5M
Cj-Zj 0 6+23M 0 1+16M 0

Dear learner, what are the next entering and leaving variables?
The largest positive number in the net evaluation row is 6 + 2 3 M. This shows that the next
variable to enter the basis is x2. And the leaving variable is s1. Therefore, we develop the next
tableau by taking the above facts into consideration.

Third Tableau
Basis Cj 6 8 0 0 -M RHS(Quantity)
x1 x2 s1 s3 A2
x2 8 0 1 1 0 0 4
A2 -M 0 0 -23 16 1 73
x1 6 1 0 -1 3 -1 6 0 83
Zj 6 8 6+2 3M -1 – M/6 -M 48 -7 3M
Cj-Zj 0 0 -6- 2 3M 1+M/6 0

65

Ashenafi A. (PhD Research Fellow) Page 65

Still we do have a positive number in the net evaluation row. Therefore, we have to search for a
better solution. We need to determine the next variable to enter the solution mix and the one
which must leave the basis. The next tableau will be developed by excluding A2 from the basis
and bringing s3 to the basis.

Final Tableau
Basis Cj 6 8 0 0 RHS(Quantity)
x1 x2 s1 s3
x2 8 0 1 1 0 4
s3 0 0 0 -4 1 14
x1 6 1 0 -1 0 5
Zj 6 8 2 0 62
Cj-Zj 0 0 -2 0

Look at the above tableau. There is no positive number in the net evaluation row. This means, we
cannot get an improved solution by developing the next tableau. Therefore, the fourth tableau is
the final tableau. The optimal solution is:

x1=5
x2=4
s3=14
Profit=Birr 62

Dear learner, please interpret the optimal solution given above.

b) Solving Maximization Problems with Mixed Constraints

All the logics we have seen so far on simplex procedure holds true again. The only difference is
the way we treat the new variables we introduced to the model (i.e. the artificial variable A and
its coefficient M).

Summary of the Steps to Create Tableau Form

If the original formulation of the LP contains one or more constraints with negative right
hand side values, multiply each of the constraints by -1 and this will change the direction
of the inequalities. This step will provide an equivalent linear program with non-negative
right hand side values.
For a less than or equal to (  ) constraints, add a slack variable to obtain equality
constraint. The coefficient of the slack variable in the objective function is assigned a
value of zero. It provides the tableau form for constraint and the slack variable becomes
one of the basic variables in the initial basic feasible solution.
For a greater than or equal to (  ) constraints, subtract a surplus variable to obtain
equality constraint, and then add an artificial variable to obtain the tableau form. The
coefficient of the surplus variable in the objective function is assigned a value of zero.
The coefficient of the artificial variable in the objective function is assigned a value of –
M, where M is a very large number. The artificial variable becomes one of the basic
variables in the initial basic feasible solution.

66

Ashenafi A. (PhD Research Fellow) Page 66

 For equality (=) constraints, add an artificial variable to obtain the tableau form. The
coefficient of the artificial variable in the objective function is assigned a value of -M.
The artificial variable becomes one of the basic variables in the initial basic feasible
solution.

In nutshell, the above expressions can be summarized in the following table;

Type of To put into standard form Coefficient in Initial Basic
constraint the objective function feasible solutions
≤ add a slack variable 0 slack variables
= add an artificial variable -M artificial variable
becomes one of
them
≥ subtract a surplus variable –Mforartificial artificial variable
and add an artificial variables and 0 becomes one of
variable for surplus variables them

d) Solving minimization problems

For the most part, solution of minimization problems using simplex are handled in the same way
as maximization problems with mixed constraints. The two key exceptions are:
That M coefficients of artificial variables in the objective function are given positive signs
instead of negative signs as noted previously, and the selection of a variable to enter the
solution is based on the largest negative value in row Cj-Zj.
Note also that minimization problems always require at least one artificial variable.

The coefficient of artificial variables, M, when we indicate them in the objective function, has
same meaning as other coefficients of variables in the objective function. It indicates the extent
of the artificial variable‘s per unit contribution to the level of the objective function. The values
in the net evaluation row indicate the extent of possible increase to the level of the objective
function as a result of bringing an additional unit of the variable in that column to the solution
mix.

To sum up, in minimization problems, while the variable with the largest negative value in the
net evaluation row enters the next tableau, the variable with the least positive ratio will leave the
solution mix like the maximization case.

Example 5: Solve the following linear programming model using the simplex procedure.
Minimize cost: 6x1 + 9x2
Subject to:
10x1 + 4x2  400
x2  14
x1, x2  0
We start the procedure by writing the model in its standard form. The standard form for the
above model is:

67

Ashenafi A. (PhD Research Fellow) Page 67

 Minimize cost: 6x1 + 9x2 + 0s1 + 0s2 +MA1 + MA2
Subject to:
10x1 + 4x2 - s1 + A1 = 400
x2 – s2 + A2 = 14
x1, x2 , s1 ,s2 ,A1 ,A2  0

Then we develop the initial simplex tableau. The initial simplex tableau is just representing the
standard form of the model using the simplex tableau. It is developed as follows:

Initial simplex tableau

Basis Cj 6 9 0 0 M M RHS(Quantity)
x1 x2 s1 s2 A1 A2
A1 M 10 4 -1 0 1 0 400
A2 M 0 1 0 -1 0 1 14
Zj 10M 5M -M -M M M 414M
Cj-Zj 6–10M 9–5M M M 0 0

Once we developed a simplex tableau, testing it if it is giving us the optimal solution should be
our usual procedure. This is done by looking at values in the net evaluation row. In our
discussion at the beginning of this section, we have said that we develop the next tableau in a
minimization problem by selecting the variable with the largest negative value to be the next
variable to enter the next tableau. Therefore, checking the tableau for optimality will be done by
checking if there is any negative value in the net evaluation row. There are two negative numbers
in this row. But we have selected 6 –10M to be the largest negative number. This is because M is
just a very large number. So, whenever we compare numbers that involve M, we just disregard
any constant number associated with it, but give attention to the coefficient of M. Then, the one
with the greatest coefficient will be selected to be the largest number. Look at the numbers in the
net evaluation row of the initial simplex tableau. 10 is the greatest coefficient of M. This made 6
– 10M to be selected being the largest negative number and an indicator for the next variable to
enter the basis. Thus x1 is the entering variable. A1 is the leaving variable. Why? Look at the
following tableau:

Second simplex tableau

Basis C 6 9 0 0 M RHS(Quantity)
x1 x2 s1 s2 A2
x1 6 1 2/5 -1/10 0 0 40
A2 M 0 1 0 -1 1 14
Zj 6 12/5+M -3/5 -M M 240+14M
Cj-Zj 0 33/5-M 3/5 M 0

The second tableau is not the optimal tableau because there is a negative number in the net
evaluation row. This indicates that we can get a better solution by bringing the variable having
this negative value (x2 in this case) to the basis. Therefore, the 3rd tableau is developed by taking
this concept into consideration. Look at the following tableau.

68

Ashenafi A. (PhD Research Fellow) Page 68

 Third simplex tableau
Basis Cj 6 9 0 0 RHS(Quantity)
x1 x2 s1 s2
x1 6 1 0 -1/10 2/5 34.4
x2 9 0 1 0 -1 14
Z 6 9 -3/5 -6.6 332.4
Cj-Zj 0 0 3/5 6.6

Since there is no any negative value in the net evaluation row, the 3rd tableau is the optimal
tableau. Therefore, the optimal solution to the above linear programming model is: x1 = 34.4 and
x2 = 14. The level of the objective function is 332.4. There is no surplus variable in the final
tableau. This means, the value of s1 and s2 at optimal solution is zero.

Activity 7: Writing a minimization LP model in standard from

Minimize cost: 5x1 + 10x2
Subject to:
x1 + 3x2  40
x1  10
x1, x2  0
Required: Write the LP model in standard form.

Answer: Deduct surplus variables from the constraint function and the inequality will become an
equation. Moreover, the surplus variables should be represented in the objective function with
zero coefficients as they contribute nothing to the objective of a firm. Surpluses are also assumed
to be non-negative like decision variables.

2.6.THE POST OPTIMALITY ANALYSIS

In the previous section, we have discussed about simplex approach to solving an LP problem.
This section discusses about sensitivity analysis and the concept of a dual linear programming
formulation. Both of these topics are important because they enrich decision making process by
providing decision makers with greater insight in LP analysis. Sensitivity analysis involves an
examination of the potential impact of changes in any of the parameters (i.e., numerical values)
of a problem (e.g., a change in the amount of scarce resources available). The dual involves
setting up and solving an LP problem that is a ―mirror image‖ of an LP problem that has been
formulated.

2.6.1. SENSATIVITY ANALYSIS

Sensitivity Analysis is concerned with the study of ‗Sensitivity ‗of the optimal solution of an
LPM with discretion variables (changes) in parameters. The degree of sensitivity of the solution
due to those variations can range from no change at all to a substantial change in the optimal
solution of the given LPP. Thus, in sensitivity analysis, we determine the range over which the
LP model parameters can change without affecting the current optimal solution. The process of
studying the sensitivity of the optimal solution of an LPP is called post-optimal analysis.

The two sensitivity analysis approaches are:

I. Trial and error approach and

69

Ashenafi A. (PhD Research Fellow) Page 69

 Analytical approach

I. Analytical Approach
Here in this module we prefer to use the analytical approach and it will be discussed in detail
below. Five types of discrete changes in the original LP model may be investigated during the
sensitivity analysis using analytical approach.
Changes of the coefficients of the objective function (cj)
Changes of the RHS Quantity (bj)
Changes of the input-output coefficient
Add/delete constraints
The addition of a new variable to the problem

1.CHANGES OF THE COEFFICIENTS OF THE OBJECTIVE FUNCTION (cj)

The decision variables can be:
i. Basic (in the solution)
ii. Non-basic (out-of the solution)

Note: Instead of resolving the entire problem as a new problem with new parameters, we may
take the original optimal solution table as an initial solution table for the purpose of knowing
ranges both lower and upper within which a parameter may assume value.

a) Range for the coefficients of basic decision variables

The range of optimality is the range over which a basic decision variable coefficient in the
objective function can change without changing the optimal solution mix. However, this change
will change only the optimal value of the objective function.

Example1: Below is a linear program problem and its optimal solution:

Maximize Z=5x1 +4.5x2 +x3
Subject to:
x1+15.8x2 < 150
5x1+6.4x2+15x3 < 77
2.8x2+11.8x3 < 36
x1, x2, x3 ≥0

The optimal tableau for this solution is:

Cj 5 4.5 1 0 0 0
BV X1 X2 X3 S1 S2 S3 RHS
X1 5 1 1.053 0 0.067 0 0 10
X3 1 0 0.67 1 -0.022 0.067 0 1.8
S3 0 0 1.924 0 0.258 -0.773 1 15.12
Zj 5 5.342 1 0.311 0.067 0 51.8
Cj - Zj 0 -0.842 0 -0.311 -0.067 0
Required: Determine the range of optimality for the coefficients of the basic-decision
variables.

70

Ashenafi A. (PhD Research Fellow) Page 70

Solution:
Analysis of basic decision variables
The analysis will be conducted on products on X1 and X3 which are in the basic solution. Divide
each Cj - Zj row entry for variables not in the solution (for instance, by X2, S1 and S2 values) by
the associated variable aij from X1 or X3 row.

Analysis of C1 (for
X1) Steps:
Take the Cj - Zj row of the optimal solution of the non-basic variables
Take the X1 row of the non-basic variables
Find the quotient of ( −)/( 1 )

X2 S1 S2
Cj-Zj row -0.842 -0.311 -0.067
X1 row 1.053 0.067 0
− -0.8 -4.64
1

Upper Limit
The smallest positive number in the ( − )/( 1 )row tells how much the profit of X1 can be increased before the solution is changed.
Upper Limit= C1+the smallest positive value of ( −)/( 1 )
=5+∞=∞
Note: C1=5(See in the original problem given above)
Lower Limit
The largest negative number closest (negative amount closest to 0)
Lower Limit= C1 +the largest negative value of( − )/( 1 )

=5+ (-0.8) = 4.2

Therefore, the range of optimality for the coefficient of X1 is 4.2≤ C1 ≤ ∞.

(The coefficient of X1 in the objective function can change between 4.2 and ∞ without
changing the optimal solution mix X1=10, X3=1.8 and S3=15.12)

II.Analysis for C3 (for X3)

X2 S1 S2
Cj - Zj row -0.842 -0.311 -0.067
X3 row 0.67 -0.022 0.067
− -1.26 14.13 -1
3

Upper Limit= Cj (for X3 )+The smallest positive value of ( −)/( 3 )

=1+ (14.13) =15.13

71

Ashenafi A. (PhD Research Fellow) Page 71

 Note: Cj (for X3) =5(Look in the OF of the LP problem)
Lower Limit= Cj (for X3 )+The largest negative value of ( − )/( 3 )

=1+ (-1) = 0

Therefore, the range of optimality for the coefficient of X3 is 0 ≤ C3 ≤15.13

(The coefficient of X3 in the objective function can change between 0 and 15.13 without
changing the optimal solution mix X1=10, X3=1.8 and S3=15.12).However, this change will
change only the optimal value of the objective function (i.e. Max Z will change).

Activity Take the original LPM and its final/optimal solution tableau.
Maximize Z=50x1 +120x2
Subject to:
x1+4x2 < 80
3x1+x2< 60
x1, x2 > 0

Determine the range of optimality for the coefficient of the basic variables.
Optimal Solution
Cj 50 40 0 0
BV X1 X2 S1 S2 RHS
X1 5 1/2 1 1/4 0 20
S2 0 5/2 0 -1/4 1 40
Zj 60 120 30 0 $2,400
Cj - Zj $-10 $0 $-30 $0

Solution
The range of optimality for C2(X2)’s profit coefficient is: $100 ≤ C2≤ ∞

b) The range for the non-basic variables

If there is a variable Cj, not participating in the optimal basis, then, in order for the variable to be
included in the optimal solution, its coefficient in the objective function will have to change
from the existing Cj to a new level Cj(new).

Cj(new)> Zj

The range of insignificance is the range over which Cj rates for non-basic variables can vary
without causing a change in the optimal solution mix (variable) is called the range of
insignificance.

72

Ashenafi A. (PhD Research Fellow) Page 72

 Example 2: Below is a linear program problem and its optimal solution:
Maximize Z=5x1 +4.5x2 +x3
Subject to:
15 x1+15.8x2 < 150
5x1+6.4x2+15x3 < 77
2.8x2+11.8x3 < 36
x1, x2 , x3 > 0

The optimal tableau for this solution is:

Cj 5 4.5 1 0 0 0
BV X1 X2 X3 S1 S2 S3 RHS
X1 5 1 1.053 0 0.067 0 0 10
X3 1 0 0.67 1 -0.022 0.067 0 1.8
S3 0 0 1.924 0 0.258 -0.773 1 15.12
Zj 5 5.342 1 0.311 0.067 0 51.8
Cj - Z j 0 -0.842 0 -0.311 -0.067 0
Required: Calculate the range of insignificance for the coefficient of non-basic variable.

Solution
X2 (C2) in the above optimal solution is not part of the solution where its value is 0, therefore
is non-basic variable.The analysis for the range of insignificance for X2 (C2) is shown below:

C2=4.5 and Zj( for C2(X2) =5.342

Cj(new for C2(X2)>5.342==> Cj(new for C2 (X2).If the profit contribution of X2 (C2) is greater
than 5.342,then X2 (C2) will be included in the solution.

Thus, ∞≤X2 (C2) < 5.342 is the range of insignificance for X2 (C2).
X2 (C2) can vary with in the given range without causing a change in the optimal solution
mix(X1=10, X1=0, X3=1.8, S1= S2=0 and S3=15.12).

2. CHANGE IN THE RIGHT HAND-SIDE QUANTITY (RHS) OR CHANGE IN THE

AVAILABILITY OF RESOURCE (CAPACITY) (bj)
Shadow prices:
How much should a firm be willing to pay to make additional resources available?

Shadow prices signify the change in the optimal value of the objective function for 1 unit
increases in the value of the RHS of the constraint that represent the availability of scarce
resources. The negative of the number of Cj - Zj row in its slack variable columns provide as
with shadow prices. Or: shadow prices are found in the Zj row of the final simplex tableau in the
slack variable columns.
RHS range is the range over which shadow prices remain valid.

73

Ashenafi A. (PhD Research Fellow) Page 73

 Example 3: Consider the following LPM & its optimal tableau:
Maximize Z=3x1+4x2
Subject to:
x1+5x2 < 15
2x1 + x2 <
8 x2 < 2 x1,
x2 > 0

The optimal tableau is given as:

Cj 3 4 0 0 0
BV X1 X2 S1 S2 S3 RHS
X1 3 1 0 -0.143 0.714 0 3.57
S3 0 0 0 -0.286 0.428 1 1.143
X2 4 0 1 0.286 -0.428 0 0.857
Zj 3 4 0.714 0.428 0
Cj - Z j 0 0 -0.714 -0.428 0

Required:
Determine the shadow price for each constraint
Determine the RHS ranges over which the shadow prices are valid

Solution:
Analysis of the 1st constraint (b1)
Steps:
Take RHS(Quantity) column
Take respective slack or surplus value added to the constraint(Si)
iii. Find the quotient of ⁄1

RHS(Q) S1
1

3.57 -0.143 -24.96

1.143 -0.286 -3.99
0.857 0.286 3.00

Lower Limit=b1-the smallest positive number in the ⁄ 1 column =15-3=12

Upper Limit=b1-the largest negative number in the ⁄ 1 column =15-(-3.99) =18.99

Therefore, 12< b1< 18.99 (The range of resource 1 over which the shadow price $0.714 per unit
is valid).

74

Ashenafi A. (PhD Research Fellow) Page 74

 Analysis of the 2nd constraint (S2)
RHS(Q) S2 RHS
S2

3.57 0.714 5
1.143 0.428 2.67
0.857 -0.428 -2
Lower Limit=b2-the smallest positive number in the RHS ⁄ S2 column =8-(2.67)= 5.33

Upper Limit=b2-the largest negative number in the RHS ⁄ S2 column =8-(-2) =10

Therefore, 5.33< b1< 10 (The range of resource 2 over which the shadow price $0.428 per unit is
valid).
Analysis of the 3rd constraint (S3)
RHS(Q) S3 RHS S3

3.57 0 Undefined
1.143 1 1.143
0.857 0 Undefined
Lower Limit=b3-the smallest positive number in the RHS ⁄ S3 column =2-(1.143)= 0.857
Upper Limit=b3-the largest negative number in the RHS ⁄ S3 column =2-(-∞) =∞

Therefore, 0.857< b3< ∞ (The range of resource 3 over which the shadow price $0 per unit is
valid).

Activity 9: Answer the following questions

Maximize Z=50x1+40x2
Subject to:
3 x1+5x2 < 150 (Assembly time)
x2 < 20 (Portable display)
8x1 + 5x2 < 300 (Warehouse
space) x1, x2 > 0

The final (optimal) simplex tableau is:

Cj 50 40 0 0 0
BV X1 X2 S1 S2 S3 RHS
X2
40 0 1 8/25 0 -3/25 12
S2 0 0 0 -8/25 1 3/25 8
X1 50 1 0 -5/25 0 5/25 30
Zj 50 40 14/5 0 26/5 1980
Cj - Zj 0 0 -14/5 0 -26/5
Required: Determine the shadow prices for the three constraints for the High Tech Co.

75

Ashenafi A. (PhD Research Fellow) Page 75

Answer:
The Zj values for the three slack variables are 14/5, 0 and 26/5, respectively.
Thus, the shadow price for the assembly time constraint is 14/5(i.e.1 additional assembly
time over the existing 150 is $2.8)
The shadow price for the portable display constraint is 0.
The shadow price for the Warehouse space constraint is26.5

Therefore, we see that obtaining more warehouse space will have the biggest positive impact on
High Tech‘s profit.

NB: With a greater than or equal to constraint, the value of the shadow price will be less than or
equal to zero because a 1 unit increase in the value of the RHS cannot be helpful; it makes it
more difficult to satisfy the constraint. As a result, for a maximization problem the optimal value
of the objective function can be expected to decrease when the RHS side of a greater than or
equal to constraint is increased.

Solve the following LPP using simplex method. A firm that manufactures both lawn mowers
and snow blowers:
X1 =the number of lawn mowers
X2 =the number of snow blowers

Maximize Z=$30x1+$80x2
Subject to:
x1+4x2 < 1000 Labor hours available
6x1 + 2x2 < 1,200lb of steel available
x2 < 20 snow blower engine available
x1, x2 > 0

What is the best product mix? What is the optimal profit?

What are the shadow prices? When the optimal solution has been reached, which resource has
the highest marginal value?
Over what range in each of the RHS values are these shadows valid?
What are the ranges over which the objective function coefficients can vary for each of the
two decision variables?

Answer:
x1=100, x2=200 and profit =$19,000

The shadow price for 1 additional labor=$15

The shadow price for 1 additional pound of steel=0
The shadow price for 1 additional snow blowers engine made available =$20
Thus, snow blower engine have the highest marginal value at the optimal solution.

The shadow price for labor hours is valid from 800 hours to 1,066.66 hours
The shadow price for pounds of steel is valid from 1,000pounds up to an infinite number
of pounds

76

Ashenafi A. (PhD Research Fellow) Page 76

 • The shadow price for snow blower engines ranges from 180 engines up to 250 engines
Without changing the current solution mix, the profit coefficient for the mowers can range
from $0 to 40, while the coefficient for the blowers can range from $60 to infinity.

2.6.2. DUALITY
The term ‗dual‘ in a general sense implies two or double. Every linear programming problem
can have two forms. The original formulation of a problem is referred to as its Primal form. The
other form is referred to as its dual LP problem or in short dual form. In general, however, it is
immaterial which of the two problems called primal or dual, since the dual of the dual is primal.
The dual involves setting up and solving an LP problem that is almost a ‗mirror image‘ of an LP
problem that has been formulated. Both in its formulation and solution, the dual is the flip flop
version of the primal.

In the context of LP, duality implies that each LP problem can be analyzed in two different ways,
but having equivalent solution. For example, consider the problem of production planning. By
using the primal LP problem, the production manager attempts to optimize resource allocation by
determining quantities for each product to be produced that will maximize profit. But through
dual LP problem approach, he attempts to achieve production plan that optimizes resource
allocation so that each product is produced at that quantity such that its marginal opportunity cost
equals its marginal return. Thus, the main focus of dual is to find for each resource its best
marginal value or shadow price. This value reflects the scarcity of resources, i.e., the maximum
additional prices to be paid to obtain one additional unit of the resources to maximize profit
under the resource constraints. If resource is not completely used, i.e., there is slack, then its
marginal value is zero.

The shadow price is also defined as the rate of change in the optimal objective function value
with the respect to the unit change in the availability of a resource. Precisely for any constraint,
we have,
Shadow Price = change in the optimal objective function value
unit change in the availability of a resource
Analysis of the dual can also enable a manager to evaluate the potential impact of a new product,
and it can be used to determine the marginal values of resources (i.e. constraints). Relative to a
product, a manager would want to know what impact adding a new product would have on the
solution quantities and the profit; relative to resources, a manager can refer to a dual solution to
determine how much profit one unit of each resource equivalent to. Whereas the primal gives
solution results in terms of the amount of profit gained from producing products, the dual
provides information on the value of the constrained resources in achieving that profit.

2.6.2.1.Formulating the Dual

There are two important forms of primal and dual problems, namely the symmetrical ( or
canonical) and the standard form.

Symmetrical form
Suppose the primal LP problem is given in the form

77

Ashenafi A. (PhD Research Fellow) Page 77

 Maximize Zx = c1x1+ c2x2 + ... +cnxn
Subject to
a11x1+a12x2+... + a1nxn  b1
a21x1+a22x2+... + a2nxn  b2

am1x1+am2x2+... + amnxn  bm
x1,x2... xn  0
and then the corresponding dual LP problem is defined as:

Minimize Zy = b1y1+ b2y2 + ... +bmym

Subject to
a11y1+a21y2+... + am1ym  c1
a12y1+a22y2+... + am2ym  c2

a1ny1+an2y2+... + amnym  cn
y1,y2... ym  0

The above pair of LP problems can be expressed in the general LP model form as

Primal Dual

n m
cjxj biyi
= =
Max Zx = j 1 Min Zy = i 1

Subject to constraints Subject to constraints

n m
aijxj  bi ajiyi  cj
j =1
; i = 1, 2,.., m i=1 ; i= 1,2, ..., n
aij = aji
and xj  0; j = 1,2,…,n and yi  0; j = 1,2,…,m

The following rules which guide the formulation of the dual problem will give you the summary
of the general relationship between primal and dual LP problems.

Primal-Dual Relationship
Primal Dual
Objective is minimization Objective is maximization and vice versa
> type constraints < type constraints
N o of columns N o of rows
N o of rows N o of columns
N o of decision variables N o of constraints
N o of constraints N o of decision variables
Coefficient of Object function RHS value
RHS value Coefficient of Object function

78

Ashenafi A. (PhD Research Fellow) Page 78

Example 1: Consider this Primal problem:
Minimize 40x1 + 44x2 + 48x3
Subject to: 1x1 + 2x2 + 3x3 ≥ 20
4x1 + 4x2 + 4x3 ≥30
x1, x2, x3 ≥ 0
The dual of this problem is:
Maximize Z = 20y1 + 30y2
Subject to:
1y1 + 4y2 ≤40
2y1 + 4y2 ≤ 44
3y1 + 4y2 ≤ 48
y1, y2 ≥ 0

The following table shows how the primal problem is transformed into its dual.

Primal Dual
Objective Minimize 40x1 + 44x2 + Maximize 20y1 + 30y2
function and Subject to 48x3 Subject to
right-hand 1  20 1  40
side values 2  30 2  44
3  48
Primal Dual
Constraint 1 1x1 + 2x2 + 3x3  1 1y1 + 4y2 
coefficients 2 4x1 + 4x2 + 4x3  2 2y1 + 4y2 
3 3y1 + 4y2 

We can see from the table that the original objective was to minimize, whereas the objective of
the dual is to maximize. In addition, the coefficients of the primal‘s objective function become
the right-hand side values for the dual‘s constraints, whereas the primal‘s right-hand side values
become the coefficients of the dual‘s objective function.

Note that the primal has three decision variables and two constraints; whereas the dual has two
decision variables and three constraints. The constraint coefficients of the primal are constraint
coefficients of the dual, except that the coefficients of the first ―row‖ of the primal become the
coefficients of the first ―column‖ of the dual, and the coefficients of the second ―row‖ of the
primal become the coefficients of the second ―column‖ of the dual. When the primal problem is
a maximization problem with all < constraints, the dual is a minimization problem with all >
constraints.

2.6.2.2.Formulating the Dual when the Primal has Mixed Constraints

In order to transform a primal problem into its dual, it is easier if all constraints in a
maximization problem are of the < variety, and in a minimization problem, every constraint is of
the > variety.

79

Ashenafi A. (PhD Research Fellow) Page 79

To change the direction of a constraint, multiply both sides of the constraints by -1. For example,

-1(2x1 + 3x2 ≥ 18) is -2x1-3x2 ≤ -18

If a constraint is equality, it must be replaced with two constraints, one with a ≤ sign and the
other with a ≥ sign. For instance,
4x1 + 5x2 = 20
will be replaced by two equations as follows:
4x1 + 5x2 < 20
4x1 + 5x2 > 20
Then one of these must be multiplied by -1, depending on whether the primal is maximization or
a minimization problem.

EXAMPLE2: Formulate the dual of this LP model.

Maximize Z = 50X1+ 80X2
Subject to:
C1 3X1+ 5X2≤ 45
C2 4X1+ 2X2 ≥16
C3 6X1+6X2= 30
X1,X2≥0
SOLUTION
Since the problem is a max problem, put all the constraints in to the ≤ form. Subsequently, C2
and C3 will be first adjusted in to ≤ constraints.
-C2 will be multiplied by -1:
-1(4X1+ 2X2 ≥16) becomes -4X1- 2X2≤ -16
- C3 is equality, and must be restated as two separate constraints. Thus, it becomes:
6X1+6X2≤ 30 and 6X1+6X2≥ 30.
-C4 Then the second of these must be multiplied by -1.
-1( 6X1+6X2≤ 30) becomes -6X1-6X2≤- 30

After making the above adjustments, rewrite the LP model again.

Maximize Z = 50X1+ 80X2

Subject to:
C1 3X1+ 5X2≤ 45
C2 -4X1- 2X2≤ -16
C3 6X1+6X2≤ 30
C4 -6X1-6X2≤- 30
X1,X2≥0

The dual of the above problem will be:

Minimize 45y1 - 16y2 + 30y3 – 30y4
Subject to
C1 3y1 -4y2 + 6y3 – 6y4 ≥ 50
C2 3y1 - 2y2 + 6y3 – 6y4 ≥ 80
y1, y2, y3, y4 ≥ 0

80

Ashenafi A. (PhD Research Fellow) Page 80

 2.6.2.3.Comparison of the Primal and Dual Simplex Solutions Cross -referencing the
values in the primal and dual final simplex tableaus is shown as follows.
PRIMAL HOW LEBELED/ WHERE CORRESPONDENCE IN
FOUND IN THE PRIMAL THE DUAL

-Decision variable 1 s1,s2,s3,…….

x ,x 2 x 3 ,…..
- slack variable s1,s2,s3,……. y1,y2,y3,……..
- shadow prices Zrow under slack column Quantity column in decision
variable rows.
- solution quantities Quantity column C-Z row under slack and
decision variable columns

Example 3: To show that the flip-flopping of values between the primal and the dual carries
over to their final simplex tableaus, let us look at the following tables. The first table contains the
final tableau for the dual and the second one contains the final tableau for the primal.

Final tableau of Dual solution to the Microcomputer problem

C 100 22 39 0 0 M M Quantity
Basis y1 y2 y3 s1 s2 a1 a2
Primal
y3 39 16/3 0 1 1/3 -2/3 -1/3 2/3 40/3
y2 22 -6 1 0 -1 1 1 -1 10 shadow
Z 76 22 39 -9 -4 9 4 740
C-Z 24 0 0 9 4 9-M 4-M

Primal solution quantities

Final tableau of Primal solution to the Microcomputer problem:

C 60 50 0 0 0 Quantity
Basis x1 x2 s1 s2 s3
s1 0 0 0 1 6 -16/3 24
x1 60 1 0 0 1 -1/3 9
x2 50 0 1 0 -1 2/3 4
Z 60 50 0 10 40/3 740
C-Z 0 0 0 -10 -40/3

The primary concern with a simplex solution is often three fold:

Which variables are in solution?
How much of each variable is in the optimal solution?
What are the shadow prices for the constraints?

Let‘s consider how we can obtain the answers to these questions from the dual solution. Notice
that the solution quantities of the dual are equal to the shadow prices of the primal (i.e. 40/3 and
10). Next, notice that the values of the solution quantities of the primal (i.e., 24, 9, and 4) can be

81

Ashenafi A. (PhD Research Fellow) Page 81

found in the bottom row of the dual. Now, in the primal solution, s1 equals 24. In the dual, the 24
appears in the y1 column. The implication is that a slack variable in the primal solution becomes
a real variable (i.e., a decision variable) in the dual. The reverse is also true: A real variable in
the primal solution becomes a slack variable in the dual.

Therefore, in the primal solution we have x1 = 9 and x2 = 4; in the dual, 9 appears under s1 in
the bottom row, and 4 appears under s2. Thus, we can read the solution to the primal problem
from the bottom row of the dual: In the first three columns of the dual, which equate to slack
variables of the primal, we can see that the fist slack equals 24 and the other two are zero. Under
the dual‘s slack columns, we can read the primal values of the decision variables. In essence,
then, the variables of the primal problem become the constraints of the dual problem, and vice
versa.

2.6.2.4. Managerial Significance of Duality

The importance of the dual LP problem in terms of the information which it provides about
the value of the resources. The economic analysis is concerned with deciding whether or not
to secure more resource and how much to pay for these additional resources. The
significance of the study dual is as follows;
The dual variables provide the decision maker a basis for deciding how much to pay for additional
unit of resources.
The maximum amount that should be paid for one additional unit of a resource is called its shadow
price (also called simplex multiplier).
The total marginal value of the resources equals the optimal objective function value. The dual
variables equal the marginal value of resources (shadow prices).

Note: The value of ith dual variable is the rate at which the primal objective function value will
increase as the ith resource increases, assuming that all other data is unchanged.

2.7. SUMMARY
Linear programming models are used to find optimal solutions to a special class of operations
research problems known as constrained optimization. An LP model consists of a
mathematical statement of the objective function and a set of mathematical statements of the
restrictions (the constraints). The objective function and the constraints consist of symbols
that represent the decision variables (for example, x1, x2, x3, etc) and numerical values called
parameters. In order for linear programming models to be used, the problems must involve a
single objective, a linear objective function and linear constraints, and have known and
constant numerical values. In general, variables are not allowed to have negative values.
These restrictions are referred to as non-negativity constraints.
Linear programming is used in business to maximize the linear functions of a large number
of variables, subject to certain constraints. Among the applications of linear programming
models are problems that involve product mix, portfolio selection, distribution, assignments,
and production and inventory planning.
Solutions to an LP model can be seen being classified into two. The first one is the graphical
approach. This model is applicable for models having only two decision variables. If the
number of variables is greater than two, we can not use the graphical approach. In this case,
another approach – the simplex procedure will be used. Therefore, the simplex method is the

82

Ashenafi A. (PhD Research Fellow) Page 82

 most famous approach to solving linear programming models. The simplex procedure
involves developing a series of tableaus, each of which describes the solution at a corner
point of the feasible solution space, beginning with the origin. Each solution is progressively
better than the previous one. The process is continued until no further improvement can be
realized, at which time the optimal solution can be identified. The simplex procedure ends
with a brief discussion of special issues such as unbounded problems, degeneracy, and
multiple optimal solutions. This unit also described how to convert = and  constraints to
standard form so that problems with those constraints can be solved manually using simplex.
Sensitivity analysis is an extension of linear programming beyond identifying the optimal
solution. It enables managers to answer ―what if…?‖ questions concerning the impact of
potential changes to model; parameters (constants). This unit examined two kinds of
potential changes: a change in the right hand side value of a constraint and a change in an
objective function coefficient. It must be stressed that sensitivity analysis as described here
requires that only one parameter change be involved, and that all other parameters will
remain fixed at their current values. Sensitivity analysis requires the information provided in
the final simplex tableau.

2.8. Self Assessment Questions-2

Tom is a habitual shopper at garage sales. Last Saturday he stopped at one where there were
several types of used building materials for sale. At the low prices being asked, Tom knew
that he could resell the items in another town for a substantial gain. Four things stood in his
way: he could only make one round trip to resell the goods; his pickup truck bed would hold
only 1000 pounds; the pickup truck bed could hold at most 70 cubic feet of merchandise; and
he had only $200 cash with him. He wants to load his truck with the mix of materials that
will yield the greatest profit when he resells them.

Item Cubic feet per unit Price per unit Weight per unit Can resell
for
2 x 4 studs 1 $0.10 5 pounds $0.80
4 x 8 plywood 3 $0.50 20 pounds $3.00
Concrete 0.5 $0.25 10 pounds $0.75
blocks

State the decision variables (give them labels). State the objective function. Formulate the
constraints of this problem. DO NOT SOLVE, but speculate on what might be a good solution
for Tom. You must supply a set of quantities for the decision variables. Provide a sentence or
two of support for your choice.

A financial advisor is about to build an investment portfolio for a client who has $100,000 to
invest. The four investments available are A, B, C, and D. Investment A will earn 4 percent
and has a risk of two "points" per $1,000 invested. B earns 6 percent with 3 risk points; C
earns 9 percent with 7 risk points; and D earns 11 percent with a risk of 8. The client has put
the following conditions on the investments: A is to be no more than one-half of the total
invested. A cannot be less than 20 percent of the total investment. D cannot be less than C.
Total risk points must be at or below 1,000.

83

Ashenafi A. (PhD Research Fellow) Page 83

 Identify the decision variables of this problem. Write out the objective function and
constraints. Do not solve.

A manager must decide on the mix of products to produce for the coming week. Product A
requires three minutes per unit for molding, two minutes per unit for painting, and one
minute for packing. Product B requires two minutes per unit for molding, four minutes for
painting, and three minutes per unit for packing. There will be 600 minutes available for
molding, 600 minutes for painting, and 420 minutes for packing. Both products have
contributions of $1.50 per unit. Answer the following questions; base your work on the
solution panel provided.

What combination of A and B will maximize contribution?

What is the maximum possible contribution?
Are any resources not fully used up? Explain.

John's Locomotive Works manufactures a model locomotive. It comes in two versions--a

standard (X1), and a deluxe (X2). The standard version generates $250 per locomotive for
the standard version, and $350 per locomotive for the deluxe version. One constraint on
John's production is labor hours. He only has 40 hours per week for assembly. The standard
version requires 250 minutes each, while the deluxe requires 350 minutes. John's milling
machine is also a limitation. There are only 20 hours a week available for the milling
machine. The standard unit requires 60 minutes, while the deluxe requires 120.

Required: Formulate as a linear programming problem, and solve using the graphical
solution method.

Phil Bert's Nuthouse is preparing a new product, a blend of mixed nuts. The product must be
at most 50 percent peanuts, must have more almonds than cashews, and must be at least 10
percent pecans. The blend will be sold in one-pound bags. Phil's goal is to mix the nuts in
such a manner that all conditions are satisfied and the cost per bag is minimized. Peanuts cost
$1 per pound. Cashews cost $3 per pound. Almonds cost $5 per pound and pecans cost $6
per pound. Identify the decision variables of this problem. Write out the objective and the set
of constraints for the problem. Do not solve.

Rienzi Farms grows sugar cane and soybeans on its 500 acres of land. An acre of soybeans
brings a $1000 contribution to overhead and profit; an acre of sugar cane has a contribution
of $2000. Because of a government program no more than 200 acres may be planted in
soybeans. During the planting season 1200 hours of planting time will be available. Each

84

Ashenafi A. (PhD Research Fellow) Page 84

 acre of soybeans requires 2 hours, while each acre of sugar cane requires 5 hours. The
company seeks maximum contribution (profit) from its planting decision. Solve graphically,
using the corner point method.

A Juice Company has available two kinds of food Juices: Orange Juice and Grape Juice.
The company produces two types of punches: Punch A and Punch B. One bottle of punch A
requires 20 liters of Orange Juice and 5 liters of Grape Juice.1 Bottle of punch B requires 10
liters of Orange Juice and 15 liters of Grape Juice. From each of bottle of Punch A a profit of
$4 is made and from each bottle of Punch B a profit of $3 is made .Suppose that the company
has 230 liters of Orange Juice and 120 liters of Grape Juice available

Required:
Formulate this problem as a LPP
How many bottles of Punch A and Punch B the company should produce in order to
maximize profit? (Using the simplex method)
What is this maximum profit?

Solve the following LPPs using the simplex method

Maximize Z=3x1 +5x2
Subject to:
x2 < 6
3x1+2x2 < 18
x1, x2 > 0

9. Solve the following LPPs using the simplex method

Maximize Z=20x1 +10x2
Subject to:
5x1+4x2 < 250
2x1+5x2 < 150
x1, x2 > 0

Solve the following LPM using simplex method

Minimize Z=10x1 +5x2
Subject to:
2x1 + 5x2 > 150
3x1+ x2 > 120
x1, x2 >0
Solve the following LPM using simplex
method Minimize Z=4x1 +5x2
Subject to:
x1 + 2x2 > 80

85

Ashenafi A. (PhD Research Fellow) Page 85

 3x1+ x2 > 75
x1, x2 >0
Below is a linear programming problem and its final (optimal)
solution. Maximize Z= 2x1+4x2+x3+x4
Subject to: X1+3x2+x4
4
2x1+x2  3
X2+4x3+x4  3
x1, x2, x3  0

Final (optimal) tableau

C 2 4 1 1 0 0 0
BV X1 X2 X3 X4 S1 S2 S3 RHS
X2 4 0 1 0 2/5 2/5 -1/5 0 1
X1 2 1 0 0 -1/5 -1/5 3/5 0 1
X3 1 0 0 1 3/20 -1/10 1/20 1/4 1/2
Zj 2 4 1 27/20 11/10 9/10 1/4 13/2
Cj-Zj 0 0 0 -7/20 -11/10 -9/10 -1/4

Required: Perform analysis and determine the range:

Basic variable of objective function (range of optimality)
Non-basic variable of objective function (range of insignificance)
RHS value
If the quantity of first constraint increased from 4 to 9, will the final solution
change, if so what are the values of new solution?

86

Ashenafi A. (PhD Research Fellow) Page 86

 CHAPTER THREE
TRANSPORTATION AND ASSIGNMENT PROBLEMS

3.0. INTRODUCTION
This unit describes two special purpose algorithms: the transportation model and the assignment
model. The chapter has two main sections. Section one discusses with transportation models. It
provides general overview of the transportation method, formulation of a transportation model,
finding an initial feasible solution, evaluation of a solution for optimality and developing an
improved solution. Section two is about an assignment model. In each section, there are in-text
questions and learning activities to initiate your study and to check yourself whether you understood
the contents of each section well. Moreover, at the end of the unit, there are checklists, summaries
and self assessment questions that will facilitate your study and ensure your understanding of the
concepts in the unit. Therefore, please work hard to understand this unit very well and to appreciate
the application of linear programming to transportation and assignment problems.

After successful completion of this chapter, learners should be able to:

describe the nature of transportation and assignment problems;
formulate models for transport problems;
find an initial feasible solution to a transportation problem;
develop an improved solution to a transportation problem;
deal with special cases of transportation problems;
explain assignment problems; and
Solve assignment problems manually.

3.1. TRANSPORTATION PROBLEMS

The transportation method is usually applied to distribution type problems, in which, supplies
of goods that are held at various locations are to be distributed to other receiving locations. The
purpose of using an LP model would be to identify a distribution plan that would minimize the
cost of transporting the goods from the warehouse to the retail stores taking into account
warehouse supplies and store demands as well as transportation costs.

Steps in solving transportation Models

The following steps are following in solving transportation problems:
Step1: formulate the transportation model of the problem
Step 2: Identify the initial feasible solution using either of:
2.1. North-west corner method
2.2. Least cost method/Intuitive method
2.3. Vogel’s Approximation method
Step 3: Evaluate/Test for optimality using
3.1. Stepping stone method
3.2. MODI
Step 4: Repeat step 3 until optimal solution is reached

87

Ashenafi A. (PhD Research Fellow) Page 87

 3.1.1. FORMULATING THE TRANSPORTATION MODELS
One important application of linear programming has been in the area of transportation of
resources from place to place to meet a specific set of requirements. A transportation problem
typically involves a set of sending locations, which are referred to origins and a set of receiving
locations, which are referred to as destinations. In order to develop a model of a transportation
problem, it is necessary to have the following information:

Origins and destinations.

Supply quantity (capacity) of each origin.
Demand quantity of each destination.
Unit transportation costs for each route.

The transportation algorithm requires the following assumptions:

All goods are homogeneous so that any origin is capable of supplying any destination.
Transportation costs are a direct linear function of the quantity shipped over any route.

The transportation problem arises frequently in planning for the distribution of goods and
services from several supply locations to several demand locations. Usually the quantity of goods
available at each supply location (origin) is limited and there is a specified amount needed
(demand) at user location (destination). With a variety of shipping routes and differing cost for
the routes, the objective is to know how many units should be shipped from each origin to each
destination so that all destination demands are satisfied and total transportation costs are
minimized.

Example 1: ABC sand and gravel pit has contracted to provide topsoil for three residential
housing developments. Topsoil can be supplied from three different ―farms‖ as follows:

Farm weekly capacity (cubic yard)

A 100
B 200
C 200
Demand for the topsoil generated by the construction project is:
Project weekly demand (cubic yard)
1 50
2 150
3 300
The manager of the sand and gravel pit has estimated the cost per cubic yard to ship over each of
the possible routes:
Cost per cubic yard
From To
Project #1 Project #2 Project #3
A Birr 4 Birr 2 Birr 8
B Birr 5 Birr 1 Birr 9
C Birr 7 Birr 6 Birr 3

Required: Formulate the transportation model of the problem

88

Ashenafi A. (PhD Research Fellow) Page 88

Solution
This constitutes the information needed to solve the problem. The next step is to arrange the
information into a transportation table as follows:

Project Project Project Supply

#1 #2 $3
Farm A 4 2 8 100
Farm B 5 1 9 200
Farm C 7 6 3 200
Demand 50 150 300

3.1.2. FINDING AN INITIAL FEASIBLE SOLUTION

The feasible solution is one in which assignments are made in such a way that all supply and
demand requirements are satisfied. In general, the number of non-zero (occupied) cells should
equal one less than the sum of the number of rows and the number of columns in a transportation
table. In the case of a table with 3 rows and 3 columns, the number of occupied cells should be
3+3-1 = 5 in order to be able to use the transportation algorithm. Sometimes, fewer occupied or
completed cells appear in a solution. When that happens, the solution is referred to as a
degenerate solution; such a solution requires modification in order to be able to determine if it is
optimal.

A solution to a transportation problem consists of quantities that are assigned to the various
routes. These values can range from zero, which implies that no units will be shipped over that
route, to a maximum that equals the smaller of the two quantities: the row (supply) and column
(demand) totals. The logic of the maximum quantity is simple. The quantity shipped cannot
exceed the available supply in a row, and it should not exceed the amount demanded (column
total). Such a solution is feasible solution.

A number of different approaches can be used to find an initial feasible solution. Some of these
are described as follows.

1. The North West Corner Method(NWCM)

The northwest-corner method is a systematic approach for developing an initial feasible solution.
Its main advantage is that it is simple to use and easy to understand. Its main drawback is that it
does not take transportation costs into account. Consequently, such a solution may require
additional effort to obtain the optimal solution.

The northwest-corner method gets its name because the starting point for the allocation process
is the upper left hand (northwest) corner of the transportation table. For the above problem, this
would be the cell that represents the route from farm A to project #1. The following set of
principles guides the allocation.

89

Ashenafi A. (PhD Research Fellow) Page 89

 Begin with the upper left-hand cell, and allocate as many units as possible to that cell. This
will be the smaller of the row supply and the column demand. Adjust the row and column
quantities to reflect the allocation.
Remain in a row or column until its supply or demand is completely exhausted or satisfied,
allocating the maximum number of units to each cell in turn, until all supply has been
allocated (and all demand has been satisfied because we assume total supply and demand
are equal).

Example 2: For ABC problem (the problem above), find an initial feasible solution using
NWCM and compute total transportation cost.

Solution
It can be solved by using the following steps:
Beginning in cell A-1, allocate 50 units, exhausting demand in column 1 and leaving 50 units
of supply in row A.
Staying in row A, move to cell A-2, where supply is now 50 and demand is 150 units. Allocate
50 units to this cell, exhausting the supply of row A and leaving 100 units in column 2.
Staying in column 2, move down to cell B-2, where supply is 200 units. Allocate 100 units to
this cell, exhausting demand in column 2 and leaving 100 units of supply in row B.
Staying in row B, move to cell B-3 and allocate 100 units, exhausting that row‘s supply and
leaving 200 units in column 3.
Staying in column 3, move down to cell C-3 and allocate 200 units, exhausting both the row
and column quantities.

Initial feasible solution using northwest–corner method

Project #1 Project #2 Project #3 Supply
To
From

Farm A 50 4 50 2 8 100
(first) (second)

Farm B 5 1 9 200
100(3rd) 100(4th)

7 6 3 200
Farm C 200(last)

50 150 300 500

Demand
In terms of minimizing total transportation cost, the solution may or may not be optimal. We
shall make that determination very shortly. At this point, let's simply compute the total
transportation cost. The total cost is found by multiplying the quantities in ―completed‖
(nonempty) cells by the cell‘s unit costs and, then summing those amounts. Thus:

90

Ashenafi A. (PhD Research Fellow) Page 90

Total transportation cost = 50(4) =Birr 200
50(2) =Birr 100
100(1) =Birr 100
100(9) =Birr 900
200(3) =Birr 600
Total transportation cost =Birr 1,900

As noted earlier, the main drawback of the northwest corner method is that it does not consider
cell (route) costs in making the allocation. Consequently, if this allocation is optimal, that can be
attributed to chance rather than the method used.

Example 3: Consider the following transportation problem:

To

Fro
Store 1 Store 2 Store 3 Store 4 Supply

Plant 1 19 30 50 10
7
70 30 40 60
Plant 2 9

Plant 3 40 8 70 20 18
Demand 5 8 7 14 34

a. Develop an initial feasible solution using the NWCM

b. Compute the total cost for this solution.

Solution
a. Table: Initial feasible solution
To
Store 1 Store 2 Store 3 Store 4 Supply
From

Plant 1 19 30 50 10
7
5 2
70 30 40 60
Plant 2 6 3 9
40 8 70 20
Plant 3 4 14 18

Demand 5 8 7 14 34

Check that the solution is feasible or not:

==>m + n-1; m=3 and n=4  3+4-1= 6 cells occupied (Feasible solution)

91

Ashenafi A. (PhD Research Fellow) Page 91

 The total transportation cost of the initial feasible solution derived by the NWCM is:
Route Unit Per unit Total
From To Shipped X cost ( $) = Cost ( $)
Plant 1 Store 1 5 19 95
plant 1 Store 2 2 30 60
Plant 2 Store 3 6 30 180
Plant 2 Store 4 3 40 120
Plant 3 Store 4 4 70 280
Plant 3 Store 4 14 20 280
Total Cost= $ 1015

Note: NWCM does not consider the cost factor for allocation.

Activity 1: Determine an initial basic feasible solution to the following transportation problem
using NWCM. Compute the total cost for this solution.
Destination
A B C Supply
S1 2 7 14 5
S 3 3 1 8
2
S3 5 4 7 7
S4 1 6 2 14
Demand 7 9 18

ANSWER: X11=5, X21=2, X22=6, X32=3, X33=4, X43=4, and Total cost =$102

2. The Least Cost Method(LCM)/An intuitive approach

This approach uses lowest cell cost as the basis for selecting routes. The steps are as follows.
Identify the cell that has the lowest unit cost. If there is a tie, select a route that
maximum shipment can be allocated, if there is a tie still, select one arbitrarily. Allocate
a quantity to this cell that is equal to the lower of the available supply for the row and
the demand for the column.
Cross out the cells in the row or column that has been exhausted (or both, if both have
been exhausted) and adjust the remaining row or column total accordingly.
Identify the cell with the lowest cost from the remaining cells. Allocate a quantity to this
cell that is equal to the lower of the available supply of the row and the demand for the
column.
Repeat step 2 and 3 until all supply and demand has been exhausted.

Example 4: Find an initial feasible solution using LCM and calculate the associated total
transportation cost for the following transportation model.

92

Ashenafi A. (PhD Research Fellow) Page 92

 Project Project Project Supply
#1 #2 $3
Farm A 4 2 8 100
Farm B 5 1 9 200
Farm C 7 6 3 200
Demand 50 150 300

Solution:
The cell with the lowest cost is Farm B to Project #2 (B-2), in which the cost is Birr 1. Farm B
has 200 tons of gravel, where as project #2 has a demand for only 150 tons. Consequently we
allocate 150 tons (the lesser of the two quantities) to cell B-2. Because this exhausts the demand
of project #2, we cross out the cells in that column and the demand, and we change the row total
to 50.

Of the cells that remain, C-3 has the lowest (A-2 cannot be used because it has been crossed out).
The supply of farm C is 200 tons, the demand of project #3 is 300 tons. The lesser of these is 200
tons, so that quantity is placed in cell C-3. Since the supply of farm C has been completely used,
the cells in row 3 are crossed out along with the row total. The remaining demand for project #3
is 100 tons, so that column total must be adjusted accordingly. Of the remaining cells, A-1 has
the lowest cell cost. The column total is 50 and the row total is 100; hence, the quantity 50 is
assigned to that cell. Because the demand of project #1 has been satisfied, the cells in the column
must be crossed out, and 50 units must be subtracted from the supply of farm A. at this point,
only two remaining cells have not been crossed out: cells A-3 and B-3. Cell A-3 has the lowest
cost, so it is next in line for allocation. The remaining supply is 50 units, and the remaining
demand is 100 units; consequently, the quantity 50 is placed in A-3. This completes the use of
supply for farm A, and it leaves a demand of 50 units for project #3. The last remaining cell,
then, receives a quantity of 50 units, canceling the remaining supply and demand for both its row
and column.
To Project #1 Project #2 Project #3 Supply

From
Farm A 4 2 50 (4th ) 8 100
50 (3rd )

Farm B 5 150 (1st ) 50 (5th ) 9 200 50

1

Farm C 7 3 200 (2nd ) 3 200

Demand 50 150 300 500

100

93

Ashenafi A. (PhD Research Fellow) Page 93

This solution may or may not be optimal. In the next section, the procedure for testing for
optimality will be described. For now, let us simply compute the total cost of this solution and
compare it to that of the northwest corner solution.

Cell Unit Cost (in Birr) Number of units Transportation cost (in Birr)
A-1 4 50 200
A-3 8 50 400
B-2 1 150 150
B-3 9 50 450
C-3 3 200 600
500 Birr 1,800

Compared to the plan generated using the northwest-corner method, this one has a total cost that
is Birr 100 less. The fact that this plan is less costly than the previous one was expected as the
previous did not involve the use of cost information in allocating units.

Example 5: Find an initial feasible solution using LCM and calculate the associated total
transportation cost for the following transportation model.

Destinations
Factory
W1 W2 W3 W4 Capacity
F1 3 2 7 6 5000
F2 7 5 2 3 6000
F3 2 5 4 5 2500
Demand 6000 4000 2000 1500 13500

Solution:

Factory
W1 W2 W3 W4 Capacity
3 2 7 6
F1 5000
1000 4000
7 5 2 3
F2 6000
2500 2000 1500
2 5 4 5
F3 2500
2500
Demand 6000 4000 2000 1500 13500

94

Ashenafi A. (PhD Research Fellow) Page 94

 Routes Units Unit Cost Total Cost
From To Shipped
F1 W1 1000 3 $ 3000
F1 W2 4000 2 8000
F2 W1 2500 7 17500
F2 W3 2000 2 4000
F2 W4 1500 3 45000
F3 W1 2500 2 5000

Total transportation cost $42,000

m= 3, n=4 ==> 3+4-1 =6 occupied calls (Feasible) Least- Cost method is better
than the NWCM because it considers cost factories.

Example 6: Develop the initial feasible solution for the following Transportation Problem using
the least-cost method (LCM)

D E F G Supply
A 1 5 3 4 100
B 4 2 2 5 60
C 3 1 2 4 120
Demand 70 50 100 60 280

Solution
The 1st allocation be made to the cell with the least-cost. Cells AD & CD both have the lowest
cost f $1. Cell AD is selected 1st because more units can be allocated to it (70) than to cell CE
(50).

Cell CF is filled in 1st since a larger quantity (120-50-70) can be placed there. Then, the
remaining requirement of 30 for column F is allocated to cell BF & source B's supply is reduced
to 30.

The initial solution by the least -cost method

D E F G Supply
A 1 5 3 4 100
70
B 4 2 2 5 60
30 30
C 3 1 2 4 120
50 70
demand 70 50 100 60 280

m=3, n=4 ==> 3+4-1 = 6 occupied cells (feasible)

95

Ashenafi A. (PhD Research Fellow) Page 95

 Routes Units Shipped Unit Cost Total Cost
From To
A D 70 1 $ 70
B F 30 2 60
B G 30 5 150
C E 50 1 50
C F 70 2 140
Total transportation cost $470

Activity 2: Find an initial feasible solution using LCM and calculate the total transportation cost.
R S T Supply
A 1 2 3 100
B 4 1 5 110
Demand 80 120 60

ANSWER: Total transportation cost=80x1+10x2+10x3+110x1+50x0=$240

3.Vogel’s approximation method (VAM)

In this method, each allocation is made on the basis of the opportunity (or penalty) cost that
would have incurred if allocation in certain cells with minimum unit transportation cost were
missed. In this method, allocation is made so that the penalty cost is minimized. The advantage
of this method is that it gives an initial solution which is nearer to an optimal solution or is the
optimal solution itself. The steps in VAM are as follows:
Compute the penalties for each row and column by taking the difference between the
smallest and the next smallest unit transportation cost in the same row and column. This
difference indicates the penalty or extra cost which has to be paid if one fails to allocate
to the cell with minimum unit transportation cost.
Select the row or column with the largest penalty and allocate as much unit as possible
in the cell having the least cost in the selected row or column satisfying the conditions.
If there is a tie in the values of penalties, it can be broken by selecting the cell where
maximum allocation can be made.
Adjust the supply and demand and cross out the satisfied row or column. If a row and a
column are satisfied simultaneously, only one of them is crossed out and the remaining
row (column) is assigned a zero supply (demand). Any row or column with zero supply
or demand should not be used in computing future penalties.
Repeat 1 to 3 until the entire available supply at various sources and demand at various
destinations are satisfied.

Example 7: For the ABC transportation problem above, find an initial feasible solution using
VAM and calculate the total transportation cost.

Solution
The following table shows the computation of penalty for various rows and columns

96

Ashenafi A. (PhD Research Fellow) Page 96

 Project Project
Project Supply Row Penalty
1 2 3
A 50 50 8 100 2 2 4 8 8
4 2 8 50 0

B 5 150 1 50 9 200 4 4 4 9 -
50 0
C 7 6 200 3 200 3 - - - -
0
Demand 50 150 300
0 0 100
50 0
Column 1 1 5
Penalty 1 1 1
1 - 1
- - 1
- - 8

The highest penalty occurs in the third column. The minimum cost in this column is Birr 3.
Hence, 200 units will be shipped from farm C to project 3 and the 3rd row is crossed out.
Proceeding in this way:
150 units will be shipped from farm B to project 2 and the second column is crossed out.
50 units will be shipped from farm B to project 1 and the first column is crossed out
rd
100 units will be shipped from farm A to project 3 and the 3 column is crossed out.

Let us compute the total cost of this solution and compare it to that of the previous solutions.
Cell Unit Cost (in Birr) Number of units Transportation cost (in Birr)
A-1 4 50 200
A-3 8 50 400
B-2 1 150 150
B-3 9 50 450
C-3 3 200 600
500 Birr 1,800

Example 8: Consider the following transportation problem

Destination
Origin 1 2 3 4 Supply
1 20 22 17 4 120
2 24 37 9 7 70
3 32 37 20 15 50
Demand 60 40 30 110 240
Required: find the initial feasible solution using Vogel‘s Approximation Method and compute
the total transportation cost.

Solution:
Now, compute the penalty for various rows and columns which is shown in the following table:

97

Ashenafi A. (PhD Research Fellow) Page 97

 Destination Column
Origin 1 2 3 4 Supply Penalty
Penalty
1 20 22 17 4 120 13
2 24 37 9 7 70 2
3 32 37 20 15 50 5
Demand 60 40 30 110 240
Row Penalty 4 15 8 3

Look for the highest penalty in the row or column, the highest penalty occurs in the second
column and the minimum unit cost i.e. cij in this column is c12=22. Hence assign 40 to this cell
i.e. x12=40 and cross out the second column (since second column was satisfied. This is shown
in the following table:

Destination Row
Origin 1 2 3 4 Supply Penalty

1 20 40 22 17 4 120 13
0
2 24 37 9 7 70 2
3 32 37 20 15 50 5
Demand 60 40 30 110 240
Column 4 15 8 3

The next highest penalty in the uncrossed-out rows and columns is 13 which occur in the first
row and the minimum unit cost in this row is c14=4, hence x14=80 and cross out the first row.
The modified table is as follows:
Destination Row
Origin 1 2 3 4 Supply Penalty

1 20 40 22 17 80 4 120 13
80
2 24 37 9 7 70 2
3 32 37 20 15 50 5
Demand 60 40 30 110 240
Column 4 - 8 3

The next highest penalty in the uncrossed-out rows and columns is 8 which occurs in the third
column and the minimum cost in this column is c23=9, hence x23=30 and cross out the third
column with adjusted capacity, requirement and penalty values. The modified table is as follows:

98

Ashenafi A. (PhD Research Fellow) Page 98

 Destination Row
Origin 1 2 3 4 Supply Penalty

1 20 40 22 17 80 4 120 -
80
2 24 37 30 9 7 70 2
3 32 37 20 15 50 5
Demand 60 40 30 110 240
Column 8 - 11 8

The next highest penalty in the uncrossed-out rows and columns is 17 which occurs in the
second row and the smallest cost in this row is c24=15, hence x24=30 and cross out the fourth
column with the adjusted capacity, requirement and penalty values. The modified table is as
follows:

Destination Row
Origin 1 2 3 4 Supply Penalty

1 20 40 22 17 80 4 120 -
80
2 24 37 30 9 30 7 70 17
40
3 32 37 20 15 50 5
Demand 60 40 30 110 240
Column 8 - - 8

The next highest penalty in the uncrossed-out rows and columns is 17 which occurs in the
second row and the smallest cost in this row is c21=24, hence xi21=10 and cross out the second
row with the adjusted capacity, requirement and penalty values. The modified table is as
follows:

Destination Row
Origin 1 2 3 4 Supply Penalty

1 20 40 22 17 80 4 120 -
80
2 10 24 37 30 9 30 7 70 17
3 32 37 20 15 50 5
Demand 60 40 30 110 240
50
Column 8 - - 8

99

Ashenafi A. (PhD Research Fellow) Page 99

 The next highest penalty in the uncrossed-out rows and columns is 17 which occurs in the third
row and the smallest cost in this row is c31=32, hence xi31=50 and cross out the third row or
first column. The modified table is as follows:

Destination Row
Origin 1 2 3 4 Supply Penalty

1 20 40 22 17 80 4 120 -
80
2 10 24 37 30 9 30 7 70
3 50 32 37 20 15 50 32
Demand 60 40 30 110 240
50
Column 32 - -

Summary of the above cost allocations

Destination Row Penalty
Origin 1 2 3 4 SS

1 20 40 22 17 80 4 120 13 13 - - -
80
2 10 24 37 30 9 30 7 70 2 2 2 17 -
3 50 32 37 20 15 50 5 5 5 5 32
DD 60 40 30 110 240
50
Column 4 15 8 3
4 - 8 3
Penalty 8 - 11 8
8 - - 8
32 - - -

The transportation cost corresponding to this choice of basic variables is:

22*40+4*80+9*30+7*30+24*10+32*50=$3,520

Activity 3: Finding initial feasible solution

A dairy firm has three plants located in different regions. The daily milk production at each plant
is as follows:
Plant 1: 6 million liters.
Plant 2: 1 million liters, &
Plant 3: 10 million liters
Each day the firm must fulfill the needs of its four distribution centers. Minimum
requirement at each center is as follows.

100

Ashenafi A. (PhD Research Fellow) Page 100

 Distribution center 1: 7 million liters
" " 2: 5 " "
" " 3: 3 " "
" " 4: 2 " "

Cost of shipping one million liters form each plant to each distribution center is given in the
following table in hundreds of dollar.

Distribution Center
D1 D2 D3 D4
P1 2 3 11 7
Plant
P2 1 0 6 1
P3 5 8 15 9

Required: Find the initial basic feasible solution by:

North-west corners method
LCM
VAM if the object is to minimize the total transportation cost

Answer:
Total cost = $11, 600
Total cost= $11,200
VAM Solution is shown below

D1 D2 D3 D4 Supply Row Penalty

P1 1 2 5 3 11 7 6 1 1 5 - - -
P2 1 0 6 1 1 1 1 - - - - -
P3 6 5 8 3 15 1 9 10 3 3 4 4 4 5
Demand 7 5 3 2
Column 1 3 5 6
Penalty 3 5 4 2
3 - 4 2
5 - 15 9
5 - - 9
5 - - -
m+n -1 +3+4-1 =6 ==>the solution is non-degenerate.
The total transportation cost=$(1x2+5x3+1x1+6x5+3x15+1x9) x100 = $10,200

3.1.3. EVALUATING/TESTING A SOLUTION FOR OPTIMALITY

The solutions found in the previous three methods are initial solutions and are not final one.
Hence, further improvement should be made to reach an optimal solution as they only indicate
initial feasible solution. Performing evaluation test or optimality test helps the company to
further reduce total transportation cost.

101

Ashenafi A. (PhD Research Fellow) Page 101

The test for optimality for a feasible solution involves a cost evaluation of empty cells (i.e. routes
to which no units have been allocated) to see if an improved solution is possible. We shall
consider two methods for cell evaluation.

The stepping stone method

The modified distribution (MODI) method

The stepping stone method involves a good deal more effort than the MODI method, as you will
note. However, it provides an intuitive understanding of the evaluation process. Moreover, when
a solution is not optimal, the distribution plan must be revised by reallocating units into and out
of various cells and only the stepping stone method can be used for the reallocation. It is,
therefore, necessary to be able to use the stepping stone approach, although the preferred choice
is first to use the MODI method and then, the stepping stone method, if necessary.

1) The Stepping Stone Method

The stepping stone method involves tracing a series of closed paths in the transportation table,
using one such path for each empty cell. The path represents a shift of one unit into an empty
cell, and it enables the manager or analyst to answer a ―what if‖ question: What impact on total
cost would there be if one unit were shifted into a cell. If the shift would result in a cost savings,
the stepping stone path also can be used to determine the maximum number of units that can be
shifted into the empty cell, as well as modifications to other completed cells needed to
compensate for the shift into the previously unused cell.

The name stepping-stone relates to an analogy of crossing a pond or stream by moving from
stone to stone; in the case of a transportation solution the ―stones‖ are the completed (occupied)
cells. Consider the initial feasible solution we found using the northwest corner method.

Steps
Step 1: Determine an initial basic feasible solution using any one of the above three methods.
Step 2: Make sure that the number of occuqwpied cells is exactly equal to m+n-1, where m is the
number of rows and n is the number of columns.
Step 3: Select an unoccupied cell.
Step 4: Beginning at this cell, trace a closed path using the most direct route through at least
three occupied cells used in a solution and then back to the original occupied cell and moving
with only horizontal and vertical moves. The cells at the turning points are called "Stepping
Stones" on the path.
Step 5: Assign plus (+) and minus (-) signs alternatively on each corner cell of the closed path
just traced, starting with the plus sign at unoccupied cell to be evaluated.
Step 6: Compute the net change in the cost along the closed path by adding together the unit cost
figures found in each cell containing a plus sign and then subtracting the unit costs in each
square containing the minus sign.
Step 7: Check the sign of each of the net changes. If all the net changes computed are greater
than or equal to zero, an optimum solution has been reached. If not, it is possible to improve the
current solution and decrease the total transportation cost.
Step 8: Select the unoccupied cell having the most negative net cost change and determine the
maximum number of units that can be assigned to a cell marked with a minus sign on the closed

102

Ashenafi A. (PhD Research Fellow) Page 102

path corresponding to this cell. Add this number to the unoccupied cell and to all other cells on
the path marked with a plus sign. Subtract this number from cells on the closed path marked with
a minus sign.
Step 9: Repeat the procedure until you get an optimum solution.

Example 9:.Below is an initial feasible solution found by the NWCM. Test/evaluate for
optimality using Stepping stone Method and calculate the associated total transportation cost.

Project #1 Project #2 Project #3 Supply

Farm A 50 4 50 8 100
- + 2
Farm B + 5 100 100 9 200
- 1
Farm C 7 6 200 3 200

Demand 50 150 300 500

Solution:
We know it is not optimal because the intuitive method generated an initial solution that has a
lower total cost. However, it will be instructive to analyze the north-west corner solution in order
to see how the stepping stone method works and to gain some insight into the process of
developing an improved solution. Only the unoccupied cells need to be evaluated because the
question at this point is not how many units to allocate a particular route but only if converting a
cell from zero units to non - zero (a positive integer) would decrease or increase total costs. The
unoccupied cells are: A-3, B-1, C-1, and C-2. They must be evaluated one at a time, but in no
particular order.

Let us begin with cell B-1. We start by placing a plus sign (+) in the cell being evaluated, which
stands for the addition of one unit to the cell. In order to maintain the column total of 50, we
must subtract one unit from an occupied cell; cell A-1 is the only option. This is designated by
placing a minus sign (-) in cell A-1. Because we subtracted one unit from row A, we must
compensate for this, which we can do by adding a unit (placing a + sign) in cell A-2. Similarly,
we compensate for the addition of one unit to column 2 by subtracting a unit from cell B-2, and
place a minus sign (–) in that cell to reflect this. Because we initially added one unit to row B in
cell B-1, this last subtraction also compensates for that, and we have traced a completed path,
which we can use to evaluate B-1.

Rules for tracing stepping - stone path

All unoccupied cells must be evaluated; evaluate cells one at a time.
Except for the cell being evaluated, only add or subtract in occupied cells. (It is permissible
to skip over occupied cells to find unoccupied cell from which the path can continue).
A path will consist of only horizontal and vertical moves, starting and ending with the empty
cell that is being evaluated.
Alternate + and - signs, beginning with a + sign in the cell being evaluated.

103

Ashenafi A. (PhD Research Fellow) Page 103

The general implication of the plus and minus signs is that cells with + signs mean one unit
would be added. Cells with a - sign indicate one unit would be subtracted. The net impact of such
a one unit shift can be determined by adding the cell costs with signs attached and noting the
resulting value.

Thus, for cell B-1 we have a net change of +2

-
5 4 7-5=+2
1
-5

This means that for each unit shifted into cell B-1 in this way (which is the only way a shift
could be made), the total cost would increase by Birr 2. Consequently, such a shift would not be
desirable. Turning our attention to cell C-1, we begin its evaluation by placing a + sign in that
cell.
Table 1
Project #1 Project #2 Project #3 Supply
Farm A 50 4 50 2 8 100
- +
Farm B 5 100 1 100 9 200
- +
Farm C 7 6 200 3 200
+ -
Demand 50 150 300 500

We can move horizontally or vertically to an occupied cell. Suppose we move to cell C-3 and
place – sign there. Next, we move vertically to cell B-3, place a + sign in it, then horizontally to
B-2, place a – sign there, move up to cell A-2, place a + sign there. The evaluation of this path is
+10.
+ -
7 3
9 1
18-8=+10
2 4
+18 -8

Hence, using this path for reallocation would increase total cost by Birr 10 per unit. Again this
path would not be desirable. Evaluation paths for empty cells A-3 and C-2 are shown below:

A-3 C-2
- -
2 1
19 93
+911 +15 -4
-2 +11

104

Ashenafi A. (PhD Research Fellow) Page 104

 Summary of cell evaluations
Cell Cell Evaluator(Kij)
A-3 +8-9+1-2= -2
B-1 +5-1+2-4= +2
C-1 +7-3+9-1+2-4= +10
C-2 +6-3+9-1= +11

The negative value of cell A-3 indicates an improved solution is possible. For each unit we can
shift into that cell, the total cost will decrease by Birr 2. How many units can be reallocated into
that cell? The next table shows an improved second table and based on this table, another cell
evaluation will be done to see if an optimal solution is reached or not. As it was indicated above,
if the all cell evaluator values are greater than or equal to zero (kij≥0), optimal solution is
reached.

Table 2
Project #1 Project #2 Project #3 Supply
Farm A 50 4 2 50 8 100

Farm B 5 150 1 50 200

9
Farm C 200 3 200

7 6

Demand 50 150 300 500

Cell Cell Evaluator(Kij)

A-2 +2-8+9-1= +2
B-1 +5-9+8-4= 0
C-1 +7-4+8-3= +8
C-2 +6-1+9-3= +11

Based on the above cell evaluation result all result indicates zero or positive number, hence
optimal solution is reached. The next step is calculating total transportation cost for this
optimal solution.

Total transportation cost

50(4) +150(1) +50(8) +50(9) +200(3) =$1800

NB. Total transportation cost can be reduced by $100 due to optimality test performed and
it can‘t be reduced beyond this value as optimality test /evaluation tells us this is optimal.

2) The Modified Distribution (MODI) method

The MODI (modified distribution) methods of evaluation a transportation solution for optimality
involves the use of index numbers that are established for the rows and columns. These are based
on the unit costs of the occupied cells. The index numbers can be used to obtain the cell
evaluations for empty cells without the use of stepping-stone paths.

105

Ashenafi A. (PhD Research Fellow) Page 105

There is one index number for each column and one for each row. These can be conveniently
displayed along the top and right edges of a matrix. The index numbers are determined in such a
way that for any occupied cell, the sum of the row index and the column index equals to the
cell‘s unit transportation cost.

Row index+ Column index = Cell cost

Ui + Vj = Cij

The index numbers are determined sequentially in a manner dictated by the position of occupied
cells. The process always begins by assigning a value of zero as the index number of row 1.

This method will be illustrated by developing index numbers for the initial feasible solution for
the above problem generated by the north - west corner method. We begin by assigning a value
of zero as the index for row 1. Once row index has been established, it will enable us to compute
column index numbers for all occupied cells in that row. Similarly, once a column index number
has been determined, index numbers for all rows corresponding to occupied cells in that column
can be determined.

Steps
Step1: Determine an initial basic feasible solution using any one of the three methods given
above.
Step2: Determine the values of dual variables, ui and vj, using ui + vj = cij NB. By definition,
U1=0
Step3: Compute the opportunity cost using cij – ( ui + vj ).
Step4: Check the sign of each opportunity cost. If the opportunity costs of all the unoccupied
cells are either positive or zero, the given solution is the optimum solution. On the other hand, if
one or more unoccupied cell has negative opportunity cost, the given solution is not an optimum
solution and further savings in transportation cost are possible.
Step5: Select the unoccupied cell with the smallest negative opportunity cost as the cell to be
included in the next solution.
Step6: Draw a closed path or loop for the unoccupied cell selected in the previous step. Please
note that the right angle turn in this path is permitted only at occupied cells and at the original
unoccupied cell.
Step7: Assign alternate plus and minus signs at the unoccupied cells on the corner points of the
closed path with a plus sign at the cell being evaluated.
Step8: Determine the maximum number of units that should be shipped to this unoccupied cell.
The smallest value with a negative position on the closed path indicates the number of units that
can be shipped to the entering cell. Now, add this quantity to all the cells on the corner points of
the closed path marked with plus signs and subtract it from those cells marked with minus signs.
In this way an unoccupied cell becomes an occupied cell.
Step9: Repeat the whole procedure until an optimum solution is obtained.

Example 10: Below is an initial feasible solution found by the NWCM. Test/evaluate for
optimality using MODI and calculate the associated total transportation

106

Ashenafi A. (PhD Research Fellow) Page 106

 Project #1 Project #2 Project #3 Supply Ui
Farm A 50 4 50 2 -2 8 100 U1=0

Farm B 5 100 1 100 9 200 U2=-1

+2

Farm C 7 6 200 3 200 U3=-7

+10 +11

Demand 50 150 300 500

Vj V1=+4 V2=+2 V3=+10

The index number for column 1 is based on the fact that the sum of its value and the row index
number must equal the cell cost of Birr 4 for cell A-1. Thus, 0+V1=4, so V1=+4. Similarly using
occupied cell A-2, the index number of column 2 is 0 + V2 = 2. V2 = +2.

Knowledge of the index number for column 2 enables us to compute the index number for row B
using the unit cost for occupied cell B-2: U2 +2 = 1, so U2=-1. The value U2 of then enables us
to compute the index number for column 3: V3 -1 = 9 so V3=+10. The remaining index number,
that of row 3, can be determined using the unit cost of occupied cell C-3 and the column 3 index
number, V3:U3+10=3, so U3=-7. We can now readily determine the cell evaluations
(improvement potentials) for each of the unoccupied cells using the relationships:

Cell Evaluation=Cell cost-(Row index column index)

Or Cell Evaluation (Kij)=Cij-(Ui+Vj)

Cell Cell Evaluator

Cell A-3 8-(0+10) = -2
Cell B-1 5-(4-14) = +2
Cell C-1 7-(-7-4) = +8
Cell C-2 6-(-7-2) = +11

These determinations can be made in any order. For example, the cell evolution for A-3 is 8-0-
10= -2. This implies an improvement (decrease in total cost) of Birr 2 per unit for units that can
be shifted into cell A-3. Similarly for empty cell B-1, the improvement potential is 5-4-(-1) =+2
which indicates any units shifted into this cell would increase total cost by Birr 2 each. For
unoccupied cell C-1, the evaluation is 7-(-7)-4 = +10, and for cell C-2, the cell evaluation is 6-(-
7)-2=+11. Note that they agree with the values we computed earlier using the stepping stone
method.

When cell evaluations are positive or zero, an optimal solution has been found. If one or more is
negative, the cell with the largest negative should be brought into solution because that route has
the largest potential for improvement per unit. In this case, we found that cell A-3 had an

107

Ashenafi A. (PhD Research Fellow) Page 107

evaluation of -2, which represented an improvement potential of Birr 2 per unit. Hence, an
improved solution is possible.

Developing an improved solution

Developing an improved solution to a transportation problem requires focusing on the
unoccupied cell that has the largest negative cell evaluation. In the above problem, the only
negative evaluation is for cell A-3. Improving the solution involves reallocating quantities in the
transportation table. More specifically, we want to take the advantage of the improvement
potential of cell A-3 by transferring as many units as possible into that cell. The stepping stone
method for that cell is necessary for determining how many units can be reallocated while
retaining the balance of supply and demand for the table. The stepping stone path also reveals
which cells must have quantity changes and both the magnitude and the direction of the changes.
The stepping stone path for A-3 is reproduced below.
Table 1
To Project #1 Project #2 Project #3 Supply
From

Farm A 50 4 50 - 2 + 8 100

Farm B 5 100 + 1 100 - 9 200

7 6 200 3 200
Farm C
Demand 50 150 300

The + signs in the path indicate units to be added, the - sign indicate units to be subtracted. The
limit on subtraction is the smallest quantity in a negative position along the cell path. There are
two quantities in negative positions, 50 and 100. Because 50 is the smaller quantity, that amount
will be shifted in the following manner. Subtract 50 units from each cell on the path with a -ve
sign and add 50 units to the quantities of each cell with a + sign in it. A quick check reveals that
the sums of quantities in each row and in each column are equal to original row and column
totals. The result is shown as:

108

Ashenafi A. (PhD Research Fellow) Page 108

 Table 2
Project #1 Project #2 Project #3 Supply Ui
Farm A 50 4 2 50 8 100 U1=0

Farm B 5 150 1 50 9 200 U2=+1

Farm C 7 6 200 3 200 U3=-5

Demand 50 150 300

Vj V1=+4 V2=0 V3=+8

With each iteration, it is necessary to evaluate the empty cells to see if further improvement is
possible. This requires use of either the MODI or the stepping stone method. Both will yield the
same value.

We begin by setting the index number for row 1 equal to zero. The column 1 index number is
found using the equality: 0+V1=4. Solving, we find V1=+4. Similarly, for column 3, 0+ V3=8,
so V3=+8. Using cell B-3 as a ―pivot,‖ the index number for row 2 can be found from the
equality U2+ V3=9. Because V3 was found to be +8, this means that U2=+1. This value now
allows us to compute the index number for column 2 because U2+ V2=1; 1+ V2=1, so V2=0.
Lastly, the index number for row 3 can be determined on the basis of the unit cost of cell C-3
because U3+ V3=3, U3+ 8=3. Thus U3 =-5. These values provide the basis for computing
evaluations for the empty cells using the relationship:

Cell Evaluation=Cell cost-(Row index column index)

Or Cell Evaluation (Kij)=Cij-(Ui+Vj)

Cell Cell Evaluator(Kij)

Cell A-2 2-(0+0) = +2
Cell B-1 5-(1+4) = 0
Cell C-1 7-(-5+4) = +8
Cell C-2 6-(-5+0) = +11

Note that the net evaluation index for every unoccupied cell is now greater or equal to zero and
thus we have reached the optimal solution.

Total transportation cost of optimal solution is:

50(4) +150(1) +50(8) +50(9) +200(3) =$1800

109

Ashenafi A. (PhD Research Fellow) Page 109

 Example 11: For example consider the transportation problem given below:
D1 D2 D3 D4 D5 Supply
A 1 9 13 36 51 50
B 24 12 16 20 1 100
C 14 33 1 23 28 150
Demand 100 70 50 40 40

Step 1: First we have to determine the basic feasible solution. The basic feasible solution using
least cost method is:

Initial solution by LCM

D1 D2 D3 D4 D5 Supply

A 50 1 9 13 36 51 50

B 24 60 12 16 20 40 1 100

C 50 14 10 33 50 1 40 23 28 150

Demand 100 70 50 40 40

x11=50, x22=60, x25=40, x31=50, x32=10, x33=50 and x34=40; and total cost
is =$ 2,760

Step 2: The dual variables u1, u2, u3 and v1, v2, v3, v4, v5 can be calculated from the
corresponding cij values,

Step 3: By convention the dual variable U1 is arbitrarily assigned to zero that is U1=0. The
values of the variables calculated are
C11=U1+V1
1=0+V1
V1=1

C32=U3+V2
33=13+V2
V2=20

C33=U3+V3
1=13+V3
V3= -12

C34=U3+V4
23=13+V4
V4=10

110

Ashenafi A. (PhD Research Fellow) Page 110

 C22=U2+V2
12=U2+20
U2= -8

C25=U2+V5
23= -8+V5
V5=15

D1 D2 D3 D4 D5 Supply Ui

A 50 1 9 13 36 51 50 U1=0

B 24 60 12 16 20 40 1 100 U2= -8

C 50 14 10 33 50 1 40 23 28 150 U3=13

Demand 100 70 50 40 40
Vj V1=1 V2=20 V3= -12 V4=10 V5=15

Step 4: Now we calculate cij – (ui + vj) values for all the cells where xij=0 (.i.e. Unoccupied
cells by the basic feasible solution)
That is:
Cell(1,2)= c12-(u1+v2) = 9-(0-20) = -11
Cell(1,3)= c13-(u1+v3) = 13-(0+-12) =1
Cell(1,4)= c14-(u1-v4) = 36-(0+10) = 26
Cell(1,5)= c15-(u1-v5) = 51-(0+15) = 37
Cell(2,1)= c21-(u2-v1) = 24-(-8+1) = 31
Cell(2,3)= c23-(u2-v3) = 16-(-8+-12)=36
Cell(2,4)= c24-(u2-v4) = 20-(-8+10) = 18
Cell(3,5)= c35-(u3-v5) = 28-(13+15) = 0

Note that in the above calculation all the cij – ui – vj ≥ 0 except for cell (1, 2) where C12 – (u1
+v2) = 9-(0-20) = -11.

Thus in the next iteration x12 will be a basic variable changing one of the present basic
variables non-basic. We also observe that for allocating one unit in cell (1, 2) we have to reduce
one unit in cells (3, 2) and (1, 1) and increase one unit in cell (3, 1). The net transportation cost
for each unit of such reallocation is -33 -1 + 9 +14 = -11

The maximum that can be allocated to cell (1, 2) is 10 otherwise the allocation in the cell (3, 2)
will be negative. Thus, the revised basic feasible solution is x11=40, x12=10, x22=60, x25=40,
x31=60, x33=50, x34=40 an corresponding total transportation cost is=$2610,

Therefore, total transportation cost was reduced by $150 (i.e.; 2760-2610)

111

Ashenafi A. (PhD Research Fellow) Page 111

 3.1.4 SPECIAL CASES IN TRANSPORTATION PROBLEM
A number of special issues are discussed in this sub-section in order to round out your
understanding of the transportation model. They are
Determining if there are alternative optimal solutions.
Defining and handling degeneracy (too few occupied cells to permit evaluations of a
solution).
Avoiding unacceptable or prohibited route assignments.
Dealing with problems in which supply and demand are not equal (unbalanced
transportation problem).

Alternate optimal solutions: Sometimes transportation problems have multiple optimal

solutions. In such instances, it can be useful for a manager to be aware of alternative
solutions, because this gives the manager an option of bringing non-quantitative
considerations into the decision. In the case of the transportation problem, the existence of an
alternative solution is signaled by an empty cell's evaluation equal to zero. In fact, you may
have noted that cell B-1 had an evaluation equal to zero in the final solution to the ABC
problem above.

Example 12: Check if the following transportation problem might have more than one
solution. We can find out what that alternate solution is by reallocating the maximum number
of units possible around the stepping stone path for that cell. That path is given below:

Project #1 Project #2 Project #3 Supply

50 - 4 2 50 + 8 100
Farm A
+ 5 150 1 50 9 200
Farm B -

7 6 200 3 200
Farm C
50 150 300

112

Ashenafi A. (PhD Research Fellow) Page 112

 Alternate optimal solution
Project #1 Project #2 Project #3 Supply
Farm A 4 2 100 8 100

Farm B 50 5 150 1 9 200

Farm C 7 6 200 3 200

Demand 50 150 300

As a check, note that the total cost is the same as before, Birr 1800. Please check it up by your
own. Make sure that the cell evaluation index is non-negative for empty cells in the alternate
optimal solution above.

Degeneracy: A transportation problem is a degenerate when there are too few occupied cells
to enable all the empty cells to be evaluated. In the case of the stepping stone method, this
means that there will be at least one empty cell for which an evaluation path cannot be
constructed. For the MODI method, it means that it will be impossible to determine all of the
row and column index numbers. It is relatively simple to determine if a solution is
degenerate. A solution is degenerate if the number of occupied cells is less than the number
of row plus the number of columns minus one.

Test for degeneracy: Number of occupied cells must equal: R + C - 1

Where R = number of rows
C = number of Columns

A quick check of the alternate solution to the ABC problem developed in the preceding section
will reveal that the solution was degenerate. The number of completed cells is 4, while the
required number of completed cell is 3+3-1= 5. This presented no difficulty because we were
not concerned with evaluating the empty cells; we simply were interested in comparing the total
cost of that solution with total cost of the original optimal solution to verify that the two yielded
the same total cost.

However, try to trace a stepping stone path for any of the empty cells and you will understand
the nature of the problem. It should be mentioned that this particular case is somewhat a typical
in that usually some paths can be traced but not all of them. Similarly, if you attempt to
compute index numbers for the rows and columns, you will be unable to compute them for row
B, column 1 or column 2.

Obviously, some modification has to be made in order to determine if a given solution is

optimal. The modification is to treat some of the empty cells as occupied cells. This is
accomplished by placing a delta () in one of the empty cells. The delta represents an extremely

113

Ashenafi A. (PhD Research Fellow) Page 113

 small quantity (e.g., 0.001 units); it is so small that supply and demand for the row and column
involved will be unaffected even without modifying other quantities in the row and column, and
so small that total cost will not change.

The purpose of the delta is to enable evaluation of the remaining empty cells. The choice of
location for the delta can be somewhat tricky: Some empty cells may be unsuitable if they do
not enable evaluations of the remaining empty cells. Moreover, the delta cannot be placed in a
cell which later turns out to be in a negative position of a cell path involved in reallocation,
because delta will be the ―smallest quantity in a negative position‖ and shifting that minute
quantity around the cell path will leave the solution virtually unchanged. Consequently, a
certain amount of trial and error may be necessary before a satisfactory location can be
identified for delta.

The technique can be demonstrated for the degenerate alternate solution of the ABC problem.
Suppose that after some experimentation, cell A-1 has been selected for the location of delta.
Not all choices would be acceptable. For example, try placing the delta in cell C-2 and compute
the improvement potentials for the empty cells. Remember that delta cannot be in a negative
position of a negative cell. The resulting index numbers generated using MODI and the
improvement potential for empty cells based on delta in cell A-1 are shown below and this
confirms that the solution is optimal.

ABC alternate solution modified for degeneracy

V1=+4 V2=0 V3=+8
Project #1 Project #2 Project #3 Supply
Farm A  4 +2 2 100 8 100

Farm B 50 5 150 1 0 9 200

Farm C +8 7 +11 6 200 3 200

Demand 50 150 300

Unacceptable Routes: In some cases, certain origin-destination combinations may be

unacceptable. This may be due to weather factors, equipment breakdowns, labor problems, or
skill requirements that either prohibit, or make undesirable certain combinations. Suppose
that in ABC problem route A-3 was suddenly unavailable because of recent flooding. In
order to prevent that route from appearing in the final solution (as it originally did), the
manager could assign a unit cost to that cell that was large enough to make the route
uneconomical and hence, prohibit its occurrence. One rule of thumb would be to assign a
cost that is 10 times the largest cost in the table. Thus, because the largest cost is Birr 9, a
unit cost of Birr 90 could be assigned instead of the original cost of Birr 8 per unit. Then this
revised problem could be solved using north - west corner, intuitive, or VAM for an initial
solution and either stepping - stone or MODI to evaluate the initial solution and any possible

114

Ashenafi A. (PhD Research Fellow) Page 114

 reallocations. The optimal solution is shown below. The prohibited route may appear in a
non-optimal solution, but it will be eliminated by the time the optimal solution is reached.

Solution to ABC problem with a prohibited route

V1=+4 V2=2 V3=+10
Project #1 Project #2 Project #3 Supply
Farm A 50 4 50 2 +80 90 100

Farm B +2 5 100 1 100 9 200

Farm C +10 7 +11 6 200 3 200

Demand 50 150 300

Dear student, please calculate the total transportation cost from the optimal solution given above.

Unequal supply and demand: Up to this point, the example has involved cases in which
supply and demand were equal (balanced transportation problem). As you might guess,
there are situations in which the two are not equal (unbalanced transportation problem).
When such a situation is encountered, it is necessary to modify the original problem so that
supply and demand are equal. This is accomplished by adding either a dummy column or a
dummy row, a dummy row is added if supply is less than demand and a dummy column is
added if demand is less than supply. The dummy is assigned unit costs of zero for each cell,
and it is given a supply (if a row) or a demand (if a column) equal to the difference between
supply and demand. Quantities in dummy routes in the optimal solution are not shipped.
Rather, they serve to indicate which supplier will hold the excess supply, and how much, or
which destination will not receive its total demand, and how much it will be short.

Let us consider an example. Suppose that farm C in the ABC problem has experienced an
equipment breakdown, and it will be able to supply only 120 cubic yards of topsoil for a period
of time. Therefore, total supply will be 80 units less than total demand. This will require adding
a dummy origin with a supply of 80 units. The modified problem and the final solution are
shown below:

Project #1 Project #2 Project #3 Supply

Farm A 4 2 8 100
Farm B 5 1 9 200
Farm C 7 6 3 120
Dummy 0 0 0 80
Demand 50 150 300 500

115

Ashenafi A. (PhD Research Fellow) Page 115

Solution to ABC problem with unbalanced transportation problem

Project #1 Project #2 Project #3 Supply

Farm A 50 4 2 50 8 100

Farm B 5 150 1 50 9 200

Farm C 7 6 120 3 120

Dummy 0 0 80 0 80
Demand 50 150 300 500

We interpret the solution indicating that project # 3 will be short by 80 units per week until the
equipment is repaired. If the intuitive approach is used to obtain the initial feasible solution when
a dummy is involved, make assignments to the dummy last. Hence, begin by assigning units to
the cell with the lowest non-zero cost, then the next lowest non-zero cost, and so on. For the
ABC problem, this would mean that units would be assigned first to B-2 because its cost of Birr
1 is the lowest non-zero cell cost.

e) Maximization in transportation problem: Although transportation model is used to

minimize transportation cost, it can also be used to get a solution with an objective of
maximizing the total value or returns. Since the criterion of optimality is maximization, the
converse of the rule for minimization will be used. The rule is: A solution is optimal if all –
opportunity costs (cell evaluations) for the unoccupied cell are zero or negative. Or else, such
problems can be solved by adding one additional step at the beginning. Identify the cell with
the largest profit and subtract all the other cell profit from that value. Then replace the cell
profits with the resulting values. These values reflect the opportunity costs that would be
incurred by using routes with unit profit that are less than the largest unit profits. Replace the
original unit profits with these opportunity costs and solve in the usual way for the minimum
opportunity cost solution. This will be identical to maximizing the total profit.

Activity 4: Evaluation/Optimality test

Consider the following transportation problem in our context
To
Store 1 Store 2 Store 3 Supply
From
Warehouse A 12 20 15 50
Warehouse B 9 11 4 15
Warehouse C 20 14 8 55
Demand 25 50 45 120
a) Develop an initial feasible solution using the north-west corner and least cost methods.

116

Ashenafi A. (PhD Research Fellow) Page 116

 Compute the total transportation cost for each initial feasible solution.
Find the optimal solution.
What is the total cost for at optimal solution?
What is its application of this transportation problem in our context?

Comment: Review the previous readings and try yourself.

3.2.ASSIGNMENT PROBLEM
3.2.1. INTRODUCTION
The assignment problem is a special case of transportation problem in which the objective is to
assign a number of origins to the equal number of destinations at the minimum cost (or
maximum profit). Assignment problem is one of the special cases of the transportation problem.
It involves assignment of people to projects, jobs to machines, workers to jobs and teachers to
classes etc., while minimizing the total assignment costs. One of the important characteristics of
assignment problem is that only one job (or worker) is assigned to one machine (or project).
Hence the number of sources are equal the number of destinations and each requirement and
capacity value is exactly one unit.

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

3.2.2.MATHEMATICAL STATEMENT OF PROBLEM

An assignment problem is a special type of linear programming problem where the objective is
to minimize the cost or time of completing a number of jobs by a number of persons.
Furthermore, the structure of an assignment problem is identical to that of a transportation
problem.

3.2.3.APPLICATION AREAS OF ASSIGNMENT PROBLEM

Though assignment problem finds applicability in various diverse business situations, we discuss
some of its main application areas:
In assigning machines to factory orders.
In assigning sales/marketing people to sales territories.
In assigning contracts to bidders by systematic bid-evaluation.
In assigning teachers to classes.
In assigning accountants to accounts of the clients.

In assigning police vehicles to patrolling areas.

117

Ashenafi A. (PhD Research Fellow) Page 117

 Jobs
Persons j1 j2 ------ jn
I1 X11 X12 ----- X1n
I2 X21 X22 ----- X2n
--- ----- ---- ----- -----
In ----- ---- ----- Xnn

Cij is the cost of performing jth job by ith worker.

Xij is the number ith individual assigned to jth job.

Total cost = X11 * C11 + X12 * C12 + ----- + Xnn * Cnn.

Mathematically the assignment problem can be expressed as

The objective function is

Minimize C11X11 + C12X12 + ------- + CnnXnn.

This can also be written as:

n n
= ∑ ∑ Cij Xij

i=1 j=1

subject to the constraints

n
Xij = 1 for all i (resourse availability) j=1

n
Xij = 1 for all i (activity requirement) i=1

and Xij = 0 or 1, for all i to activity j.

3.2.4.SOLUTION METHODS
The assignment problem can be solved by the following four methods :
Enumeration method
Simplex method
Transportation method
Hungarian method

1. Enumeration method
In this method, a list of all possible assignments among the given resources and activities is
prepared. Then an assignment involving the minimum cost, time or distance or maximum profits
is selected. If two or more assignments have the same minimum cost, time or distance, the

118

Ashenafi A. (PhD Research Fellow) Page 118

problem has multiple optimal solutions. This method can be used only if the number of
assignments is less. It becomes unsuitable for manual calculations if number of assignments is
large.

Simplex method As
discussed in chapter two.

Transportation method
As assignment is a special case of transportation problem it can also be solved using
transportation model discussed in previous chapter. But the degeneracy problem of solution
makes the transportation method computationally inefficient for solving the assignment problem.

4. The Hungarian Assignment Method(HAM)

The Hungarian Algorithm developed by Koning has been used with a good deal of success on
these problems and is summarized as follows.

THE HUNGARIAN ASSIGNMENT METHOD (HAM)

Step 1. Determine the cost table from the given problem.
If the no. of sources is equal to no. of destinations, go to step 3.
If the no. of sources is not equal to the no. of destination, go to step2.

Step 2. Add a dummy source or dummy destination, so that the cost table becomes a square
matrix. The cost entries of the dummy source/destinations are always zero.
Step 3: Row Reduction-Select the smallest row values and deduct it from deduct it from all row
values.
Step 4.Column Reduction- In the reduced matrix obtained in the step 3, locate the smallest
element of each column and then subtract the same from each element of that column. Each
column and row now have at least one zero.

Step 5. In the modified matrix obtained in the step 4, search for the optimal assignment as follows:
Examine the rows successively until a row with a single zero is found. Enrectangle this row (□)
and cross off (X) all other zeros in its column. Continue in this manner until all the rows have
been taken care of.
Repeat the procedure for each column of the reduced matrix.
If a row and/or column has two or more zeros and one cannot be chosen by inspection then
assign arbitrary any one of these zeros and cross off all other zeros of that row / column.
Repeat (a) through (c) above successively until the chain of assigning ( ) or cross (X) ends.

Step 6. If the number of assignment ( ) is equal to n (the order of the cost matrix), an optimum
solution is reached.
If the number of assignment is less than n(the order of the matrix), go to the next step.

Step7. Draw the minimum number of horizontal and/or vertical lines to cover all the zeros of
the reduced matrix, using the following sub steps:
Mark all the rows that do not have assignments.

119

Ashenafi A. (PhD Research Fellow) Page 119

 Mark all the columns (not already marked) which have zeros in the marked rows.
iii. Mark a l l t h e r o w s ( not a l r e a d y marked) that h a v e assignments
in marked
iv. columns.
Repeat steps 4 (ii) and (iii) until no more rows or columns can be marked.
Draw straight lines through all unmarked rows and columns.
NB. You can also draw the minimum number of lines by inspection

Step 8. Develop the new revised cost matrix as follows:

Find the smallest element of the reduced matrix not covered by any of the lines.
Subtract this element from all uncovered elements and add the same to all the elements laying at the
intersection of any two lines.
Step 9. Go to step 6 and repeat the procedure until an optimum solution is attained.

See diagrammatic Representation of Hungarian Approach

120

Ashenafi A. (PhD Research Fellow) Page 120

 121

Ashenafi A. (PhD Research Fellow) Page 121

 Example 13: Three computer programmers should be assigned to program three computer
programs. The time taken by all of these three programmers to complete the program is shown
below:
Program 1 Program 2 Program 3
Programmer 1 100 110 120
Programmer 2 90 80 130
Programmer 3 105 80 90

Required: Find the best way of assignment that minimizes the time taken to complete the
computer programs using Hungarian Assignment Method (HAM) and compute the minimum
time.

Solution
Step 1: Row Reduction
Program 1 Program 2 Program 3
Programmer 1 0 10 20
Programmer 2 10 0 50
Programmer 3 25 0 10

Step 2: Column Reduction

Program 1 Program 2 Program 3
Programmer 1 0 10 10
Programmer 2 10 0 40
Programmer 3 25 0 0

Step 3: Make an Assignment

Program 1 Program 2 Program 3
0 10 10

Programmer 2 10 0 40

Programmer 3 25 0 0

This is an optimal assignment, because the total number of rectangle (assignment) (□) is equal to
the total number of rows or columns.

Assignment
Programmer 1 to Program 1
Programmer 2 to Program 2
Programmer 3 to Program 3
Zmin=100+80+90
=270 Minutes

122

Ashenafi A. (PhD Research Fellow) Page 122

 Example 14: A job has four men available for work on four separate jobs. Only one man can
work on any one job. The cost of assigning each man to each job is given in the following
table. The objective is to assign men to jobs such that the total cost of assignment is minimum.

Jobs
Persons 1 2 3 4
A 20 25 22 28
B 15 18 23 17
C 19 17 21 24
D 25 23 24 24

Solution:
Step 1: Row Reduction
Identify the minimum element in each row and subtract it from every element of that row.
Jobs
Persons 1 2 3 4
A 0 5 2 8
B 0 3 8 2
C 2 0 4 7
D 2 0 1 1

Step 2: Column Reduction

Identify the minimum element in each column and subtract it from every element of that
column.
Jobs
Persons 1 2 3 4
A 0 5 1 7
B 0 3 7 1
C 2 0 3 6
D 2 0 0 0

Step 3: Make the assignment for the reduced matrix obtain from steps 1 and 2 in the
following way:
Examine the rows successively until a row with exactly one unmarked zero is found.
Enclose this zero in a box □ as an assignment will be made there and cross (X) all
other zeros appearing in the corresponding column as they will not be

123

Ashenafi A. (PhD Research Fellow) Page 123

 considered for future assignment. Proceed in this way until all the rows have been
examined.
After examining all the rows completely, examine the columns successively until a
column with exactly one unmarked zero is found. Make an assignment to this single
zero by putting square □ around it and cross out (X) all other assignments in that row
proceed in this manner until all columns have been examined.
Repeat the operations (a) and (b) successively until one of the following situations
arises:
All the zeros in rows/columns are either marked□ or crossed (X) and there is exactly
one assignment in each row and in each column. In such a case optimum assignment
policy for the given problem is obtained.
➢ There may be some row (or column) without assignment, i.e., the total
number of marked zeros is less than the order of the matrix. In such a case
proceed to next step 4.

Step 4: Draw the minimum number of vertical and horizontal lines necessary to cover all the
zeros in the reduced matrix obtained from step 3 by adopting the following procedure:

Step 5: Select the smallest element from all the uncovered elements. Subtract this smallest
element from all the uncovered elements and add it to the elements, which lie at the intersection
of two lines. Thus, we obtain another reduced matrix for fresh assignment.

124

Ashenafi A. (PhD Research Fellow) Page 124

 Jobs
Persons 1 2 3 4
A 0 4 0 6
B 0 2 6 0
C 3 0 3 6
D 3 0 0 0

Go to step 3 and repeat the procedure until you arrive at an optimum assignment.

Final Table

Since the number of assignments is equal to the number of rows (& columns), this is the
optimal solution.
The total cost of assignment = A1 + B4 + C2 + D3
Substitute the values from original table:
20+17+24+17=$78.

Example 15: A departmental head has four subordinates, and four tasks to be performed. The
subordinates differ in efficiency, and the tasks differ in their intrinsic difficulty. His estimate,
of the time each man would take to perform each task, is given the matrix below

Men
Persons 1 2 3 4
A 18 26 17 11
B 13 28 14 26
C 38 19 18 15
D 19 26 24 10

Solution:
Step 1: Identify the minimum element in each row and subtract it from every element of that
row, we get the reduced matrix

125

Ashenafi A. (PhD Research Fellow) Page 125

 Men
Persons 1 2 3 4
A 7 15 6 0
B 0 15 1 13
C 23 4 3 0
D 9 16 14 0

Step 2 Identify the minimum element in each column and subtract it from every element of
that column.
Men
Persons 1 2 3 4
A 7 11 5 0
B 0 11 0 13
C 23 0 2 0
D 9 12 13 0

Step 3 Make the assignment for the reduced matrix obtain from steps 1 and 2 in the following
way: Now proceed as in the previous example

Optimal assignment is:

A→G,
B→ E,
C→ F and
D →H
The minimum total time for this assignment scheduled is 17
+13+19+10 or 59 man-hours.

Activity5: Solve this assignment problem, S o as to minimize the time in

hours.

Time-matrix (Time in hrs)

Men
Persons 1 2 3 4
A 6 12 3 7
B 13 10 12 8
C 2 5 15 20
D 2 7 8 13

Answer: Try yourself

3.2.5 SPECIAL CASES IN ASSIGNMENT PROBLEM

1. Multiple Optimum Solutions
This situation of more than one optimal solutions the manager has a elasticity in decision
making. Here the manager can choose any of the solutions by his will and experience.

126

Ashenafi A. (PhD Research Fellow) Page 126

 2. Maximization case in Assignment Problem
Some assignment problems entail maximizing the profit, effectiveness, or layoff of an
assignment of persons to tasks or of jobs to machines. The Hungarian Method can also solve
such problems, as it is easy to obtain an equivalent minimization problem by converting every
number in the matrix to an opportunity loss. The conversion is accomplished by subtracting all
the elements of the given effectiveness matrix from the highest element. It turns out that
minimizing opportunity loss produces the same assignment solution as the original
maximization problem.

Example 16: Five different machines can do any of the five required jobs, with different profits
resulting from each assignment as given below:

Machines
Jobs A B C D E
1 30 37 40 28 40
2 40 24 27 21 36
3 40 32 33 30 35
4 25 38 40 36 36
5 29 62 41 34 39

Find out the maximum profit possible through optimum assignment.

Solution:
Here, the highest element is 62. So we subtract each value from 62.
Machines
Jobs A B C D E
1 32 25 22 34 22
2 22 38 35 41 26
3 22 30 29 32 27
4 37 24 22 26 26
5 33 0 21 28 23
Now use the Hungarian Method to solve the above problem. And the maximum profit through
this assignment is 214.

Example 17. XYZ Ltd. employs 100 workers of which 5 are highly skilled workers that can
be assigned to 5 technologically advanced machines. The profit generated by these highly
skilled workers while working on different machines is as follows:
Machines
Workers I II III IV VI
A 40 40 35 25 50
B 42 30 16 25 27
C 50 48 40 60 50
D 20 19 20 18 25
E 58 60 59 55 53

127

Ashenafi A. (PhD Research Fellow) Page 127

Solve the above assignment problem so as to maximize the profits of the company.

3. Unbalanced Assignment Problem

It is an assignment problem where the number of persons is not equal to the number of jobs.If
the number of persons is less than the number of jobs then we introduce one or more dummy
persons (rows) with zero values to make the assignment problem balanced. Likewise, if the
number of jobs is less than the number of persons then we introduce one or more dummy jobs
(columns) with zero values to make the assignment problem balanced

Example 18: Solve the following problem using HAM

Jobs
Persons 1 2 3 4
A 20 25 22 28
B 15 18 23 17
C 19 17 21 24

Solution:
Since the number of persons is less than the number of jobs, we introduce a dummy person
(D) with zero values. The revised assignment problem is given below:

Jobs
Persons 1 2 3 4
A 20 25 22 28
B 15 18 23 17
C 19 17 21 24
D 0 0 0 0
(dummy)
Now use the Hungarian Method to solve the above problem.

Example 19. In a typical assignment problem, four different machines are to be assigned to
three different jobs with the restriction that exactly one machine is allowed for each job. The
associated costs (in rupees ‗‘000) are as follows:

Jobs
Persons 1 2 3
A 60 80 50
B 50 30 60
C 70 90 40
D 80 50 70

128

Ashenafi A. (PhD Research Fellow) Page 128

 4. Prohibited Assignment
Sometimes it may happen that a particular resource (say a man or machine) cannot be assigned
to perform a particular activity. In such cases, the cost of performing that particular activity by a
particular resource is considered to be very high (written as M or ∞) so as to prohibit the entry
of this pair of resource-activity into the final solution.

Activity 6: Solving assignment problem

Scott and associates, Inc., is an accounting firm that has three new clients. Project leaders will be
assigned to the three clients. Based on the different backgrounds and experiences of the leaders,
the various leader-client assignments differ in terms of project completion times. The possible
assignments and estimated completion times in days are

Client
Project leader
1 2 3
Mr. X 10 16 32
Mr. Y 14 22 40
Mr. Z 22 24 34
What is the optimal assignment of project leaders? What is the total time required?

Comments: The optimal assignment is an assignment with the minimum possible time to
complete the project. Apply the steps mentioned above and find the optimal assignment of the
three project leaders to the three clients.

3.3. SUMMARY
This Chapter describes two types of problems that lend themselves to solution using linear
programming techniques: the transportation problem and the assignment problem. Transportation
type problems often involve the distribution of goods, whereas assignment type problems
involve the matching or pairing of two sets of items. Usually, such problems have different costs
for different distribution alternatives or different pairings, and the objective is to identify the plan
which minimizes total cost. For each type of problem, a procedure for formulating the problem,
in a way that lends itself to a solution using a simple algorithm is described.

To distribute the goods with the least cost, initial tableau representing the initial distribution can
be made by north west corner method, VAM or intuitive (least cost) methods. This initial tableau
might be at its optimal state or not. To check for optimality we can use stepping stone method or
modified distribution method. The stepping stone and MODI are important tools used to evaluate
whether the shift of goods from occupied cells to unoccupied cells can result in a reduction of
total transportation cost or not.

The basic assumption within which we can solve assignment problems using the linear
programming is ―assigning one job, project, etc… to one worker, machine, etc…‖Because of
the feature of assignment problems, special-purpose solution procedures have been specifically
designed to solve the assignment problem; one such procedure is called the Hungarian method
which involves what is called matrix reduction.

129

Ashenafi A. (PhD Research Fellow) Page 129

 3.4.Self Assessment Questions-3
For the data below, construct an initial feasible solution using the northwest corner rule.

COSTS Dest. 1 Dest. 2 Dest. 3 Dest. 4 Supply

Source 1 12 18 9 11 105

Source 2 19 7 30 15 145

Source 3 8 10 14 16 50

Demand 80 60 70 90 300 \ 300

For the transportation problem below, construct an initial feasible solution using the intuitive/LCM
method.
COSTS Dest. 1 Dest. 2 Dest. 3 Dest. 4 Supply
Source 1 12 18 9 11 45

Source 2 19 7 30 15 145

Source 3 8 10 14 16 50

Demand 80 30 70 60 240 \240

Consider the transportation problem in the data set and optimal solution below. Verify by
hand or by calculator (show your work) the value of the objective function.

COSTS Dest. 1 Dest. 2 Dest. 3 Dest. 4 Supply

Source 1 12 18 9 11 100

Source 2 21 7 30 15 150

Source 3 8 10 14 16 50

Demand 80 60 70 90 300 \ 300

Consider the transportation problem in the data set and optimal solution below. Calculate
improvement indexes on each empty cell. Is this solution optimal?

130

Ashenafi A. (PhD Research Fellow) Page 130

 COSTS Dest. 1 Dest. 2 Dest. 3 Dest. 4 Supply
Source 1 12 18 9 11 100
Source 2 21 7 30 15 150
Source 3 8 10 14 16 50
Demand 80 60 70 90 300 \ 300

Find the minimum cost shipping solution for the transportation problem data set in the table
below. Provide a table of shipping quantities and the minimum value for the objective
function.
COSTS Dest. 1 Dest. 2 Supply
Source 1 6 10 20
Source 2 8 3 30
Source 3 4 6 15
Source 4 12 11 20
Source 5 7 9 25
Demand 70 40 110 \ 110

6. Consider the transportation data set for a minimization problem below.

COSTS Dest. 1 Dest. 2 Dest. 3 Supply

Origin 1 5 4 3 30
Origin 2 4 2 6 40
Origin 3 3 7 9 20
Demand 15 50 25 90\90
Calculate the initial solution using the northwest corner rule.
Calculate improvement indexes, iterate, and solve for the optimal shipping pattern
using stepping stone method.

Find the minimum cost solution for the transportation problem detailed in the table below.
Explain carefully the meaning of any quantity in a ―dummy‖ row or column.
COSTS Dest 1 Dest 2 Dest 3 Supply
Source 1 30 12 5 20
Source 2 10 14 12 30
Source 3 20 11 25 75
Dummy 0 0 0 30
Demand 40 60 55 155 \ 155

The Shamrock Transportation Company has four terminals: A, B, C, and D. At the start of a
particular day, there are 8, 8, 6, and 3 tractors available at those terminals, respectively.
During the previous night, trailers were loaded at plants R, S, T, and U. The number of

131

Ashenafi A. (PhD Research Fellow) Page 131

 trailers at each plant is 2, 12, 5, and 6, respectively. The company dispatcher has
determined the distances between each terminal and each plant, as follows. How many
tractors should be dispatched from each terminal to each plant in order to minimize the total
number of miles traveled?

COSTS Dest. 1 Dest. 2 Dest. 3 Dest. 4 Supply

Origin 1 21 44 25 40 8
Origin 2 42 24 45 22 8
Origin 3 78 32 42 55 6
Origin 4 44 42 32 30 3
Demand 2 12 5 6 25\25

A) Find the minimum cost solution for the transportation problem detailed in the table below.

COSTS Dest 1 Dest 2 Dest 3 Supply

Source 1 18 12 5 20
Source 2 10 14 12 30
Source 3 9 11 15 75
Demand 40 30 55 125 \ 125

Before your solution can be implemented, you discover that the combination Source 3 –
Destination 1 is unavailable, due to political turmoil in the country where Source 3 is
located. Solve the revised problem. How much is cost increased by this complication?

A firm has established a distribution network for the supply of a raw material critical to
its manufacturing. Currently there are two origins for this raw material, which must be
shipped to three manufacturing plants. The current network has the following
characteristics:

COSTS Plant 1 Plant 2 Plant 3 Supply

Raw material source 1 $6 $8 $9 400
Raw material source 2 $4 $7 $3 600

Demand 500 500 500 1500 \ 1000

The firm has identified two potential sites for a third raw material source; these are
identified as Candidate A and Candidate B. From A, the costs to ship would be $ 9 to Plant
1, $10 to Plant 2, and $12 to Plant 3. From B, these costs would be $11, $14, and $8. The
new source, wherever it is located, will have a capacity of 500 units. Set up—but DO NOT
SOLVE— this problem as though you were going to solve it with transportation problem
software.

132

Ashenafi A. (PhD Research Fellow) Page 132

 A firm has established a distribution network for the supply of a raw material critical to
its manufacturing. Currently there are two origins for this raw material, which must be
shipped to three manufacturing plants. The current network has the following
characteristics:

COSTS Plant 1 Plant 2 Plant 3 Supply

Raw material source 1 $6 $8 $9 400
Raw material source 2 $4 $7 $3 600

Demand 500 500 500 1500 \ 1000

The firm has identified two potential sites for a third raw material source; these are
identified as Candidate A and Candidate B. From A, the costs to ship would be $ 9 to Plant
1, $10 to Plant 2, and $12 to Plant 3. From B, these costs would be $11, $14, and $8. The
new source, wherever it is located, will have a capacity of 500 units. Solve with the
transportation method. Which site should be selected?

A manufacturer of semiconductor "wafers" has been attempting to convert its operations to

practices more in keeping with JIT principles. The firm is now paying much more attention
to the transit time between one processing stage and the next. The plant has a somewhat
haphazard pattern of machine locations, partly because the machines were purchased and
installed at different times, partly from a shortage of floor space, and partly from previous
experiments with work cells. The bottom line is this: there are four machines that perform a
certain processing phase, and three machines that perform the next phase. All units of a
large class of wafers go through these two phases. The table below displays the transit time,
in minutes, from each machine of the first phase to each machine of the second. Machine 3
is not really 100 minutes away from machine B; the company has prohibited that
combination because of quality problems associated with that specific pairing. Supply and
demand quantities are in wafers processed per week. Develop a transit time minimizing
solution for this firm. What is the total transit time of this solution? Which machines are
fully utilized? Which machines have some capacity unused or requirements unfilled? Was
the prohibition on the 3-B combination honored?

COSTS Phase 2, Phase 2, Phase 2, Supply

machine A machine B machine C
Phase 1, machine 1 7.5 2.5 4.0 700
Phase 1, machine 2 3.0 6.0 6.5 600

Phase 1, machine 3 2.5 100.0 8.0 1000

Phase 1, machine 4 5.0 8.5 7.0 700

Demand 1200 800 1500 3500 \ 3000

133

Ashenafi A. (PhD Research Fellow) Page 133

 Four companies viz. W, X, Y and Z supply the requirements of three warehouses viz. A, B
and C respectively. The companies‘ availability, warehouses requirements and the unit cost of
transportation are given in the following table. Find an initial basic feasible solution using
North West Corner Method
Least Cost Method
Vogel Approximation Method (VAM)

Warehouses
Company A B C Supply
W 10 8 9 15
X 5 2 3 20
Y 6 7 4 30
Z 7 6 9 35
Demand 25 26 49

14. Savanna Store is interested to purchase the following type and quantities of dresses

Dress Type V W X Y Z

Quantity 150 100 75 250 200

Four different dress makers are submitted the tenders, who undertake to supply not more than
the quantities indicated below:

Dress A B C D
Maker
Dress 300 250 150 200
Quantity

Savanna Store estimates that its profit per dress will vary according to the dress maker as
indicates in the following table:

V W X Y Z
A
2.75 3.5 4.25 2.25 1.5

B 3 3.25 4.5 1.75 1

C 2.5 3.5 4.75 2 1.25

D 3.25 2.75 4 2.5 1.75

Determine how should the orders to be places for the dresses so as to maximize the profit.

134

Ashenafi A. (PhD Research Fellow) Page 134

 A shop has four machinists to be assigned to four machines. The hourly cost of having
each machine operated by each machinist is as follows:
Machine (cost/hr.)
Machinist A B C D
1 $12 $11 $8 $14
2 10 9 10 8
3 14 8 7 11
4 6 8 10 9
However, because he does not have enough experience, machinist 3 cannot operate machine B.
Determine the optimal assignment and compute total minimum cost.

The Bunker Manufacturing firm has five employees and six machines and wants to assign the
employees to the machines to minimize cost. A cost table showing the cost incurred by each
employee on each machine follows:
Machine
Employee A B C D E F
1 $12 $7 $20 $14 $8 $10
2 10 14 13 20 9 11
3 5 3 6 9 7 10
4 9 11 7 16 9 10
5 10 6 14 8 10 12
Because of union rules regarding departmental transfers, employee 3 cannot be assigned to
machine E, and employee 4 cannot be assigned to machine B. Solve this problem, indicate the
optimal assignment, and compute total minimum cost.

An electronics firm produces electronic components, which it supplies to various electrical

manufacturers. Quality control records indicate that different employees produce different
numbers of defective items. The average number of defects produced by each employee for
each of six components is given in the following table:
Component
Employee A B C D E F
1 30 24 16 26 30 22
2 22 28 14 30 20 13
3 18 16 25 14 12 22
4 14 22 18 23 21 30
5 25 18 14 16 16 28
6 32 14 10 14 18 20

Determine the optimal assignment that will minimize the total average number of defects
produced by the firm per month.

The football coaching staff at Tech focuses its recruiting on several key states, including
Georgia, Florida, Virginia, Pennsylvania, New York, and New Jersey. The staff includes seven
assistant coaches, two of whom are responsible for Florida, a high school talent-rich state,
whereas one coach is assigned to each of the other five states. The staff has been together for a

135

Ashenafi A. (PhD Research Fellow) Page 135

long time and at one time or another, all the coaches have recruited in all the states. The head
coach has accumulated some data on the past success rate (i.e., percentage of targeted
recruits signed) for each coach in each state, as shown in the following table:
State
Coach GA FL VA PA NY NJ
Allen 62 56 65 71 55 63
Bush 65 70 63 81 75 72
Crumb 46 53 62 55 64 50
Doyle 58 66 70 67 71 49
Evans 77 73 69 80 80 74
Fouch 68 73 72 80 78 57
Goins 72 60 74 72 62 61

Determine the optimal assignment of coaches to recruiting regions that will maximize the overall
success rate and indicate the average percentage success rate for the staff with this assignment.

136

Ashenafi A. (PhD Research Fellow) Page 136

 CHAPTER FOUR
DECISION THEORY/ANALYSIS

4.0. INTRODUCTION
In this chapter, you will study decision theory. Decision theory represents a generalized
approach to decision making, which often serves as the basis for a wide range of managerial
decision making. The decision model includes a list of courses of action that are available and
the possible consequences of each course action. An important factor in making a decision is the
degree of certainty associated with the consequences. This can range anywhere from complete
certainty to complete uncertainty and it generally affects the way a decision is reached. This unit
presents two commonly used decision theory approaches: a payoff table and a decision tree.
They provide structure for organizing the relevant information in a format conducive to making
rational decisions.

Dear student! In the previous chapter dealing with LP, models were formulated and solved in
order to aid the manager in making decision. The solutions to the models were represented by
values for the decision variables. However, these LP models are formulated under the
assumption that certainty existed. In actual practice, however, many decision making situations
occur under conditions of uncertainty. For example, the demand for a product may be not 100
units next week, but 50 or 200 units, depending on the market (which is uncertain).

Making decisions in an atmosphere of increasing time pressure, uncertainty, and conflicting

expert opinions creates challenges for any manager. Making such leadership decisions in crisis
situations is even more demanding. To a great extent, the success or failure that you (or any
individual) experience even in day-to-day life depends on the decisions made by you. Decision
making in today‘s environment and especially in business is far more complex than earlier and
the cost of making errors is too high. Thus to make a right decision a systematic approach is
necessary. Decision analysis provides an analytic and systematic approach to decision making.

After studying this unit, you should be able to:

define decision making;
appreciate the processes of decision making;
differentiate the types of decision making models and environments;
identify the various criteria for decision making;
differentiate the tools used in decision making analysis;
construct a payoff table and decision tree;
learn how to use decision trees, to find out limitations & advantages of decision tree analysis

137

Ashenafi A. (PhD Research Fellow) Page 137

4.1. CHARACTERISTICS OF DECISION THEORY
Decision theory problems are characterized by the following:
List of alternatives: A finite number of decision alternatives are available to the decision
maker at the time of making a decision. The number and type of such alternatives may
depend on th e previous decisions made and the consequences of those decisions. These
alternatives are also called courses of actions, acts or strategies and are under control and
known to you that is you will determine what courses of action are possible.

Decision alternatives must be mutually exclusive (clearly distinct among themselves) and,
ideally, collectively exhaustive (cover all reasonable options open to management).
Determining a realistic set of action alternatives demands creativity and experience on the
nature of the problem under consideration. Managerial intuition is extremely valuable at this
stage of the analysis.

States of nature: - the set of possible future conditions, or events, beyond the control of the
decision maker, that will be the primary determinants of the eventual consequence of the
decision. The states of nature, like the list of alternatives, must be mutually exclusive and
collectively exhaustive.
Decision theory will require the decision maker to develop a mutually exclusive and
collectively exhaustive list of possible future events. These future events are referred to as
states of nature and they depend upon certain factors which are beyond the control. There
may be a great deal of uncertainty with respect to the occurrence of the states of nature.

Payoffs: A payoff is a quantitative measure of the result of taking a particular course of

action combined with the occurrence of a particular state of nature. It is the net gain or loss
obtained as the outcome of the decision that accrues from a given combination of decision
alternatives and events. They are also known as conditional profit values. In business,
payoffs are usually expressed using monetary values. In decision analysis, however, one can
make use of either monetary values or abstract utilities.

The payoffs might be profits, revenues, costs, or other measures of value. Usually the
measures are financial. Usually payoffs are estimated values. The more accurate these
estimates, the more useful they will be for decision making purposes and the more likely, it is
that the decision maker will choose an appropriate alternative. The number of payoffs
depends on the number of alternative/state of nature combination.

Degree of certainty: - the approach often used by a decision maker depends on the degree of
certainty that exists. There can be different degrees of certainty. One extreme is complete
certainty and the other is complete uncertainty. The later exists when the likelihood of the
various states of nature are unknown. Between these two extremes is risk (probabilities are
unknown for the states of nature). Knowledge of the likelihood of each of the states of nature
can play an important role in selecting a course of active.

Decision criteria: - the decision maker‘s attitudes toward the decision as well as the degree
of certainty that surrounds a decision. Example; maximize the expected payoffs.

138

Ashenafi A. (PhD Research Fellow) Page 138

 4.2. THE PAYOFF TABLE
A payoff table is a device a decision maker can use to summarize and organize information
relevant to a particular decision. It includes a list of alternatives, the possible future states of
nature, and the payoffs associated with each of the alternative/state of nature combinations. If
probabilities for the states of nature are available, these can also be listed. The general format of
the table is illustrated below:

States of nature
S1 S2 S3
A1 V11 V12 V13
AlternativesA2 V21 V22 V23
A3 V31 V32 V33
where:
Ai = the ith alternative
Sj = the jth states of nature
Vij = the value or payoff that will be realized if alternative i is chosen and event j
occurs.

Decision situations can be categorized in to three classes:

Situation of certainty,
Situations where probabilities cannot be assigned to future occurrences and
Situations where probabilities can be assigned to future occurrences.

In this chapter we will discuss each of these classes of decision situations separately.

4.3. DECISION MAKING UNDER CERTAINTY

The simplest of all circumstances occurs when decision making takes place in an environment of
complete certainty. When a decision is made under conditions of complete certainty, the
attention of the decision maker is focused on the column in the payoff table that corresponds to
the state of nature that will occur. The decision maker then selects the alternative that would
yield the best payoff, given that state of nature.

EXAMPLE1: Doctor Thomas has been thinking of opening his own private clinic. The problem
is how large the clinic size should be under various economic scenarios. After a careful analysis,
Dr.Thomas has developed the following profit (in thousands) table.

Alternatives (Clinic Market Conditions(States of nature)

Size) Poor(S1) Fair(S2) Good(S3)
Small(A1) 4 16 12
Medium(A2) 5 6 10
Large(S3) -1 4 15
Therefore, if we know that Fair market condition (S2) will occur, the decision maker then can
focus on the first row of the payoff table. Because alternative A1 has the largest profit (16), it
would be selected.

139

Ashenafi A. (PhD Research Fellow) Page 139

4.4. DECISION MAKING UNDER COMPLETE UNCERTAINTY
Under complete uncertainty, the decision maker either is unable to estimate the probabilities for
the occurrence of the different state of nature, or else he or she lacks confidence in available
estimates of probabilities, and for that reason, probabilities are not included in the analysis.

A decision making situation includes several components- the decision themselves and the actual
event that may occur future, known as state of nature. At the time the decision is made, the
decision maker is uncertain which state of nature will occur in the future, and has no control over
them.

Decisions made under these circumstances are at the opposite end of the spectrum from the
certainty case just mentioned. Once the decision has been organized in to a payoff table, several
criteria are available making the actual decision. There are several approaches (criteria) to
decision making under complete uncertainty. Some of these discussed in this section include:
maximax, maximin, minimax regret, Hurwicz, and equal likelihood.

4.4.1. MAXIMAX CRITERIA

With the maiximax criterion, the decision maker selects the decision that will result in the
maximum of the maximum payoffs.( In fact this is how this criterion derives its name- maximum
of maximum). Tha maximax is very optimistic. The decision maker assumes that the most
favorable state of nature for each decision alternative will occur. For example, the investor would
optimistically assume that good economic conditions will prevail in the future. The best payoff
for each alternative is identified, and the alternative with the maximum of these is the designated
decision.

Example 2: For the previous problem which alternative should be selected under maximax
criteria?

S1 S2 S3 Row Maximum
A1 4 16 12 16*maximum
A2 5 6 10 10
A3 -1 4 15 15
Decision: A1 will be chosen.

Note: If the pay off table consists of costs instead of profits, the opposite selection would be
indicated: The minimum of minimum costs. For the subsequent decision criteria we encounter,
the same logic in the case of costs can be used.

4.4.2. MAXIMIN CRITERIA

This approach is the opposite of the previous one, i.e. it is pessimistic. This strategy is a
conservative one; it consists of identifying the worst (minimum) payoff for each alternative, and,
then, selecting the alternative that has the best (maximum) of the worst payoffs. In effect, the
decision maker is setting a floor on the potential payoff by selecting maximum of the minimum;
the actual payoff cannot be less than this amount. It involves selecting best of the worst.

140

Ashenafi A. (PhD Research Fellow) Page 140

Example 3: For the previous problem which alternative should be selected under maximin
criteria?

S1 S2 S3 Row minimum
A1 4 16 12 4
A2 5 6 10 5*maximum
A3 -1 4 15 -1

Decision: A2 will be chosen.

Note: If it were cost, the conservative approach would be to select the maximum cost for
each decision and select the minimum of these costs.

4.4.3. MINIMAX REGRET

Both the maximax and maximin strategies can be criticized because they focus only on a single,
extreme payoff and exclude the other payoffs. Thus, the maximax strategy ignores the possibility
that an alternative with a slightly smaller payoff might offer a better overall choice. For example,
consider this payoff table:

S1 S2 S3 Row Max.
A1 -5 16 -10 16*max
A2 15 15 15 15
A3 15 15 15 15

A similar example could be constructed to demonstrate comparable weaknesses of the maximin

criterion, which is also due to the failure to consider all payoffs.

An approach that does take all payoffs in to consideration is Minimax regret. In order to use this
approach, it is necessary to develop an opportunity loss table. The opportunity loss reflects the
difference between each payoff and the best possible payoff in a column (i.e., given a state of
nature). Hence, opportunity loss amounts are found by identifying the best payoff in a column
and, then, subtracting each of the other values in the column from that payoff. Therefore, this
decision avoids the greatest regret by selecting the decision alternative that minimizes the
maximum regret.

Example 4: For the previous problem which alternative should be selected under minimax
regret criteria?
S1 S2 S3
A1 4 16 12
A2 5 6 10
A3 -1 4 15

141

Ashenafi A. (PhD Research Fellow) Page 141

 Opportunity loss table:
S1 S2 S3
A1 5-4=1 16-16=0 15-12=3
A2 5-5=0 16-6=10 15-10=5
A3 5-(-1)=6 16-4=12 15-15=0

The values in an opportunity loss table can be viewed as potential ―regrets‖ that might be
suffered as the result of choosing various alternatives. A decision maker could select an
alternative in such a way as to minimize the maximum possible regret. This requires
identifying the maximum opportunity loss in each row and, then, choosing the alternative
that would yield the best (minimum) of those regrets.

S1 S2 S3 Max. Loss
A1 5-4=1 16-16=0 15-12=3 3*minimum
A2 5-5=0 16-6=10 15-10=5 10
A3 5-(-1)=6 16-4=12 15-15=0 12

Decision: A1 will be chosen.

Although this approach makes use of more information than either Maximin or Maximax, it
still ignores some information, and, therefore, can lead to a poor decision.
EXAMPLE:
Opportunity loss table
S1 S2 S3 S4 Max. Loss
A1 0 0 0 24 24
A2 15 15 15 0 15*minimum
A3 15 15 15 0 15*minimum

4.4.4. PRINCIPLE OF INSUFFICIENT REASON/ EQUAL LIKELIHOOD/

LAPLACE CRTERIA
The Minimax regret criterion‘s weakness is the inability to factor row differences. Hence,
sometimes the minimax regret strategy will lead to a poor decision because it ignores certain
information.

The principle of insufficient reason offers a method that incorporates more of the information. It
treats the states of nature as if each were equally likely, and it focuses on the average payoff for
each row, selecting the alternative that has the highest row average.

Example 4: For the previous problem which alternative should be selected under principle of
insufficient reason criteria?
S1 S2 S3 Row Average
A1 4 16 12 32/3=10.67* max
A2 5 6 10 21/3=7
A3 -1 4 15 18/3=6
Decision: A1 is selected

142

Ashenafi A. (PhD Research Fellow) Page 142

The basis for the criterion of insufficient reason is that under complete uncertainty, the decision
maker should not focus on either high or low payoffs, but should treat all payoffs (actually, all
states of nature), as if they were equally likely. Averaging row payoffs accomplishes this.

4.4.5. THE HURWITZ CRITERION

The Hurwitz criterion strikes a compromise between the maximax and maximin criterion. The
principle underlying this decision criterion is that the decision maker is neither totally optimistic,
nor totally pessimistic.
The Hurwicz criterion attempts to find a middle ground between the extremes posed by the
optimist and pessimist criteria. Instead of assuming total optimism or pessimism, Hurwicz
incorporates a measure of both by assigning a certain percentage weight to optimism and the
balance to pessimism.

A weighted average can be computed for every action alternative with an alpha-weight α,
called the coefficient of realism. "Realism" here means that the unbridled optimism of
maximax is replaced by an attenuated optimism as denoted by the α. Note that 0 ≤ α ≤ 1. Thus,
a better name for the coefficient of realism is coefficient of optimism. An α = 1 implies
absolute optimism (maximax) while an α = 0 implies absolute pessimism (maximin).

Selecting a value for α simultaneously produces a coefficient of pessimism 1 -α , which reflects

the decision maker's aversion to risk. A Hurwicz weighted average H can now be computed for
every action alternative Ai in A as follows:

H (Ai ) = α (column maximum) + ( 1 - α ) (column minimum)

for positive-flow payoffs (profits, income)

H (Ai ) = α (column minimum) + ( 1 - α ) (column maximum)

for negative-flow payoffs (costs, losses)

The Hurwitz criterion requires that for each alternative, the maximum payoff is multiplied by 
and the minimum payoff be multiplied by 1-.

Example 5: For the previous problem which alternative should be selected under principle of
insufficient reason criteria and if  = 0.4.

A1 = (0.4x16) + (0.6x4) = 8.8

A2 = (0.4x10) + (0.6x5) = 7
A3 = (0.4x15) – (0.6x1) = 5.4

Decision: A1 is selected

143

Ashenafi A. (PhD Research Fellow) Page 143

A limitation of Hurwicz criterion is the fact that  must be determined by the decision maker.
Regardless of how the decision maker determines, it is still a completely a subjective measure
of the decision maker‘s degree of optimism. Therefore, Hurwicz criterion is a completely
subjective decision making criterion.

Activity 2: Which decision making under uncertainty is sounds more appropriate to you?
Why?
______________________________________________________________________________
______________________________________________________________________________

4.5. DECISION MAKING UNDER RISK (WITH PROBABILITIES)

Dear learner, the decision making criteria just presented was based on the assumption that no
information regarding the likelihood of the states of the nature was available. Thus, no
probabilities of occurrence were assigned to the states of nature, except in the case of the equal
likely hood criterion.

It is often possible for the decision maker to know enough about the future state of nature to
assign probabilities to their occurrences. The term risk is often used in conjunction with partial
uncertainty, presence of probabilities for the occurrence of various states of nature. The
probabilities may be subjective estimates from managers or from experts in a particular field, or
they may reflect historical frequencies. If they are reasonably correct, they provide the decision
maker with additional information that can dramatically improve the decision making process.

Given that probabilities can be assigned, several decision criteria are available to aid the decision
maker. Some of these are discussed below.

4.5.1. EXPECTED MONETARY VALUE (EMV)

The EMV approach provides the decision maker with a value which represents an average payoff
for each alternative. The best alternative is, then, the one that has the highest EMV. The average
or expected payoff of each alternative is a weighted average:

k
EMVi = Σ Pj.Vij
i=1

Where:
EMVi = the EMV for the ith alternative
Pi = the probability of the ith state of nature
Vij = the estimated payoff for alternative i under state of nature j.
Note: the sum of the probabilities for all states of nature must be 1.

Example 6: For the previous problem which alternative should be selected under EMV criteria
and if probabilities of 20%, 30% & 50% were assigned to S1,S2 & S3 respectively ?

Probability 0.20 0.50 0.30

144

Ashenafi A. (PhD Research Fellow) Page 144

 S1 S2 S3 Expected payoff
A1 4 16 12 12.40*maximum
A2 5 6 10 7
A3 -1 4 15 6.30

Decision: A1 will be chosen.

Note that it does not necessarily follow that the decision maker will receive a payoff equal to the
expected monetary value of a chosen alternative. Similarly, the expected payoffs for either of the
other alternatives do not equal any payoffs in those rows. What, then, is the interpretation of the
expected payoff? Simply a long-run average amount; the approximate average amount one could
reasonably anticipate for a large number of identical situations.

4.5.2. EXPECTED OPPORTUNITY LOSS (EOL)

The table of opportunity loss is used rather than a table of payoffs. Hence, the opportunity losses
for each alternative are weighted by the probabilities of their respective state of nature to
compute a long run average opportunity loss, and the alternative with the smallest expected loss
is selected as the best choice.

=Steps for calculating EOL:

Prepare a payoff matrix for each course of action and state of nature combination along
with the associated probabilities.
For each state of nature, calculate the opportunity loss values by subtracting each payoff
from the maximum payoff for that outcome.
Calculate EOL for each course of action by multiplying the probability of each state of
nature with the opportunity loss value and then adding the values.
Select a course of action for which the EOL value is minimum.

Example 7: For the previous problem which alternative should be selected under EOL criteria
and if probabilities of 20%, 30% & 50% were assigned to S1,S2 & S3 respectively ?

EOL (A1) = 0.20(1) + 0.50(0) + 0.30(3) = 1.10 *minimum

EOL (A2) = 0.20(0) + 0.50(10) + 0.30(5) = 6.50
EOL (A3) = 0.20(6) + 0.50(12) + 0.30(0) = 7.20

Note: The EOL approach resulted in the same alternative as the EMV approach (Maximizing
the payoffs is equivalent to minimizing the opportunity losses).

4.5.3. EXPECTED VALUE OF PERFECT INFORMATION (EVPI)

It can sometimes be useful for a decision maker to determine the potential benefit of knowing for
certain which state of nature is going to prevail. The EVPI is the measure of the difference
between the certain payoffs that could be realized under a condition involving risk.

Example 7: For the previous problem calculate the EVPI.

145

Ashenafi A. (PhD Research Fellow) Page 145

If the decision maker knows that S1 will occur, A2 would be chosen with a payoff of $5.
Similarly for S2 $16 (for A1) and for S3, $15 (with A3) would be chosen.

Hence, the expected payoff under certainty (EPC) would be:

EPC = 0.20(5) + 0.50(16) + 0.30(15) = 13.50

The difference between this figure and the expected payoff under risk (i.e., the EMV) is the
expected value of perfect information. Thus:

EVPI = EPC – EMV

= 13.50 – 12.40 = 1.10

Note: The EVPI is exactly equal to the EOL. The EOL indicates the expected opportunity loss
due to imperfect information, which is another way of saying the expected payoff that could be
achieved by having perfect information.

Note: The expected value approach is particularly useful for decision making when a number of
similar decisions must be made; it is a long-run approach. For one-shot decisions, especially
major ones, other methods (perhaps, maximax or maximin) may be preferable. In addition, non
monetary factors, although not included in a payoff table, may be of considerable importance.
Unfortunately, there is no convenient way to include them in an expected value analysis.

Activity 3: Making decisions under different decision making environments

ABC‘s decision to produce a new line of recreational products has resulted in the need to
construct either a small plant or a large plant. The best selection of plant size depends on how
the marketplace reacts to the new product line. To conduct an analysis, marketing
management has decided to view the possible long run demand as low, medium, or high. The
following payoff table shows the projected profit in millions of Birrs:
Plant size Long run demand

Low Medium High

Small 150 200 200
Large 50 200 500
What is the decision to be made?
What is the chance event for ABC‘s problem?
Construct a decision tree for this problem
What are the recommended decision using the optimistic, conservative, and
minimax regret approaches?
If the marketing manager knows for sure that the long run demand of recreational projects is
low; what will be his/her decision to get the best payoff?
For the payoff table given above (in problem 1), suppose the manager has assigned
probabilities of 0.3 to the occurrence of state low demand and 0.5 to the occurrence of
medium demand.
Which alternative would be chosen using maximum expected value as the
criterion?
Calculate the expected value of perfect information.

146

Ashenafi A. (PhD Research Fellow) Page 146

Comment: Read the notes given before this activity and answer the questions as correct as
possible. Note that the probability of the occurrence of high demand is 0.20 (20%) as the sum of
probabilities is always 1 (100%).

4.6. DECISION THEORY AND DECISION TREES

Decision trees are excellent tools for making financial or number based decisions where a lot of
complex information needs to be taken into account. They provide an effective structure in
which alternative decisions and the implications of taking those decisions can be laid down and
evaluated. They also help you to form an accurate, balanced picture of the risks and rewards
that can result from a particular choice.

4.6.1. What is a decision tree?

Decision tree is a classifier in the form of a tree structure. So far, the decision criteria are
applicable to situations where we have to make a single decision. There is another technique for
analyzing a decision situation. A decision tree is a schematic diagram consisting of nodes and
branches. There are some advantages of using this technique:

It provides a pictorial representation of the sequential decision process. Each process

requires one payoff table.
It is easier to compute the expected value (opportunity loss). We can do them directly on
the tree diagram.
More than one decision maker can be easily involved with the decision processes.

There are only three rules governing the construction of a decision tree:

Branches emanating from a decision node reflect the alternative decisions possible at that point.

If a decision situation requires a series of decisions, then a payoff table approach cannot
accommodate the multiple layers of decision - making. A decision tree approach becomes the
best method. However, a tree diagram will contain nothing but more squares (decision nodes).
We can still apply the expected value criterion at each decision node from the right to the left of
tree diagram (backward process).

A decision tree can be used to classify an example by starting at the root of the tree and moving
through it until an event node occurs, which provides the classification of the instance.

Decision tree induction is a typical inductive approach to learn knowledge on classification. The
key requirements to do mining with decision trees are:

147

Ashenafi A. (PhD Research Fellow) Page 147

 Attribute-value description: object or case must be expressible in terms of a fixed
collection of properties or attributes. This means that we need to discretize continuous
attributes, or this must have been provided in the algorithm.
Predefined classes (target attribute values): The categories to which examples are to be
assigned must have been established beforehand (supervised data).
Discrete classes: A case does or does not belong to a particular class, and there must be more
cases than classes.
Sufficient data: Usually hundreds or even thousands of training cases.

4.6.2. How to Draw a Decision Tree?

Decision tree will be Started with a decision that needs to be made. This decision is represented
by a small square towards the left of a large piece of paper. From this box draw out lines
towards the right for each possible solution, and write that solution along the line. Keep the
lines apart as far as possible so that you can expand your thoughts.

At the end of each solution line, consider the results. If the result of taking that decision is
uncertain, draw a small circle. If the result is another decision that needs to be made, draw
another square. Squares represent decisions, circles represent uncertainty or random factors.
Write the decision or factor to be considered above the square or circle. If you have completed
the solution at the end of the line, just leave it blank.

Starting from the new decision squares on your diagram, draw out lines representing the options
that could be taken. From the circles draw out lines representing possible outcomes. Again mark
a brief note on the line saying what it means. Keep on doing this until you have drawn down as
many of the possible outcomes and decisions as you can see leading on from your original
decision.
An example of the sort of thing you will end up with is shown below:

148

Ashenafi A. (PhD Research Fellow) Page 148

 PORTRAY OF A DECISION TREE

Once this is done, review your tree diagram. Challenge each square and circle to see if there are
any solutions or outcomes you have not considered. If there are, draw them in. If necessary,
redraft your tree if parts of it are too congested or untidy.
You should now have a good understanding of the range of possible outcomes.

4.6.3.Starting to Evaluate Your Decision Tree

Now you are ready to evaluate the decision tree. This is where you can calculate the decision
that has the greatest worth to you. Start by assigning a cash or numeric value to each possible
outcome - how much you think it would be worth to you.

Next look at each circle (representing an uncertainty point) and estimate the probability of each
outcome. If you use percentages, the total must come to 100% at each circle. If you use
fractions, these must add up to 1. If you have data on past events you may be able to make
rigorous estimates of the probabilities. Otherwise write down your best guess.

4.6.4. Calculating Tree Values

Once you have worked out the value of the outcomes, and have assessed the probability of the
outcomes of uncertainty, it is time to start calculating the values that will help you make your
decision.

We start on the right hand side of the decision tree, and work back towards the left. As we
complete a set of calculations on a node (decision square or uncertainty circle), all we need to
do is to record the result. All the calculations that lead to that result can be ignored from now on
- effectively that branch of the tree can be discarded. This is called 'pruning the tree'.

149

Ashenafi A. (PhD Research Fellow) Page 149

4.6.5. Calculating the Value of Decision Nodes
When you are evaluating a decision node, write down the cost of each option along each
decision line. Then subtract the cost from the value of that outcome that you have already
calculated. This will give you a value that represents the benefit of that decision.

Sunk costs, amounts already spent, do not count for this analysis. When you have calculated the
benefit of each decision, select the decision that has the largest benefit, and take that as the
decision made and the value of that node.

Example 8: An executive has to make a decision. He has four alternatives D1, D2, D3 and D4.
When the decision has been made events may lead such that any of the four results may occur.
The results are R1, R2, R3 and R4. Probabilities of occurrence of these results are as follows:

R1 = 0.5, R2 = 0.2, R3 = 0.2, R4 = 0.1

The matrix of pay-off between the decision and the results is indicated below:

R1 R2 R3 R4
D1 14 9 10 5
D2 11 10 8 7
D3 9 10 10 11
D4 8 10 11 13
Required: Show this decision situation in the form of a decision tree and indicate the most
preferred decision and corresponding expected value.

Solution
A decision tree, which represents possible courses of action and states of nature are shown in
the following figure. In order to analyses the tree, we start working backward from the end
branches.

150

Ashenafi A. (PhD Research Fellow) Page 150

 The most preferred decision at the decision node 1 found is by calculating expected value of
each decision branch and selecting the path (course of action) with high value.

The expected monetary value of node A, B and C is calculate as follows: EMV (A) = 0.5 x 14 +
0.2 x 9 + 0.2 x 10 + 0.1 x 5 = 11.3
EMV (B) = 0.5 x 11 + 0.2 x 10 + 0.2 x 8 + 0.1 x 7 = 9.8
EMV (C) = 0.5 x 9 + 0.2 x 10 +0.2 x 10 + 0.1 x 11 = 9.6
EMV (D) = 0.5 x 8 + 0.2 x 10 + 0.2 x 11 + 0.1 x 13 = 9.5

Since node A has the highest EMV, the decision at node 1 will be to choose the course of action
D4.

Example 9: A farm owner is seriously considering of drilling farm well. In the past, only 70%
of wells drilled were successful at 200 feet of depth in the area. Moreover, on finding no water
at 200 ft., some persons drilled it further up to 250 feet but only 20% struck water at 250 ft. The
prevailing cost of drilling is Br 50 per feet. The farm owner has estimated that in case he loss

151

Ashenafi A. (PhD Research Fellow) Page 151

 not get his own well, he will have to pay Rs. 15,000 over the next 10 years (in PV terms) to buy
water from the neighbor. The following decisions can be optimal:

do not drill any well, (ii) drill up to 200 ft. (iii) if no water is found at 200 ft., drill further up
to 250 ft.

Required: Draw an appropriate decision tree and determine the farm owner‘s strategy under
EMV approach.

Solution
The given data can easily be represented by the following decision tree diagram.
Consequence of outflow($)

There are two decision points in the tree indicated by 1 and 2. In order to decide between the
two basis alternatives, we have to fold back (backward induction) the tree from the decision
point 2, using EMV as criterion:

152

Ashenafi A. (PhD Research Fellow) Page 152

 EVALUATION OF DECISION POINTS

Decision point State of Nature Probability Cash outflows Expected cash

Outflow

Decision at point D2
1. Drill up to Water struck 0.2 $ 12,5000 $ 2,500
250 fit.

No water struck 0.8 27,500 22,000

EMV (outflows) = 24,500
2. Do not drill EMV (outflow) = $ 25,000

up to 250 ft.
The decision at D2 is : Drill up to 250 feet.
Decision at point D2
1. Drill up to Water struck 0.7 $ 10,000 $. 7,000

200 ft.
Not water 0.3 24,500 7,350
struck
EMV (outflow) = $. 14,350
2. Do not drill EMV (outflow) = $15,000

up to 200 ft.
The decision at D1 is : Drill up to 200 ft.

Thus the optimal strategy for the farm-owner is to drill the well up to 200 ft. and if no water is
struck, then further drill it up to 250 ft.

Activity 6: Applying decision tree

A businessman has two independent investments A and B available to him; but he lacks the
capital to undertake both of them simultaneously. He can choose to take A first and then stop, or
if A is successful then take B, or vice versa. The probability of success of A is 0.7, while for B
it is 0‘4. Both investments require an initial capital outlay of Br. 2,000, and both return nothing
if the venture is unsuccessful. Successful competitions of A will return Br. 3,000 (over cost),
and successful completion of B will return Br. 5,000 (over cost). Draw the decision tree and
determine the best strategy.

Answer:
The appropriate decision tree is shown below:

153

Ashenafi A. (PhD Research Fellow) Page 153

 There are three decision points in the above decision tree indicated by D1, D2 and D3.

EVALUATION OF DECISION POINTS

Decision point Outcome Probability Conditional Expected
Values Values
D3 (i) Accept A Success 0.7 Br. 3000 Br. 2100
Failure 0.3 - Br. 2000 - Br. 600
Br. 1500
(ii) Stop 0
D2 (i) Accept B Success 0.4 Br. 5000 Br. 2000
Failure 0.6 - Br. 2000 - Br. 1200
Br. 800
Stop 0
D1 (i) Accept A Success 0.7 Br .3000 + 800 Br. 2660
Failure 0.3 - Br. 2000 - Br. 600

(ii) Accept B Success 0.4 Br. 5000 + 1500 Br. 2600

Failure 0.6 - Br. 2000 - Br. 1200
Br. 1400
(iii) Do Nothing 0

Hence, the best strategy is to accept A first, and if it is successful, then accept B.

154

Ashenafi A. (PhD Research Fellow) Page 154

4.6.6. Strengths and Weakness of Decision Tree Methods
The strengths of decision tree methods are:
Decision trees are able to generate understandable rules.
Decision trees perform classification without requiring much computation.
Decision trees are able to handle both continuous and categorical variables
Decision trees provide a clear indication of which fields are most important for prediction or
classification.

The weaknesses of decision tree methods

Decision trees are less appropriate for estimation tasks where the goal is to predict the value
of a continuous attribute.
Decision trees are prone to errors in classification problems with many class and relatively
small number of training examples.
Decision tree can be computationally expensive to train. The process of growing a decision
tree is computationally expensive. At each node, each candidate splitting field must be
sorted before its best split can be found. In some algorithms, combinations of fields are used
and a search must be made for optimal combining weights. Pruning algorithms can also be
expensive since many candidate sub-trees must be formed and compared.
Decision trees do not treat well non-rectangular regions. Most decision- tree algorithms only
examine a single field at a time. This leads to rectangular classification boxes that may not
correspond well with the actual distribution of records in the decision space.

4.7.SUMMARY
Decision theory is a general approach to decision making. It is very useful for a decision maker
who must choose from a list of alternatives, knowing that one of a number of possible future
states of nature will occur and that this will have an impact on the payoff realized by a particular
alternative.

Decision models can be categorized according to the degree of uncertainty that exists relative to
the occurrence of the states of nature. This can range from complete knowledge about which
state will occur to partial knowledge (probabilities) to no knowledge (no probabilities, or
complete uncertainty). When complete uncertainty exists, the approach a decision maker takes in
choosing among alternatives depends on how optimistic or pessimistic she or he is, and it also
depends on other circumstances related to the eventual outcome or payoff. Under complete
certainty, decisions are relatively straight forward. Under partial uncertainty, expected values
often are used to evaluate alternatives. An extension of the use of expected values enables
decision makers to assess the value of improved or perfect information about which state of
nature will occur.

In decision making analysis, managers can use tools such as payoff table, or decision trees.
Sometimes, decision makers can improve the decision process by taking into account additional
information, which enables them to modify state of nature probabilities. Because there is almost
always an additional cost associated with obtaining the additional information, the decision
maker must decide whether the expected value of that information is worth the cost necessary to
obtain it.

155

Ashenafi A. (PhD Research Fellow) Page 155

 Decision trees provide an effective method of decision making because they:
clearly lay out the problem so that all choices can be viewed, discussed and challenged
provide a framework to quantify of the values of outcomes and the probabilities of
achieving them
help us to make the best decisions on the basis of our existing information and best
guesses.
As with all decision-making methods, though, decision tree analysis should be used in
conjunction with common sense. They are just one important part of your decision-making tool
kit.

4.8. Self Assessment Questions-4

A food product company is contemplating the introduction of a revolutionary new product
with new packing to replace the existing product at much higher price (S1) or a moderate
change in the composition of the existing product with a new packaging at a small increase
in price (S2) or a small change in the composition of the existing expect the word ‗New‘
with a negligible increase in price (S3). The three possible states of nature of events are
High increase in sales (N1)
No change in sales (N2) and
Decrease in sales (N3).

The marketing department of the company worked out the payoffs in term of yearly net profits
for each of the strategies for these events (expected sales). This is represented in following
table:

State of nature Payoffs (in Birr)

Strategies S1 S2 S3
A1 7,00,000 3,00,000 1,50,000
A2 5,00,000 4,50,000 0
A3 3,00,000 3,00,000 3,00,000
Which strategy should the executive concerned choose on the basis of :
Maximin criterion,
Maximax criterion,
(iii) Hurwicz criterion (with α = 0.7)

Dr. Mohan Lal has been thinking about starting his own independent nursing home. The
problem is to decide how large nursing home should be? The annual returns will depend on
both the size of nursing home and a number of marketing factors. After a careful analysis,
Dr. Mohan Lal developed the following table:
Size of Good market Fair market Poor market
Nursing home (Br (Br.) (Br.)
Small 50,000 20,000 ─ 10,000
Medium 80,000 30,000 ─ 20,000
Large 100,000 30,000 ─ 40,000
Very large 300,000 25,000 ─ 160,000
a) Develop a decision table for this decision.

156

Ashenafi A. (PhD Research Fellow) Page 156

 What is the maximax decision?
What is the maximin decision?
What is the criterion of realism decision? Use α value of 0.8.
The proprietor of a food stall has invented a new food-delicacy, which he calls Whim. He
has calculated that the cost of manufacture is Re. 1 per piece and because of its novelty it
can be sold for Br.3 per piece. It is, however, perishable, and good unsold at the end of the
day are a dead loss. He expected the demand to very between 10 and 15.

Required: How many pieces should be manufactured so that his net profit is maximum? Use
Maximin and Minimax criteria.

157

Ashenafi A. (PhD Research Fellow) Page 157

 CHAPTER FIVE
NETWORK ANALYSIS/MODEL

5.0. INTRODUCTION
In the previous chapter, we discussed about decision making. This unit describes techniques that
are widely used to plan and schedule projects. The key tool in such effort is a network diagram.
This unit introduces about project management tool such as PERT (project evaluation and
review technique) and CPM (critical path method).

In each section, there are in-text questions and learning activities to initiate your study and to
check whether you understood the contents of each section well. Moreover, at the end of the unit,
there are checklists, summaries and self assessment questions that will facilitate your study and
ensure your understanding of the concepts in the unit. Enjoy this unit!

After studying this chapter, you should be able to:

define project;
give a general description of PERT and CPM;
explain the basic concepts in network analysis;
construct simple network diagram;
list the kinds of information that a PERT and CPM analysis can provide;
analyze networks with deterministic and probabilistic times; and
describe activity “crashing” and solve simple problems.

5.1 PERT and CPM Network

This section presents the basic concepts of project management. It defines project, applies PERT
and CPM in project management.

What is a Project?
Activity 1: how do you define a project? (You can use the space left below to write your
response).
_____________________________________________________________________________
_____________________________________________________________________________

Managers are responsible for planning, scheduling and controlling projects that consists of
numerous separate jobs or tasks performed by a variety of departments and individuals. Project
managers must schedule and coordinate the various jobs or activities so that the entire project is
completed on time. Project can simply be defined as a series of interdependent tasks usually
directed toward some major output and requiring significant resources and period of time to
perform. Projects are unique, one time operations designed to accomplish a specific set of
objectives during a limited time frame. Examples of projects include: constructing a shopping
complex, installing a new computer system, moving a firm to a new location, launching a space
shuttle and introducing a new product or service to the market place. Time, cost and performance
objectives are the three key issues in project management. PERT and CPM have proven to be
extremely valuable in project management.

158

Ashenafi A. (PhD Research Fellow) Page 158

A project manager looks for procedures that will help him/her answer questions such as the
following:
What is the total time to complete a project?
What are the scheduled start and finish time for each specific activity?
Which activities are "critical" and must be completed exactly as scheduled to keep the
project on schedule?
How long can "non-critical" activities are delayed without causing an increase in total
project completion time?
What is the cost of accomplishing activities?
Basic concepts in network analysis
Activity 2: what are the basic concepts in network analysis? (You can use the space left below to
write your response).
_____________________________________________________________________________
_____________________________________________________________________________

Dear learner, the following are some of the basic concepts to be understood in network
analysis.
Events/Nodes: Events/nodes in the network diagram represent project
millstones, such as the start or the completion of an activity and occur at a
particular instant of time at which some specific part of the project has been
achieved. Events are commonly represented by circles (nodes) in the network
diagram.

Activities: Activities in the network diagram represent project operations or

tasks to be conducted. An arrow is commonly used to represent an activity with
its head indicating the direction of progress in the project. Activities are
identified by the number of their starting event and ending event. An arrow (i, j)
between two events; the initial event i represent the start of the activity and the
terminal event j represents the completion of the activity as shown below:

Activity j
i
completion
starting event
event

The activities can be further classified into the following three categories:
Predecessor activity: An activity which must be completed before one or more
other activities started.
Successor activity: An activity which started immediately after one or more of
other activities is completed.

159

Ashenafi A. (PhD Research Fellow) Page 159

 Dummy activity: An activity which does not consume either any resource and /or
time. It is depicted by dotted line in the net work diagram. A dummy activity in
the network is added only to establish the given precedence relationship among
activities of the project and is needed when (i) two or more parallel activities in a
project have same head and tail events , or (ii) two or more activities have some
(but not all) of their immediate predecessor activities in common.

Rules of network construction

Following are some of the rules which have to be followed while construction a network:
In network diagram, arrows represent activities and circles the events. The length of an arrow
is of no significance.
Each activity should be represented by only one arrow and must start and end in a circle
called event. The tail of an activity represents the start, and head the completion of work.
The event numbered 1 denotes start of the project and is called initial event. All activities
emerging from event 1 should not be preceded by any other activities. Event carrying the
highest number denotes the completion time. A network should have only one initial event and
only one terminal event.
The general rule for numbering the event is that the head event should always be numbered
larger than that at its tail.
An activity must be uniquely identified by its starting and completion event which implies
that
An event number should not be repeated.
Two activities should not be identified by the same completion vent.
Activities must be represented either by their symbols or by the corresponding ordered
pair of starting-completion events.
The logical sequence between activities must follow these rules:
An event cannot occur until all the incoming activities into have been completed.
An activity cannot start unless all the preceding activities on which it depends have
been completed.
Though a dummy activity does not consume either any resource or time, it has to
follow the rules 6(a) and 6 (b).

Example1: Consider the following example: The owner of a shopping center is planning to
modernize and expand the current 32-buisnes shopping center complex. The project is expected
to provide room for 8 to 10 new businesses. The owner of the center has to plan, schedule and
complete the expansion project with the minimum possible time.

Let us see how PERT and CPM can help. The first step is to develop a list of the activities that
make up the project.

160

Ashenafi A. (PhD Research Fellow) Page 160

 List of the activities for the shopping project
Activity Description Immediate Activity time
predecessor (in weeks)
A prepare architectural drawings - 5
B identify new tenants - 6
C develop prospectus for tenants A 4
D select contractor A 3
E prepare building permits A 1
F obtain approval for building permits E 4
G perform construction D,F 14
H finalize with tenants B,C 12
I tenants move in G,H 2
Total 51 weeks

The immediate predecessor column identifies the activities that must be completed immediately
prior to the start of that activity. You may think that the total time required to complete the
project is 51 weeks. However, as two or more activities often may be scheduled concurrently,
shortening the completion time for the project is possible. Ultimately PERT/CPM will provide a
detailed activity schedule for completing the project in the shortest time possible.

Project network is a graphical representation of projects. The activities correspond to the nodes
of the network (drawn as rectangles). The arcs (lines with arrow) show the precedence
relationships among the activities. Nodes have also been added to the network to denote the
start and the finish time of the project.

D
G

Start
C

I
B H Finish

161

Ashenafi A. (PhD Research Fellow) Page 161

 The concept of critical path
To facilitate the PERT/CPM computations, we modified the project network with activity time
is as follows:

Project network with activity times

F
4
E
1

A
5
D
3
G
14

Start
C
4
I
B 2
6 Finish
H
12

To determine the project completion time, we have to analyze the network and identify the
critical path for the network. A path is a sequence of connected nodes that leads from the start
node to the finish node.

Example2: The following are paths for the project under consideration

Path Total time required

A-E-F-G-I 5+1+4+14+2=26 weeks
A-D-G-I 5+3+14+2=24 weeks
A-C-H-I 5+4+12+2=23 weeks
B-H-I 6+12+2=20 weeks

Look for the path that requires the longest time. Because all other paths are shorter in duration,
this longest path determines the total time required to complete the project. If activities on the
longest path are delayed, the entire project will be delayed. Thus, the longest path is the
critical path. Activities on the critical path are called critical activities.

Determining the critical path

We begin by finding the earliest start time and a latest start time for all activities in the
network.

162

Ashenafi A. (PhD Research Fellow) Page 162

 Let ES = earliest start time for an activity
EF = earliest finish time for an activity
t = activity time
The earliest finish time for any activity is: EF = ES+t

Activity A can be started as soon as the project starts, so we set the earliest start time for activity A
is zero. With an activity time of 5 weeks, the earliest finish time for activity A is EF = 0+5=5

We will write the earliest start time and earliest finish times in the node to the right of the
activity letter as:

A 0 5
5 ES EF

Because an activity cannot be started until all immediately preceding activities have been
finished, the following rule can be used to determine the earliest start time for each activity. The
earliest start time for an activity is equal to the largest of the earliest finish times for all
immediately preceding activities.

Let us find the EF of all activities:

Activity ES t EF
A 0 5 5
B 0 6 6
C 5 4 9
D 5 3 8
E 5 1 6
F 6 4 10
G 10 14 24
H 9 12 21
I 24 2 26

Note that the earliest finish time for activity I, the last activity in the project, is 26 weeks.
Therefore, we now know that the total completion time for the project is 26 weeks.

163

Ashenafi A. (PhD Research Fellow) Page 163

 Project Network with Earliest Start and Earliest Finish Times

E56
F610
1 4

05
5
D58
3
G1024
14

Start
C59
4
I2426
2
B06 H921
6 12 Finish

We now continue the algorithm for finding the critical path by making backward pass through
the network. If the project can be completed in 26 weeks, we begin the backward pass with a
latest finish time of 26 for activity I. Once the latest finish time for an activity is known, the
latest start time for an activity can be computed as follows: Let LS = latest start time for an
activity
LF = latest finish time for an activity
t = activity time
Then, LS=LF-t→ LF=LS+t
Activity I: LS=LF-t=26-2=24

We will write the LS and LF values in the node directly below the earliest start and earliest
finish times. This for node I, we have:
I 24 26
2 24 26

The following rule can be used to determine the latest finish time for each activity in the
network. The latest finish time for an activity is the smallest of the latest start times for all
activities that immediately follow the activity.

164

Ashenafi A. (PhD Research Fellow) Page 164

 Activity LS LF t
A 0 5 5
B 6 12 6
C 8 12 4
D 7 10 3
E 5 6 1
F 6 10 4
G 10 24 14
H 12 24 12
I 24 26 2

Latest Start and Latest Finish Times Shown in Each Node

E56 F610
156 4 6 10

05
5 05
D58
3 7 10
G1024
14 10 24

Start
C59
4 8 12
I 2426
2 24 26
B 06 H921
6 6 12 12 12 24 Finish

After we have completed the forward and backward pass, we can determine the amount of slack
associated with each activity. Slack is the length of time an activity can be delayed without
increasing the project completion times.
Slack=LS-ES=LF-EF

For example, the slack associated with activity C is LS-ES=8-5= 3 weeks. Hence, an activity C
can be delayed up to 3 weeks and the entire project can still be completed in 26 weeks. In this
sense, activity C is not critical to the completion of the entire project in 26 weeks.

Next we consider activity E: LS-ES=5-5= 0. Thus, activity E has no slack. Thus, this activity
cannot be delayed without increasing the completion time for the entire project. In other words,
completing activity E exactly as scheduled is critical in terms of keeping the project on

165

Ashenafi A. (PhD Research Fellow) Page 165

 schedule. Thus, activity E is critical activity. In general, the critical activities are the activities
with zero slack. One of the primary contribution of PERT/CPM is the identification of the
critical activities. The project manager will want to monitor critical activities closely because a
delay in anyone of these activities will lengthen the project completion time.

Activity schedule for the shopping center project

Activity ES LS EF LF Slack Critical path?
A 0 0 5 5 0 yes
B 0 6 6 12 6 -
C 5 8 9 12 3 -
D 5 7 8 10 2 -
E 5 5 6 6 0 yes
F 6 6 10 10 0 yes
G 10 10 24 24 0 yes
H 9 12 21 24 3 -
I 24 24 26 26 0 yes

Because A, E, F, G and I have zero slack, these activities are the critical activities for the
project. The path formed by nodes (A-E-F-G-I) is the critical path in shopping center project.

Contribution of PERT/CPM
Project managers look for procedures that will help answer important questions regarding the
planning, scheduling, and controlling of projects. Let us reconsider these questions in light of
the information that the critical path calculations have given us.
How long will the project take to complete? Answer: The project can be completed
in 26 weeks.
What are the scheduled start and finish time for each activity? Answer: Consider the
activity schedule above.
Which activities are "critical" and must be completed exactly as scheduled to keep the
project on schedule? Answer: A, E, F, G, and I are the critical activities.
How long can "non-critical" activities be delayed before they cause an increase in
total project completion time? Answer: Consider the above activity schedule.

Let us summarize the PERT/CPM critical path procedure:

Develop a list of the activities that make up the project.
Determine the immediate predecessor(s) for each activity in the project.
Estimate the completion time for each activity.
Draw a project network showing the activities and immediate predecessors listed in steps
1 and 2.
Use the project network and the activity time estimates to determine the earliest start and
earliest finish times for each activity by making a forward pass through the network.
The earliest finish time for the last activity in the project identifies the total time required
to complete the project.
Use the project completion time identified in step 5 as the latest finish time for the last
activity and make a backward pass through the network to identify the latest start and
latest finish time for each activity.

166

Ashenafi A. (PhD Research Fellow) Page 166

 Determine the slack for each activity.
Determine the critical activities.
Use the information from step 5 and 6 to develop the activity schedule for the project.

Activity 3: Applying the CPM/PERT

Activity Immediate predecessor Time
A - 3
B - 5
C - 7
D C 3
E A,B 7
F E,D 3
G D 2
H F,G 2
Draw the project network
Develop the activity schedule for the project
Identify the critical path

Comment: You should draw the project network and identify the critical path. And from the
network, you can develop the schedule for the project.

5.2. PROJECT CRASHING

Activity 4: what is crashing of a project? (You can use the space left below to write your
response).
_____________________________________________________________________________
_____________________________________________________________________________

The shortening of activity times, which usually can be achieved by adding resources, is referred
to as crashing. Added resources such as more workers, overtime, and so on generally increase
project costs, so the decision to reduce activity times must take into consideration the additional
cost involved. In effect, the project manager has to make a decision that involves trading
reduced activity time for additional project cost. That is, the manager will want to identify the
activities that cost the least to crash and then crash those activities only the amount necessary to
meet the desired project completion time.

For crashing to be made, we must obtain for the following information:

Normal and crash time estimate for each activity.
Normal cost and crash cost estimate for each activity
A list of activities that are on a critical path.

We shall assume that the cost of crashing an activity per period is given by the following
formula:
crash cost - normal cost
normal time - crash time

We can summarize the method for reducing the project completion time as follows:

167

Ashenafi A. (PhD Research Fellow) Page 167

 Compute the cost of reducing activity time by one unit time for each activity.
Identify all the critical paths. Select one activity on each critical path (same activity can
be on more than one critical path) such that the total cost of crashing all these activities
by one unit time is the minimum among all such combinations of the activities. Further
the activities selected should be such that their current activity time is higher than the
crash time, i.e. no activity reduction in activity duration can take place beyond its crash
time. If there is at least one critical path, on which none of the activities can be crashed,
then no further reduction in the project completion time can take place. Reduce the
activity time for the activities selected for crashing by unit time period and recalculate
the early start time, late start time and critical paths for the network.
Repeat steps from a to c.

Example:
Activity Immediate predecessor Time (week)
A - 5
B - 3
C - 8
D A 4
E B 4

Project Network with Activity Times

A D
5 4

Start E
B 4
3
Finish

C
8

168

Ashenafi A. (PhD Research Fellow) Page 168

 The crashing data for the project are as follows:
Activity Time in weeks Cost in Birr Crash cost per
Normal Crash time Normal Crash week
time
A 5 4 600 800 200
B 3 1 400 600 100
C 8 5 900 1200 100
D 4 2 600 1200 300
E 4 3 500 700 200
3000 4500

Paths
A-D (5+4=9 weeks)
B-E (3+4= 7 weeks)
C (8 weeks)

Thus, project completion time is 9 weeks. If the project completion time is to be reduced by 1
week, it can be done only by reducing the activity duration of one of the activities on the critical
path. If there are more than one critical paths, then at least one activity on each path has to be
crashed. You may note that reducing activity time of any activity other than critical activities
will not reduce the length of critical path and hence the project duration cannot be reduced.

For our example, critical path consists of activities A and D. If the duration of either A or D is
reduced by 1 week, the length of this path will be reduced to 8 weeks and hence project duration
will be reduced to 8 weeks.

The cost of reducing the activity A and D by 1 week is Birr 200 and Birr300 respectively.
Hence, it will be beneficial to crash activity A by 1 week and total cost becomes Birr 3200.

Project Network with Activity Times after Crashing By One Week

A D

4 4

Start E
B 4
3
Finish

C
8

169

Ashenafi A. (PhD Research Fellow) Page 169

 Both the paths A-D and C are critical in this network. To reduce the project duration further, at
least one activity on each of the critical paths has to be reduced. It implies that activity C with
either activity A or activity D has to be crashed. However, it is not possible to crash activity A
as already it is being done in its crash time and no further reduction is possible. Hence, we can
reduce the project duration by 1 week by crashing C and D, which will cost Birr400. We can
repeat this, process for further reduction in the project duration.

Project Network with Activity Times

A D
4 3

Start
B E
3 4 Finish

C
7

Critical path Crash one activity at each critical path Cost increased
by
A-D D will be crashed Birr300
B-E B will be crashed as it is least cost Birr100
C C will be crashed Birr100

Project Network with Activity Times

A
D
4 2

Start B E
2 4

Finish

C
6

170

Ashenafi A. (PhD Research Fellow) Page 170

 Note: Now the project completion time is 6 weeks after we crashed activities D, B and C.

Critical path Crash one activity at each Costincreased

critical path by
A-D No further crash is possible -
B-E B will be crashed as it is least Birr100
cost
C C will be crashed Birr100

Project Network with Activity Times

D
A 2

Start E
4
B
1
Finish

C
5

Note also that the project completion time is 6 weeks if we crash activities B and C. As we
know the main purpose of crashing is to shorten the project completion time. Thus, if we don‘t
shorten the project completion time from 6 weeks, why should we crash activities B and C and
incur additional cost of crashing? Thus, we should not crash activity B and C as crashing them
doesn‘t shorten the project completion time.

Project Crashing (Summary)

Current Critical Activities Cost of Activities Additional Total cost
project Paths which can crashing selected Cost due to
completion be crashed crashing
time
9 A-D A,D 200, 300 A 200 3200
8 A-D, C D,C 300, 100 D,C 400 3600
7 A-D, C, D, C,B,E 300, 100, D,C,B 500 4100
B-E 100, 200
6 A-D, C, C,B,E 100, 100, Cannot be
B-E 200 crashed

171

Ashenafi A. (PhD Research Fellow) Page 171

 5.3. PROBABILISTIC TIME ESTIMATES
Activity 5: how do we make a probabilistic estimate of project time? (You can use the space
left below to write your response).
_____________________________________________________________________________
_____________________________________________________________________________

In certain projects, it seems unrealistic to assume that we can know with certainty the time
durations in which the activities can be completed. In such cases three time estimates are made.
These estimates are
Most Likely Time (TM): The time which is taken most frequently by the activity.
Optimistic Time (TO): The time by which activity can be completed, if everything
went well.
Pessimistic Time (TP): The time by which the activity will get completed even under
adverse conditions

From the above time estimates expected time (ET) of an activity is computed using the
following relationship:
ET = TO+ 4TM +TP
6
The expected time for an activity represents the average time it would take if the activity is
performed over and over again. It is different from most likely time. Further, consider two
distributions, with the following activity times, in weeks, activity A and B.

Activity A Activity B
TO 3 2
TM 5 4
TP 7 12

Find the expected time of completing each activity.

= 3+4(5)+7 =
ETA 5weeks
6
= 2+4(4)+12 =
ETB 5weeks
6
Both the activities have same expected time of 5 weeks but there is a higher degree of
variability associated with activity B. This variability can be measured by computing the
standard deviation of the activity, SD (i), as

SD = TP(i) − TO(i)
i
6
SD A = 7 − 3 = 2/3
6
SD B = 12−2 = 5/3
6

172

Ashenafi A. (PhD Research Fellow) Page 172

 Variance is defined as the square of the standard deviation and hence
TP(i) TO(i)
V (i) = (SD )2 =( − )2
i 6
Therefore,

V(A) = (2/3)2 = 4/9

V(B) = (5/3)2 = 25/9

Time of the Project Completion

Unlike the deterministic case, when there is uncertainty in activity times, it is not possible to
compute the time of completion of the project with certainty. Instead, we shall estimate the
expected time and variance of the project completion time. The expected project completion
time E(T) can be computed as the length of critical path in the deterministic time estimates,
when activity times are replaced by the expected times (TE) of the activities. Further a crude
estimate of the variance of the completion time can be obtained by adding the variances of the
activities on the critical path.

Activity 6: Applying the CPM/PERT

Explain the following terms in PERT/CPM:
Earliest time
Latest time
Total activity time
Critical path
Slack
Crashing
Most likely time
Optimistic time
Pessimistic time

Consider the following project:

Activity Immediate predecessor Expected time
(in months)
A - 4
B - 6
C A 2
D A 6
E C,B 3
F C,B 3
G D,E 5
The project must be completed in 18 months. Do you anticipate difficulty in meeting the
deadline? The project is complete when activities F and G both complete.
How much time will be needed to complete this project?

173

Ashenafi A. (PhD Research Fellow) Page 173

 Comment: Calculate the project duration of the activities on the critical path. If the duration of
the critical path is less or equal to than 18 month, we can meet the deadline. That is, the
duration of activities on the critical path determine the project completion time.

5.4.SUMMARY
Projects are composed of a unique set of activities established to realize a given set of
objectives during a limited life span. The non- routine nature of project activities places a set of
demands on the project manger, which are different in many aspects than those required for the
manager of more traditional operations activities, both in planning and coordinating the work.

PERT and CPM are two commonly used techniques for developing and monitoring projects.
Although each technique was developed independently and for different purposes, time and
practice have erased most of the original differences so that now little distinction can be made
between the two. Either one provides the manager with a rational approach to project planning
along with a graphical display of project activities. Both depict the sequential relationships that
exist among activities and reveal to managers which activities must be completed on time in
order to achieve timely completion of the project. Managers can use that information to direct
their attention toward the most critical activities.

A systematic way of describing a project is through identification of activities and their

interrelationships. These interrelationships can be visualized using arrow diagrams or networks.
Project management deals with time-scheduling and resource allocation for these activities. The
slacks associated with each activity can be used for better control of the project. Project
duration can be reduced by incurring additional cost in executing activities. A project cost-time
relationship can be obtained using project crashing to obtain the optimal project duration.

5.5. Self Assessment Questions 5

1. A network consists of the activities in the following list. Times are given in weeks.

Activity Preceding Time

A -- 8
B -- 3
C A 7
D A, B 3
E C 4
F D 6
Required:
Draw the network diagram.
Calculate the ES, EF, LS, LF, and Slack for each activity.
What is project completion time?

The network below represents a project being analyzed by Critical Path Methods. Activity
durations are A=5, B=2, C=12, D=3, E=5, F=1, G=7, H=2, I=10, and J=6.
What task must be on the critical path, regardless of activity durations?

174

Ashenafi A. (PhD Research Fellow) Page 174

 b. What is the duration of path A-B-E-H-J?
c. What is the critical path of this network?
d. What is the length of the critical path?
e. What is slack time at activity H?
f. What is the Late Finish of activity H?
g. If activity C were delayed by two time units, what would happen to the project duration?

3. A network consists of the following list. Times are given in weeks.

Duration
Activity Preceding (weeks)
A -- 9
B A 2
C A 12
D A 5
E B 6
F B 8
G C, F 3
H D 2
I H 8
J G, I 6
K E, J 2

Draw the network diagram.

Which activities form the critical path?

175

Ashenafi A. (PhD Research Fellow) Page 175

 How much slack exists at activities A and F?
What is the duration of the critical path?

A partially solved PERT problem is detailed in the table below. Times are given in weeks.

Activity Preceding Optimistic Probable Pessimistic Expected Variance

Time Time Time Time
A -- 7 9 14 1.361
B A 2 2 8 0
C A 8 12 16 0
D A 3 5 10 1.361
E B 4 6 8 0
F B 6 8 10 0
G C, F 2 3 4 0
H D 2 2 8 1.000
I H 6 8 16 2.778
J G, I 4 6 14 2.778
K E, J 2 2 5 0.250

Calculate the expected time for each activity. Enter these values in the appropriate
column in the table above.
Which activities form the critical path?
What is the estimated time of the critical path?
What are the project variance and the project standard deviation?
What is the probability of completion of the project after week 40?

The network below represents a project being analyzed by Critical Path Methods.
Activity durations are indicated on the network.

176

Ashenafi A. (PhD Research Fellow) Page 176

 a) Identify the activities on the critical path.
b) What is the duration of the critical path?
c) Calculate the amount of slack time at activity H.
d) If activity I were delayed by ten time units, what would be the impact on the project
duration?

6. Three activities are candidates for crashing on a CPM network. Activity details are in
the table below.

Activity Normal Time Normal Cost Crash Duration Crash Cost

X 8 days $6,000 6 days $8,000
Y 3 days $1,800 2 days $2,400
Z 12 days $5,000 9 days $7,700

a. What is the crash cost per unit time for each of the three activities?
b. Which activity should be crashed first to cut one day from the project's duration; how much
is added to project cost?
c. Which activity should be the next activity crashed to cut a second day from the
project's duration; how much is added to project cost?

7. A network consists of the following list. Times are given in weeks.

Activity Preceding Optimistic Probable Pessimistic

A -- 5 11 14
B - 3 3 9
C -- 6 10 14
D A, B 3 5 7
E B 4 6 11
F C 6 8 13
G D, E 2 4 6
H F 3 3 9

Draw the network diagram.

Calculate the expected duration and variance of each activity.
Calculate the expected duration and variance of the critical path.
Calculate the probability that the project will be completed in less than 28 weeks.

177

Ashenafi A. (PhD Research Fellow) Page 177

 CHAPTER SIX
INTRODUCTION TO GAME THEORY
6.0.INTRODUCTION
In the previous chapter, we discussed about project management. This chapter deals with
introductory concepts in game theory. Thus, the development and application of game theory is
one of the most exciting areas in operations research. This unit explains some of this theory and
show how decision makers can make strategic moves that give them an advantage over their
competitors. It has two main sections. Section one discusses with an overview of game theory and
Section two discusses about two person zero sum game.

In each section, there are in-text questions and learning activities to initiate your study and to check
whether you understood the contents of each section well. Moreover, at the end of the unit, there are
checklists, summaries and self assessment questions that will facilitate your study and ensure your
understanding of the concepts in the unit. Therefore, please work hard to understand this unit very
well and to know the basic introductory concepts of game theory.

After studying this unit, you should be able to:

define game;
define game theory;
identify the types of games that can be played by two parties;
identify basic elements of a game; and
explain the payoff matrix in a certain game.

6.1. OVERVIEW OF GAME THEORY

This section presents the basic concepts in game theory. It deals with the definition of game,
game theory and the basic elements of game. In the business and economics literature, the term
‗game‘ refers to the general situation of conflict and competition in which two or more
competitors (or participants) are involved in decision making activities in anticipation of certain
outcomes over a period of time. The competitors are referred to as players. A player may be an
individual, a group of individuals, or an organization.

6.2. DEFINITION OF A GAME AND GAME THEORY

Activity1: How do you define a game? (You can use the space left below to write your
response).
_____________________________________________________________________________
_____________________________________________________________________________

Have you tried the above question? Good! You play games all the time with your parents,
friends and enemies, and even with your teachers. Games are played in business, politics,
diplomacy and wars. The word game may convey an impression that the subject is not
important in the larger schemes of things –that it deals with trivial pursuits such as gambling
and sports when the world is full of more weighty matters such as war, business, education,
career and relationships. Actually all these weighty matters are games. Game theory is the
science of rational behavior in interactive situation. Game theory is the formal study of
decision-making where several players must make choices that potentially affect the interests of
the other players.

178

Ashenafi A. (PhD Research Fellow) Page 178

 Game theory deals with decision situations in which two intelligent opponents have conflicting
objectives. Typical examples include launching advertisement campaigns for competing
products and planning war strategies for opposing armies. These situations are in contrast with
the ones we have studied so far, where nature is viewed as a non-malevolent opponent.

Game is the science of strategic decision-making. Any situation in which individuals must make
strategic choices and in which the final outcome will depend on what each person chooses to do
can be viewed as a game. Game is an action where there are two or more mutually aware
players and the outcome for each depends on the action of all. The reason for spending time on
game theory is that it is a tool designed for investigating the behavior of rational decision
makers in setting for which each agent‘s best action depends upon what other agents are
expected to do. As a result, game theory will prove to be very useful in providing insight
concerning the strategic behavior of firms (decision makers). Game theoretic analysis is built on
two fundamental assumptions: These are:
Rationality: game theory assumes that players are interested in maximizing their payoffs.
Common knowledge: all players know the structure of the game and that their opponents are
rational.

Games that firms play can be either cooperative or non-cooperative. A game is cooperative if the
players can negotiate binding contracts that allow them to plan joint strategies. A game is non-
cooperative if negotiation and enforcement of a binding contract are not possible.

A cooperative game can be the negotiation of two firms in an industry for a joint investment to
develop a new technology. If the firms can sign a binding contract to divide the profits from their
joint investment, a cooperative outcome that makes both parties better off is possible. It is a win-
win game. An example of a non cooperative game is a situation in which two competing firms take
each other‘s likely behavior into account and independently determine a pricing or advertising
strategy to win market share.

In a game conflict, two opponents, known as players, will each have a finite or infinite number
of alternatives or strategies. Associated with each pair of strategies is a payoff that one player
pays to the other. In games, the participants make use of deductive and inductive logic in
determining a strategy for winning.

6.3. THE BASIC ELEMENTS OF A GAME

Activity 2: What are the basic elements of a game? (You can use the space left below to write
your response).
_____________________________________________________________________________
_____________________________________________________________________________

The strategic form (normal form) of a game describes an economic setting by three elements:
Players: Each decision maker in a game is called a player. These players can be individuals
(poker game), firms in the markets, or entire nation (as in the military conflict). A player is
an agent who makes decisions in a game.

179

Ashenafi A. (PhD Research Fellow) Page 179

 Strategies: Each course of action open to a player during a game is called a strategy. Strategy is
a decision rule of players. A strategy tells a player how to behave in the settings being
modeled or is a decision rule that instructs a player how to behave over the course of the
game. Strategies are simply the choices available to players.
Payoffs: The final return to the players of a game at its conclusion is called ―payoffs.‖ The
payoffs for the firms can be profit. A player‘s payoff function describes how it evaluates
different strategies. That is, given the strategies chosen by all players, a player‘s payoff
function tells the state of well being (or welfare or utility) from players having played those
strategies. It is the objective, usually numerical, that a player in a game aims to maximize. A
payoff is a number, also called utility that reflects the desirability of an outcome to a player,
for whatever reason. When the outcome is random, payoffs are usually weighted with their
probabilities. The expected payoff incorporates the player‘s attitude towards risk.

Strategy
Activity 3: What is strategy in your view? (You can use the space left below to write your
response).
_____________________________________________________________________________
_____________________________________________________________________________

The strategy for a player is the list of all possible actions (moves or courses of action) that the
player will take for every payoff (outcome) that may arise. It is assumed that the rules governing
the choices are known in advance to the players. The outcome resulting from a particular choice is
also known to the players in advance and is expressed in terms of numerical values (e.g., money,
percent of market share, utility, etc). The particular strategy (or complete plan) by which a player
optimizes his gains or losses without knowing the competitor‘s strategies is called optimal strategy.
The expected outcome when players follow their optimal strategy is called the value of the game.

Generally, there are two types strategies are employed by players in a game.
Pure strategy: It is the decision rule which is always used by the player to select the particular
strategy (course of action). Thus, each player knows in advance of all strategies out of which
the player always selects only one particular strategy regardless of the other player‘s strategy,
and the objective of the players is to maximize gains or minimize losses.
Mixed strategy: Course of action that is to be selected on a particular occasion with some fixed
probability is called mixed strategy. Thus, there is a probabilistic situation and objective of
the players is to maximize expected gains or to minimize expected losses by making choice
among pure strategies with fixed probabilities.

Activity 4: Understanding the basics of a game theory

Define game and game theory. Give examples for a game.
Explain the following terms:
Strategy
Pure and mixed strategies
Payoff
Is a war a game? Why?
Discuss about the three basic elements of a game.

180

Ashenafi A. (PhD Research Fellow) Page 180

 5) What are the two fundamental assumptions of game theory?

Comment: Read carefully what you have learned above and answer these questions as correctly
as possible.

6.4. TWO-PERSON ZERO-SUM GAMES

This section presents about the two-person-zero sum game. A game is said to be zero-sum if for
any outcome, the sum of the payoffs to all players is zero. In a two-player zero-sum game, one
player‘s gain is the other player‘s loss, so their interests are diametrically opposed. Our treatment
of the theory of games will be limited to two-person games, that is, situations of conflict where
there are only two participants. Many management situations obviously involve the participation
of many persons and, thus, are not examples of two-person games. We should remember that the
basis underlying the generation of optimal or winning strategies in conflict situations must
respect the same principle of inductive and deductive logic, regardless of the number of
participants.

6.4.1. Payoff Matrix

Activity 5: What is payoff matrix? (You can use the space left below to write your response).
_____________________________________________________________________________
_____________________________________________________________________________
Payoff matrix: The payoffs (a quantitative measure of satisfaction a given player gets at the end
of the play) in terms of gains or losses, when players select their particular strategies (courses of
action), can be represented in the form of a matrix, called the payoff matrix. Since the game is
zero sum, the gain of one player is equal to the loss of other and vice versa. In other words, one
player‘s payoff table would contain the same amounts in payoff table of other player with the
sign changed. Thus, it is sufficient to construct payoff table only for one of the players. A two-
person game is illustrated as follows:

If a player A has m strategies represented by the subscripted letters: A1, A2, A3, …Am and player
B has n strategies represented by the subscripted letters: B1, B2, B3, ………Bn. The numbers m
and n need to be equal. The total number of possible outcomes is therefore m xn. Here, it is
assumed that each player knows not only his own list of possible courses of action but also of his
opponent. That is, there is a common knowledge. For convenience, it is assumed that player A is
always a gainer whereas player B is a loser. Let aij be the payoff which player A gains from
player B if player A chooses strategy i and player B chooses strategy j. Then the payoff matrix is
shown below:
Player B’s Strategy
Player A’s
strategy
B1 B2 … Bn
A1 a11 a12… a1n
A2 a21 a22… a2n
. . . .
. . . .
. . . .
Am am1 am2 amn

181

Ashenafi A. (PhD Research Fellow) Page 181

 By conversion, the rows of the payoff matrix denote player A‘s strategies and the column
denote player B‘s strategies. Since player A is assumed to be the gainer always, so s/he wishes
to gain as large a payoff aij as possible while player B will do his best to reach a small a value
of aij as possible. Of course, the gain to player B and loss to A must be – aij.

6.4.2. Assumptions of the Game

Each player has a finite number of possible strategies (courses of action). The list may not be
the same for each player.
Player ‗A‘ attempts to maximize gains and player B minimizes losses.
The decisions of both players are made individually prior to the play with no communication
between them.
The decisions are made simultaneously and also announced simultaneously so that neither
player has an advantage resulting from direct knowledge of the other player‘s decision.
Both players know not only possible payoffs to themselves but also of each other.

6.4.3. Pure Strategies: Games with Saddle Point

Activity 6: what are games with saddle point? (You can use the space left below to write your
response).
_____________________________________________________________________________
_____________________________________________________________________________

The selection of an optimal strategy by each player without the knowledge of the competitor‘s
strategy is the basic problem of playing games. Since the payoffs for either player provide all
the essential information, therefore, only one player‘s payoff table is required to evaluate the
decisions. By convention, the payoff table for the player whose strategies are represented by
rows (say player A) is constructed. Now the objective of the study is to know how these players
must select their respective strategies so that they optimize their payoff. Such a decision making
criterion is referred to as the minimax-maximin principle. Such principle in pure strategies
game always leads to the best possible selection of a strategy for both players.

If the maximin value=minimax value, then the game is said to have a saddle (equilibrium) point
and the corresponding strategies are called optimal strategies. The amount of payoff at an
equilibrium point is known as the value of the game. A game may have more than one saddle
point.

6.4.4. Rules to determine saddle point:

Select the minimum (lowest) element in each row of the payoff matrix and write them under
‗row minima‘ heading. Then select the largest element among these elements and
enclose it in a rectangle.
Select the maximum (largest) element in each column of the payoff matrix and write them
under ‗column minima‘ heading. Then select the lowest element among these elements
and enclose it in a circle.
Find out the elements(s) which is the same in the circle as well as rectangle and mark the
position of such element(s) in the matrix. This element represents the value of the game
and is called the saddle (or equilibrium) point.

182

Ashenafi A. (PhD Research Fellow) Page 182

 6.4.5. Mixed Strategies: Games without Saddle Point
Activity 7: What are games without saddle point? (You can use the space left below to write
your response).
_____________________________________________________________________________
_____________________________________________________________________________
In certain cases, there is no pure strategy solution for a game, i.e., no saddle point exists. In all
such cases, to solve games both the players must determine an optimal mixture of strategies to
find a saddle point. The optimal strategy mixture for each player may be determined by
assigning to each strategy probability of being chosen. The strategies so determined are called
mixed strategies because they are probabilistic combination of available choices of strategy.

For solving a 2x2 game without saddle point, the following formula is also used. If payoff
matrix for player A is given by
Player B

a11 a12 
Player A  
a a
 21 22 

Then the following formulae are used to find the value of game and optimal strategies:
a a −a
22 − a21 22 12
p1 = a ; q1 = a
11 + a22 − (a12 + a21 ) 11 + a22 − (a12 + a21 )

p a a −a a
2 = 1 − p1 ; q2 = 1 − q1 and V =
11 22 21 12
, where V is the value of the game
a
11 + a22 − (a12 + a21 )

6.4.6. The Rules (Principles) Of Dominance

Activity 8: What is rule of dominance of a game? (You can use the space left below to write
your response).
_____________________________________________________________________________
_____________________________________________________________________________

The rules of dominance are especially used for the evaluation of two-person-zero-sum games
without saddle point. Certain dominance principles are stated as follows:
For player B, if each element in column, say Cr is greater than or equal to the corresponding
elements in another column, say Cs in the payoff matrix, then the column Cr is dominated
by column Cs (i.e. rth strategy is dominated by sth strategy) and therefore column Cr can
be deleted from the payoff matrix.
For player A, if each element in row, say Rr is less than or equal to the corresponding
elements in another row, say Rs in the payoff matrix, then the row Rr is dominated by
row Rs and therefore row Rr can be deleted from the payoff matrix.
A strategy say, k can be dominated if it is inferior (less attractive) to an average of two or
more other pure strategies. In this case, if the domination is strict, then strategy k can be

183

Ashenafi A. (PhD Research Fellow) Page 183

 deleted. If strategy k dominates the convex linear combination of some other pure
strategies, then one of the pure strategies involved in the combination may be deleted.

Remarks:
Rules (principles) of dominance discussed are used when the payoff matrix is a profit matrix
for the player A and a loss matrix for player B. Otherwise the principle gets reversed.
The value of the game, in general, satisfies the expression: maximin valueVminimax
value.
A game said to be a fair game if the lower (maximin) and upper (minimax) values of the
game are equal and both equals zero.
A game is said to be strictly determinable if the lower (maximin) values of the game are
equal and both equal the value of the game.

Example: For the game with payoff matrix:

Player B's Strategy
Player A's B1 B2 B3 Row Minimum
Strategy A1 -1 2 -2 -2 Maximin
A2 6 4 -6 -6

Column 6 4 -2 Minimax
Maximum
Determine the best strategies for player A and B.
Determine the value of game.
Is the game (i) fair? (ii) strictly determinable?

Solution:
In this example, gains to player A or losses to player B are represented by the positive quantities
where as losses to A and gains to B are represented by negative quantities. It is assumed that A
wants to maximize his minimum gains from B. Since the payoffs given in the matrix are what A
receives, therefore, he is concerned with the quantities which represent the row minimums.
Now A can do no worse than receive one of these values and best of them occurs when he
chooses strategy A1. This choice provides a payoff of -2 to A when B chooses strategy B3. This
refers to A‘s choice of A1 as his maximum payoff strategy because this row contains the
maximum of A‘s minimum possible payoff from his competitor B.

Player B's Strategy

Player A's B1 B2 B3 Row Minimum
Strategy A1 -1 2 -2 -2 Maximin
A2 6 4 -6 -6

Column 6 4 -2 Minimax
Maximum

Similarly, it is assumed that B wants to minimize his losses and wishes his losses to A be as
small as possible. There are column maximums that represent the greatest payments B might

184

Ashenafi A. (PhD Research Fellow) Page 184

 have to make to A. The smallest of these losses is -2, which occurs when A chooses his course
of action, A1 and B chooses his course of action B 3. This choice of B3 by B is his minimax loss
strategy because this column amount is the minimum of the maximum possible losses. In the
immediately preceding table, the quantity -2 in the A1 row and B3 column is enclosed both in a
box and a circle. That is, it is both the minimum of the column maxima and the maximum of the
row minima. This value is referred to as the saddle point. The payoff amount in the saddle
point position is also called value of the game. For this game value of the game, V = -2, for
player A. The value of the game is always expressed from the point of view of the player whose
strategies are listed in the rows. The game is strictly determinable. Also since the value of the
game is not zero, the game is not fair.

Activity 9: Understanding two person zero sum game

Explain the following terms:
Payoff matrix
Minimax and maximin criteria
Saddle point
Zero sum game
What makes a game (i) fair? (ii) strictly determinable?
In a zero sum game, the gain of one player is the loss of another. Do you agree? Why?

Comment: Read what you have learned above carefully and answer these questions as correctly
as possible.

6.5. SUMMARY
Game theory is the formal study of conflict and cooperation. Game theoretic concepts apply
whenever the actions of several agents are interdependent. These agents may be individuals,
groups, firms, or any combination of these. The concepts of game theory provide a language to
formulate structure, analyze, and understand strategic scenarios.

Games can be cooperative or non cooperative. A game is cooperative if the players can
communicate with one another and arrange binding contracts; otherwise it is non-cooperative. In
playing either kind of game, however, the most important aspect of strategy design is to understand
your opponent‘s position, and correctly predicting the likely response to your actions. In some
games, either of the player or both may have not dominant strategy. If both players have a dominant
strategy it is very simple to determine the outcome.

6.6. Self Assessment Questions-6

Attempt the following questions as required.
Determine which of the following two person zero-sum games are strictly determinable and
fair. Give the optimum strategies for each player in the case of strictly determinable.
a)
Player A Player B
B1 B2
A1 1 2
A2 4 -3

185

Ashenafi A. (PhD Research Fellow) Page 185

 b)

Player A Player B
B1 B2
A1 -5 2
A2 -7 -4
Assume that two firms are competing for market share for a particular product. Each firm is
considering what promotional strategy to employ for the coming period. Assume that the
following payoff matrix describes the increase in market share of firm A and the decrease in the
market share of firm B. Determine the optimal strategies for each firm, assuming that this a zero
sum game.

Firm B
No Promotion Moderate Much Promotion
promotion
Firm A No Promotion 5 0 -10
Moderate Promotion 10 6 2
Much promotion 20 15 10

186

Ashenafi A. (PhD Research Fellow) Page 186

REFERENCES
D.R. Anderson, D.J. Sweeney and T.A Williams, (2003), An Introduction to Management Science:
Quantitative Approaches to Decision Making, 10th edition, Thomson South-Western, New York.
Frederick S. Hillier and Gerald J. Lieberman, (1995), Introduction to Operations Research, 8th
edition, McGraw-Hill Inc. , New York
J. K Sharma, (2005), Operations Research: Problems and Solutions, 2nd edition, Macmillan India, New
Delhi.
William J. Stevenson, (1989), Introduction to Management Science, McGraw-Hill Inc.
Hamdy A Taha, (1999). Introduction to Operations Research, PHI Limited, New Delhi.
Mustafi, C.K. (1988). Operations Research Methods and Practices, Wiley Eastern Limited, New Delhi.
Levin, R and Kirkpatrick, C.A. ( 1978). Quantitative Approached to Management, Tata McGraw Hill,
Kogakusha Ltd., International Student Edition.
Peterson R and Silver, E. A. (1979). Decision Systems for Inventory Management and Production
Planning.
Handley, G and T.N. Whitin. (1983). Analysis of Inventory Systems, PHI.

187

Ashenafi A. (PhD Research Fellow) Page 187

ANSWER KEY TO SELF ASSESSMENT QUESTIONS
Self Assessment Questions-1
I. Multiple choice questions
1c 3d 5a 7d
2d 4c 6d

II. Short Answer Questions

to define the problem.
Operations Research is multidisciplinary means that operations research is derived from a
combination of other fields of studies such as mathematics, economics, statistics,
engineering, computer science, psychology, etc.
Decision making is the process of selecting the best, based on some criteria, among the
alternatives to solve a certain problem.
Deterministic models represent problems of decision making under conditions of certainty.
Probabilistic/stochastic models deal with problems of decision making under uncertainty or
risk and the probabilities of the alternative states of nature are known.

Self Assessment Questions-2

There are numerous non-optimal mixtures that may yield more profit than some of these one-
product solutions. One of these is 20 plywood and 20 blocks, which profits $60.00. The
optimal solution is no studs, 10 sheets of plywood, and 80 concrete blocks, which earns a
profit of $65.

Maximize 0.04A + 0.06B + 0.09C + 0.11D

Subject to:
A + B + C + D ≤$100,000
.5A -.5B -.5C -.5D ≤0 (rearranged from A ≤.5(A + B + C + D))
.8A -.2B -.2C - .2D ≥0 (rearranged from A ≥.2 (A + B + C + D)) -C + D ≤0 (rearranged
from D ≤C)
.002A + .003B + .007C + .008C ≤1000

Maximum contribution occurs at A=150, B=75, where contribution= $337.50. Molding time
is completely used, as is painting time. Packing time used is 1 x 150 + 3 x 75 = 375, so
Packing has 45 hours left over.

Maximize Z=250X1+350X2
Subject to:
250X1+350X2≤40
60X1+120X2≤20
X1, X2≥0

188

Ashenafi A. (PhD Research Fellow) Page 188

 Max Z=$4,100
X1=8
X2=6

Let peanuts=X1,
Cashews= X2,
Almonds=X3,and
Pecans=X4.

Minimize Z=$1X1 + $3X2 + $5X3 + $6X4

Subject to:
X1+X2+X3+X4≥1
0.5X1 -0 .5X2 -0.5X3 - 0.5 X4 ≤0
-0.1X1 -0 .1X2 -0.1X3 +0.9 X4 ≥0
X1, X2, X3, X4≥0

The optimal solution is 200 acres in soybeans and 160 acres in sugar cane. There's not
enough labor to plant all 500 acres when 200 acres is in soybeans.

189

Ashenafi A. (PhD Research Fellow) Page 189

7.

Juice needed for one bottle of

Juice Punch A Punch B Juice Available
__________________________________________________________
Orange Juice (lt) 20 10 230
Grape Juice (lt) 5 15 120
Profit per tent $4 $3

Let X1= the No of bottles of punch A produced.

X2= the No of bottles of punch B produced.

190

Ashenafi A. (PhD Research Fellow) Page 190

 LPP Model
Maximize Z=4X1 +3X2
Subject to:
20X1 +10X2 < 230 Orange Constraint
5X1 +15X2 < 120 Grape Constraint
X1, X2 > 0 Non-negativity constraint

Standard form
Maximize Z=4x1 +3x2 + 0 s1 +0 s2+ 0 s3
Subject to:
x1+3x2 + s1 +0 s2 = 230
5x1+15x2 +0s1 + s2+ = 120
x1 , x2 , s1 , s2, > 0

Where, s1 =Unused orange juice

s2 =Unused grape juice
Initial (first) tableau
Cj 4 3
BV X1 X2 S1 S2 RHS(Q)
S1 0 20 3 1 0 230
S2 0 5 15 0 1 120
Zj 0 0 0 0 0 0
Cj - Zj 4 3 0 0

Second tableau
Cj 4 3
BV X1 X2 S1 S2 RHS(Q)
X1 4 1 1/2 1/20 0 11.5
S2 0 0 25/2 -1/4 1 62.5
Zj 4 2 1/5 0 46
Cj - Zj 0 1 -1/5 0

Third and optimal Tableau

Cj 4 3
BV X1 X2 S1 S2 RHS(Q)
X1 4 1 0 3/50 -1/25 9
X2 3 0 1 -1/50 2/25 5
Zj 4 3 0.12 0.08 51
Cj - Zj 0 0 -0.12 -0.08

Maximum Z=$51
X1= 9 bottles of punch A
X2= 5 bottles of punch B
s1 =0
s2 =0

191

Ashenafi A. (PhD Research Fellow) Page 191

 X1=2, X2 =6, s1 =0 , s2=0 and Maximum Z=$36
X1=50, X2 =0, s1 =0 , s2=50 and Maximum Z=$1,000
X1=450/13, X2 =210/13 and Min Z=$540
X1=14, X2 =33 and Min Z=$221
12.
a) -Range of optimality for C1(X1) is: 0.5≤C1≤3.75
-Range of optimality for C2(X2) is: 3.125≤C2≤8.5
-Range of optimality for C3(X3) is: 0≤C3≤12
-Range of insignificance for C4(X4) is:-∞ ≤C4≤1.35
-Range for 1st (b1) constraint is: 1.5≤b1≤9
-Range for 2nd (b2) constraint is: 1.34≤b2≤8
-Range for 3rd (b3) constraint is: 1≤b3≤∞

Self Assessment Questions-3

Step 1: The smaller value at the intersection of S1 and D1 is 80. out other cells in D1. The
smaller value at the intersection of S1 and D2 is 25 (105-80); assign25 to S1D2 and rule out
other cells in S1. The smaller value at the intersection of S2 and D2 is35 (60-25); assign 35
to S2D2 and rule out other cells in S2. The smaller value at the intersection of S2 and D3 is
70; assign 70 to S2D3 and rule out other cells in D3. The remaining assignments are 40 at
S2D4 and 50 at S3D4. These actions are summarized in the table below.

NWCM Dest.1 Dest.2 Dest.3 Dest.4 Supply

Source 1 80 25 105
Source 2 35 70 40 145
Source 3 50 50
Demand 80 60 70 90 300/300

The lowest cost in the matrix is 7; assign 30 units to S2D2 and rule out all other cells in D2.
Now the lowest cost in the table is 8; assign 50 units to S3D1 and rule out all other cells in
S3. The smallest cost is now 9; assign 45 units to S1D3 and rule out all other S1. The
remaining assignments are 60 units to S2D4, 30 units to S2D1, and 25 units to S2D3. These
assignments are summarized in the table below.

LCM Dest. 1 Dest. 2 Dest. 3 Dest. 4 Supply

Source 1 45 45

Source 2 30 30 25 60 145

Source 3 50 50

Demand 80 30 70 60 240 \ 240

192

Ashenafi A. (PhD Research Fellow) Page 192

3.

COSTS Dest. 1 Dest. 2 Dest. 3 Dest. 4 Supply

Source 1 12 18 9 11 100

Source 2 21 7 30 15 150

Source 3 8 10 14 16 50

Demand 80 60 70 90 300 \ 300

Shipments Dest. 1 Dest. 2 Dest. 3 Dest. 4 Row Total

Source 1 30 0 70 0 100
Source 2 0 60 0 90 150
Source 3 50 0 0 0 50
Col. Total 80 60 70 90 300 \ 300

The objective is $12 *30 + $9*70 + $ 7*60 + $15*90 +$8*50 = $3,160

4.
COSTS Dest. 1 Dest. 2 Dest. 3 Dest. 4 Supply
Source 1 12 18 9 11 100
Source 2 21 7 30 15 150
Source 3 8 10 14 16 50
Demand 80 60 70 90 300 \ 300

Shipments Dest. 1 Dest. 2 Dest. 3 Dest. 4 Row Total

Source 1 30 0 70 0 100
Source 2 0 60 0 90 150
Source 3 50 0 0 0 50
Col. Total 80 60 70 90 300 \ 300

The solution is optimal.

5. (a) The optimal solution is shown in the solution below. (b) The minimum cost is $675.

Shipments Dest. 1 Dest. 2 Row Total

Source 1 20 0 20
Source 2 0 30 30
Source 3 15 0 15
Source 4 10 10 20
Source 5 25 0 25
Col. Total 70 40 110 \ 110
Total Cost 675

193

Ashenafi A. (PhD Research Fellow) Page 193

 The initial solution and improvement indexes appear in the first table below. The optimal
solution appears in the second table.
NWC Dest. 1 Dest. 2 Dest. 3 Row Total
Origin 1 Ship 15 Ship 15 Save 5 30
Origin 2 Add 1 Ship 35 Ship 5 40
Origin 3 Save 3 Add 2 Ship 20 20
Col. Total 15 50 25 90\90

Shipments Dest. 1 Dest. 2 Dest. 3 Row Total

Origin 1 0 5 25 30
Origin 2 0 40 0 40
Origin 3 15 5 0 20
Col. Total 15 50 25 90\90
Total Cost $255

The optimum solution, which costs $1370, appears in the solution below. A dummy row was
added since total supply was less than total demand. Thirty units are shipped from Dummy to
Destination 3. This means that Destination 3 does not receive 30 units that it demanded. The
dummy row can be interpreted as unfilled demand, or shortfall.
Shipments Dest 1 Dest 2 Dest 3 Row Total
Source 1 0 0 20 20
Source 2 25 0 5 30
Source 3 15 60 0 75
Dummy 0 0 30 30
Col. Total 40 60 55 155 \ 155
Total Cost $1370
8. The optimal solution from Excel OM is in the table below.
Shipments Dest 1 Dest 2 Dest 3 Dest 4 Row Total
Origin 1 2 0 5 1 8
Origin 2 0 6 0 2 8
Origin 3 0 6 0 0 6
Origin 4 0 0 0 3 3
Col. Total 2 12 5 6 25\25
Total Cost 677
9.
The first solution is in the first table; cost of this solution is $1,225.
COSTS Dest. 1 Dest. 2 Dest. 3 Supply
Source 1 20 20
Source 2 30 30
Source 3 40 30 5 75
125 \
12
Demand 40 30 55 5

194

Ashenafi A. (PhD Research Fellow) Page 194

 The revision is accomplished by assigning the prohibited cell a very high cost, such as
$1,000. This solution appears in the second table; its cost is $1,535. The increase in
cost is $310.
COSTS Dest. 1 Dest. 2 Dest. 3 Supply
Source 1 10 10 20
Source 2 30 30
Source 3 30 45 75
125 \
1
2
Demand 40 30 55 5

10. The setup requires two transportation problems, each with three sources.
COSTS Plant 1 Plant 2 Plant 3 Supply
Raw material source 1 $6 $8 $9 400
Raw material source 2 $4 $7 $3 600
Candidate A $9 $10 $12 500
Demand 500 500 500 1500 \ 1500

COSTS Plant 1 Plant 2 Plant 3 Supply

Raw material source 1 $6 $8 $9 400
Raw material source 2 $4 $7 $3 600
Candidate B $11 $14 $8 500
Demand 500 500 500 1500 \ 1500
11. This problem requires two transportation problems, each with three sources.
COSTS Plant 1 Plant 2 Plant 3 Supply
Raw material source 1 $6 $8 $9 400
Raw material source 2 $4 $7 $3 600
Candidate A $9 $10 $12 500
Demand 500 500 500 1500 \ 1500

COSTS Plant 1 Plant 2 Plant 3 Supply

Raw material source 1 $6 $8 $9 400
Raw material source 2 $4 $7 $3 600
Candidate B $11 $14 $8 500
Demand 500 500 500 1500 \ 1500
The solutions to these problems appear in the tables below.
COSTS Plant 1 Plant 2 Plant 3 Supply
Raw material source 1 400 0 400
Raw material source 2 100 500 600
Candidate A 500 500
Demand 500 500 500 1500 \ 1500
Total cost $9,300

195

Ashenafi A. (PhD Research Fellow) Page 195

 COSTS Plant 1 Plant 2 Plant 3 Supply
Raw material source 1 400 400
Raw material source 2 500 100 0 600
Candidate B 500 500
Demand 500 500 500 1500 \ 1500
Total cost $9,900

The solution appears in the software-generated table below. A dummy Phase 1 machine has
been added because there is not perfect balance in the capacities of these two phases. The
"cost" of this solution is a total transit time of 12,300 minutes per week. The destination
phase Machine C required 1500 wafers, but got only 1000. All other machines in both phases
are fully utilized. Yes, the quantity transferred from machine 3 to machine B is zero.
COSTS Phase 2, Phase 2, Phase 2, Supply
machine A machine B machine C
Phase 1, machine 1 700 700
Phase 1, machine 2 200 100 300 600

Phase 1, machine 3 1000 1000

Phase 1, machine 4 700 700

Phase 1, dummy 500 500

Demand 1200 800 1500 3500 \ 3500

Total cost 12,300

a. x11 = 15, x21 = 10, x22 = 10, x32 = 16, x33 = 14, x43 = 35
Minimum Cost is: 753

x13 = 15, x22 = 20, x33 = 30, x41 = 25, x42 = 6, x43 =
4 Minimum Cost is: 542

x13=15, x22=20, x33=30, x41=25, x42=6,

x43=4 Minimum Cost is: 542

14.150 dresses of V and 50 dresses of Z by Dress Maker A

250 dresses of Y by Dress Maker B
150 dress of Z by Dress Maker C
100 dress of W and 75 dresses of X by Dress Maker
D Maximum Profit is: 1687.50
1B, 2D, 3C, 4A, Z = 32
1B, 2E, 3A, 4C, 5D, 6F, Z = 36
1C, 2F, 3E, 4A, 5D, 6B, Z = 85 defects

196

Ashenafi A. (PhD Research Fellow) Page 196

 Allen-NJ, Bush-PA, Crumb-NY, Doyle-FL, Evans-GA, Fouch-FL, Goins-VA, Z = 498;
success rate = 71.1%

Self Assessment Questions-4

1.
Maximax: alternative c with a payoff of Birr 22
Maximin: alternative d with a payoff of Birr 14
Minimax regret: alternative c with minimum regret of Birr 5
2.
Maximax: alternative b or c with a payoff of Birr 10
Maximin: alternative d with a payoff of Birr 14
Minimax regret: alternative d with minimum regret of Birr 4
3.
Alternative a with EVa=Birr 15.9
EVPI=Birr 2 (17.9-15.9)
Self Assessment Questions-5
1.
(a)

(b, c)
Results
Task Early Early Late Late Slack
Start Finish Start Finish
A 0 8 0 8 0
B 0 3 7 10 7
C 8 15 8 15 0
D 8 11 10 13 2
E 15 19 15 19 0
F 11 17 13 19 2
Project 19

(a) J;
20;
A-B-G-I-J;

197

Ashenafi A. (PhD Research Fellow) Page 197

 30;
5;
24;
no impact.
Results
Task Early Early Latest Latest Slack
Start Finish Start Finish
A 0 5 0 5 0
B 5 7 5 7 0
C 5 17 10 22 5
D 0 3 10 13 10
E 7 12 17 22 10
F 3 4 13 14 10
G 7 14 7 14 0
H 17 19 22 24 5
I 14 24 14 24 0
J 24 30 24 30 0
Project 30

3.(a) Network Diagram

paths A-D-H-I-J-K and A-C-G-J-K are critical;

A has no slack; F has 2 units
32 weeks

4.
A=9.5 B=3 C=12 D=5.5 E=6 F=8 G=3 H=3 I=9 J=7 K=2.5
A-D-H-I-J-K;
36.5;
9.53, 3.09;
0.13.

5.
Critical activities are A-C-J-K;
The critical path is 26 time units;

198

Ashenafi A. (PhD Research Fellow) Page 198

 Slack at H is 9 units;
I has 11 units slack--a ten unit delay would have no impact on the project.

Task Early Start Early Finish Late Start Late Finish Slack
A 0 5 0 5 0
B 0 3 6 9 6
C 5 16 5 16 0
D 0 4 9 13 9
E 3 10 9 16 6
F 4 10 15 21 11
G 3 7 17 21 14
H 4 7 13 16 9
I 10 15 21 26 11
J 16 22 16 22 0
K 22 26 22 26 0
Project 26
6.
crash cost X = $1,000 per day; crash cost Y = $600 per day ; crash cost Z = $900 per day
select Y, adding $600; (c) select Z, adding $900
7.
(a)

199

Ashenafi A. (PhD Research Fellow) Page 199

 (b)& c
Expected Variance Std. dev. Slack Variance
Task time

A 10.5 2.25 1.5 3

B 4 1 1 8
C 10 1.778 1.333 0 1.778
D 5 0.444 0.667 3
E 6.5 1.361 1.167 8
F 8.5 1.361 1.167 0 1.361
G 4 0.444 0.667 3
H 4 1 1 0 1
Project 22.5 Project 4.139
Std. dev. 2.034

(d) z = (28-22.5)/2.03 = 2.71, (P<=28) = .997

Self Assessment Questions-6

a) Not fair, value of the game=1
Not fair, value of the game=-5
Optimal strategy for both A and B is ‗much promotion‘; value of the game is 10.

200

Ashenafi A. (PhD Research Fellow) Page 200

 MEKANE YESUS MANGEMENT AND LEADERSHIP COLLEGE
DISTANCE EDUCATION PROGRAM
OPERATIONS RESEARCH TUTOR MARKED ASSIGNMENTS (30%)
INSTRUCTION: ANSWER ALL OF THE FOLLOWING QUESTIONS ON A
SEPARATE ANSWER SHEET.
What do we mean by models and explain different types of Models
Discuss areas into which operations research models can be applied?
A company executive must determine the optimal mix of radio and TV advertisements to
purchase for the next month. Each radio advertisement costs Birr 1,000, and each TV
advertisement costs Birr 2,000. There must be a combined total of at least 1200
advertisements. Also, each radio advertisement is expected to reach 1,000 families, and each
TV advertisement is expected to reach 1,500 families. It is required that at least 1,500,000
families be reached.
Required
Formulate the linear programming model for the problem.
How many of each type of advertisement should be purchased for the next month in order
to minimize the total cost? And also determine the minimum cost. Use the graphical
approach.
Determine the surplus value for each constraint?

A manufacturing firm produces electric motors for washing machines and vacuum cleaners.
The firm has resource constraints for production time, steel and wire. The linear programming
model for determining the number of washing machine motors (X1) and vacuum cleaner motors
(X2), to produce has been formulated as follows.
Maximize Z = 70 X1 + 80 X2
Subject to:
2X1+ X2 ≤ 19 (production, hrs)
X1 + X2 ≤ 14 (steel, m2 )
X1 + 2X2≤ 20 (wire,m2 )
X1&X2≥0
The final optimal simplex tableau for this model is as follows
Cj 70 80 0 0 0

BV X1 X2 S1 S2 S3 RHS(Q)
X1 70 1 0 2/3 0 -1/3 6
S2 0 0 0 -1/3 1 -1/3 1
X2 80 0 1 -1/3 0 2/3 7
Zj 70 80 20 0 30 980
Cj-Zj 0 0 -20 0 -30

Required: Perform analysis and determine the optimal range for:

coefficients of basic objective function
coefficients of non-basic objective function
RHS values

201

Ashenafi A. (PhD Research Fellow) Page 201

5. Solve the following LPM using Simplex Method.
Minimize Z=7X1+3X2+4X3
Subject to:
4X1+7X2+ 2X3=112
3X1+4X2+ 2X3≥100
3X1+6X2+2 X3≤150
X1, X2, X3≥0
6. Below is a transportation problem and find an initial feasible solution using:
a) North West corner method
b) Least cost method
c) Vogel‘s Approximation Method
d) Calculate total transportation cost for a, b, and c.
e) Evaluate /test the initial feasible solution found by NWCM or in (a) above using
stepping stone method and compute total transportation cost associated with it.
1 2 3 4 5 Supply
A 20 25 27 20 25 50
B 18 21 22 24 20 80
C 19 17 20 18 19 100
Demand 30 40 60 40 60

7. Mekane Yesus Management and Leadership College(MYMLC) Management department

head has five instructors to be assigned to four different courses. All the instructors have taught
the courses in the past and have been evaluated by the students. The rating for each instructor for
each course is given in the following table (a perfect score is 100):
Course
Instructor Risk Management Operations Operations Statistics for
and Insurance Management Research Management I
Belay 80 75 90 85
Solomon 95 90 90 97
Endale 85 95 88 91
Sosena 93 91 80 84
Tekalign 91 92 93 88

The department head wants to know the optimal assignment (using Hungarian Assignment
Method) of instructors to courses to maximize the overall average evaluation. The instructor
who is not assigned to teach a course will be assigned to grade exams and which instructor is not
assigned to teach a course?
A given project has the following activities with their allocated time and precedence
relationship, as shown below:

Activity Immediate Time(weeks)

Predecessor
A - 3
B A 5
C A 4

202

Ashenafi A. (PhD Research Fellow) Page 202

 D B 8
E D 9
F D 7
G D 4
H B,C 8
I G,H 6
J E,F,I 9
Required:
Draw the network diagram for the project.
Calculate the activity schedule showing the ES, EF, LS, and LF and calculate
the slacks (floats for each activity.
Find the critical path.
Find the expected project completion time.

203

Ashenafi A. (PhD Research Fellow) Page 203