‫ب‬

Quantitative Business
Analysis (QBA)

Introduction
 What is QBA?-1

➢QBA, an approach to decision making based on the scientific method. It

is a relatively new discipline. Its contents and boundaries are not yet
fixed.
➢The scientific management revolution of the early 1900s, initiated by F.
W. Taylor, provided the foundation for the use of quantitative methods
in management. But modern Management Science research is generally
considered to have originated during the World War II period,
➢Decision-making is one of the main activities of a manager.
➢ Decision-making can be either qualitative or quantitative.
➢ In qualitative decision-making, intuition, and subjective judgment are
used. Past experience with similar problems is often an important factor
in choosing a qualitative approach, as is the complexity and importance
of a problem.
 What is QBA? -2

➢Managers tend to use a qualitative approach to decision-making when

(a) the problem is fairly simple
(b) the problem is familiar
(c) the cost involved are not great
(d) immediate decisions are needed.
➢ Conversely, managers generally prefer to use a quantitative
approach when one or more of the following conditions exist:
(a) the problem is complex
(b) the problem is not familiar
(c) the costs involved are substantial
(d) enough time is available to analyze the problem.
 What is QBA? -3
➢ Thus, in simple situations decisions are taken simply by common
sense, sound judgment and expertise without using mathematics.
➢ But when the problem is complex, the problem is especially
important (e.g. a great deal of money is involved) and the manager
desires a thorough analysis before attempting to make decision, the
problem is new and the manager has no previous experience from
which to draw and the problem is repetitive then the manager
cannot develop a good solution without the support of quantitative
analysis.
➢The tools of QBA are not from any one discipline rather Mathematics,
Statistics, Economics, Engineering; Psychology
 Why QBA?
➢The problem is complex and the manager cannot develop a good
solution without the aid of quantitative analysis.
➢The problem is especially important (i.e. a great deal of money is
involved) and the manager desires a thorough analysis before
attempting to make a decision.
➢The problem is new and the manager has no previous experience from
which to draw.
➢The problem is repetitive and the manager saves time and effort by
relying on quantitative procedures to make routine decision
recommendations.
 How QBA Can Help a Business Executive in
His/Her Decision Making?

Qualitative Information
i.e. Managers’ past
experience, intuition,
judgment etc Summary
Managerial and Decision
Problem Evaluation
Quantitative
Information i.e.
Mathematical
Models/QBA
 Origin/History of QBA
➢QBA concept is derived from the fact that ‘unity is strength’. When there
is a calamity to the nation, citizens of all shades of opinion, join together
to do their might to solve the problem.
➢ In the Second World War, there was a natural calamity to Great Britain
from German Forces of Hitler. Allied forces were threatened on land, sea
and air. Superior weapons and strategy of Germans became a real threat
to allied forces from German submarines, U-boats and aircraft.
➢Government appealed to the people and requested talents from all
walks of life to join together and find a solution to the problem to
overcome the threatening situation.
➢ Scientists from all relevant disciplines from all organizations joined
together in different teams. Each team is given a problem. These
combined efforts produced fantastic results.
➢This signaled to the birth of QBA/QAM/OR/MS/DS as a separate
discipline.
 Role of QBA in Business and Industry

➢We can use QBA for so many objectives. But we mention just a few
below:
➢Enables the management to decide when to buy and how much to buy.
➢Aid to decision-making and improve its quality.
➢Identification of optimum solution that is most appropriate.
➢Integrating the system as a whole.
➢Improve objectivity of analysis and clarity of thought.
➢Minimize the cost and maximize the profit.
➢Improve productivity and efficiency.
➢Success in competitions and market leadership.
➢Renders great help in optimum resource allocation.
 Scope of QBA
➢National plans and budgets.
➢Defense services operations.
➢Government developments and public sector units.
➢Industrial establishment and private sector units.
➢R & D and engineering divisions.
➢Public works department and construction of projects.
➢Business management and competition.
➢Agriculture and Irrigation projects.
➢Education and training.
➢Transportation and communication.
➢Home management and personal budgeting.
 Methods of QBA

➢Analytical Method
➢Trial and Error Method
➢Simulation Method
1. Analytical method: The method of solution depends on classical steps and
techniques in Mathematics like use of differential calculus, integration, set,
matrix and co-ordinate geometry. Examples are EOQ and graphical solution
for the product mix through linear programming. This is a deterministic
method.
2. Trial and error method: Some problems and models fail to yield a solution
through classical, mathematical or graphical methods. Trial and error method
is used here. In this method a certain algorithm is developed. One starting
point is an initial solution, which is the first approximation. The method of
solution is repeated with a certain set of rules so that initial solution is
gradually modified at each subsequent solution till optimal solution is reached.
There are certain criteria laid down to check whether the solution has become
an optimal solution. The classical example of trial and error method is simplex
method of linear programming. This is a deterministic method.
3. Simulation method: Solution of problems using principles of statistics,
sampling and probability is called simulation method. This method is applied
where the data is insufficient or where the situation is quite uncertain or when
it is impossible to generate data by direct measurement. In such situations,
samples are created as faithfully as possible to represent the real situation
called the ‘universe’. Random tables, mechanical devices and electronic
computers are used in order to establish random nature of events. Typical
example is Monte Carlo Simulation. This is a stochastic method.
 Phases of QBA

1. Problem identification
2. Model building
3. Model solution
4. Model validation
5. Model report generation
6. Model implementation
1. Problem identification: The problem must be precisely and concisely defined. In this step
not only the problem is defined but also uses, objectives and limitations of the study are
stressed in the light of the problem.
2. Model building: In this step the parameters or uncontrollable inputs of the problem must
be recognized, the decision variables or controllable inputs specified, and the objective
and constraints expressed quantitatively. If any of the uncontrollable inputs of the model
are subject to variation, the model is called a stochastic model otherwise the model is
referred to as a deterministic model. Mathematical models are built by translating verbal
statements into mathematical expressions involving the decision variables of the problem.
3. Solution of the model: In this step the mathematical model developed at earlier stage is
solved through scientific methods. Solution must satisfy the objective function as well as
constraints. There are various scientific methods or techniques available such as Linear
Programming, Transportation Model, Assignment Model, Game Theory, Queuing Theory,
and Statistical Methods etc.
4. Validation of the model: A model is said to be valid if it gives a reliable result (output) for
a set of inputs under the given conditions. Such validation is possible for a limited period
of time. As the time changes, original assumptions and conditions in which the model
was developed also change. Hence it is essential to check the validity of model from time
to time.
5. Report generation: After the model appears to give satisfactory results, a report must be
generated to convey the model’s solutions. This report should contain a statement of the
problem, the assumptions made and an indication of the general approach to problem
solution. Of course, a concise summary of a recommendation based on the model’s
results should be stated. The report should be written in such a way that the decision
maker easily understands it.
 Model

• A model is an abstract of reality, a simplified version of something. It

shows the relationship between cause and effect and between
objectives and constraints.
• Model is playing a significant role in business decision making.
 Models in QBA

Criterion of Categories of Models

Classification
1 Degree of Abstraction (a) Physical (b) Analogue
(c) Mathematical
2 Purpose (a) Descriptive (b) Explanatory
(c) Predictive (d) Perspectives
3 Special Behaviour (a) Static (b) Dynamic
Characteristics (c) Linear (d) Non-linear
4 Degree of Certainty (a) Deterministic (b) Probabilistic
5 Procedure or Method (a) Analytical (b) Simulation
of Solution
6 Form of Structure (a) Allocation (b) Inventory
(c) Queuing (d) Replacement
(e) Competitive
1. Physical or Iconic model: This is the representation of the real object
or situation. For an example, child’s toy truck is a model of a real
truck. Thus, structure of an atom, model of an aeroplane, photograph
of a machine, layout drawing of a factory etc. are examples of
physical model. These are also called static model.
2. Analogue model: These are abstract models mostly showing inter
and intra relationships between two or more parameters. Some
examples are histogram, frequency table, flow chart, Gantt chart,
price-demand graph etc. These are also called dynamic model.
3. Mathematical model: Here sets of relations are represented in the
form of mathematical equations, using symbols to present various
parameters.
4. Descriptive Model: The model which describes the situation under
study is called descriptive model. For example, a table showing
various demand levels and their respective frequencies is a
descriptive Model.
5. Explanatory Model: A model which explains behaviour of various
components of the system under study is called explanatory model.
For instance, correlation between share price and earning per share
may be termed as explanatory model.
6. Predictive Model: Predictive Model predicts future behaviour of
various variables on the basis of their present relationships; i.e. a
model making a demand forecast at a given level of price for a
commodity.
6. Prescriptive Model: These models suggest the preferred course
of action in a given situation such as linear programming model
or inventory model.
7. Static Model: Static models are concerned with determining an
answer for a particular set of fixed conditions that will probably
not change significantly in the short run.
8. Dynamic Model: Dynamic models take account of time factor
and admit the impact of changes generated by time.
9. Linear Model: A linear model is one where each of its
components is related to other variables linearly. Most of the OR
techniques presently available use linear models.
11. Non-Linear Model: A non-linear model is one in which one or
more components of the model exhibit non-linear behaviour.
12. Deterministic Model: In deterministic models, variables and their
relationships are stated exactly. Conditions of certainty and
perfect knowledge are presumed to exist. Same results are always
obtained from the same data. The break-even model, widely used
in business, is one of the best examples of a deterministic model.
13. Probabilistic Model: Probabilistic models are used to deal with
those situations in which outcomes of managerial actions cannot
be predicted with certainty. This implies that a decision or
strategy may result in one of several different payoffs, each with
a certain probability.
14. Allocation Model: The problem of allocation can arise whenever
one must select the level of certain activities which must compete
for certain scarce resources necessary tp perform those activities.
This allocation is made on the basis of the goal to be achieved,
such as maximization of profits or minimization of costs. The
linear programming model, the transportation model, or the
assignment models are examples of allocation models.
15. Inventory Model: Inventory models deal with a class of problems
involving the storage of idle resources until they are needed. That is,
inventories involve what is apparently the simplest operations that can
be conceived – holding or storing resources. The decisions required
generally entail the determination of how much of a resource to acquire
and how to acquire it. A large number of resources may be involved –
for example, in determining how many of each of a large number of
parts to purchase or produce and when to do so in order to minimize
total cost of managing, holding and storage.
16. Queuing Model: The formation of queues or waiting lines is a common
phenomenon which occurs whenever the current demand for a service –
exceeds the current capacity to provide that service. The objective of
this model is to allow one to determine the optimum number of
personnel or facilities necessary to service customers who arrive at
some random rate when considering the cost of service and the cost of
waiting. These types of problems are also called waiting line problems.
17. Replacement Model: Replacement models serve to solve a class of
problems in which equipment or other assets must be replaced because
of deterioration or complete failure. These models are considered in
two categories. The first deals with equipment that deteriorates
gradually, with time. The second category deals with items that have a
more or less constant efficiency with time. It often happens that items
are replaced not because they no longer perform to their designated
standards, but because more modern equipment performs to higher
standards.
15. Competitive Model: Competitive situations are characterized by
the fact that two or more individuals are making decisions in
situations that involve conflicting interests and in which the
outcome is controlled by the decisions of all the parties involved.
Many conflicting situations of this type are found in economic,
social, political, and military problems. Game theory, a
conceptual framework of great importance, provides solutions for
some of these problems.
 Advantages of Models-1
• Through a model, the problem under consideration becomes controllable.
• It provides some logical and systematic approach to the problem.
• It indicates the limitations and scope of an activity.
• Models help incorporating useful tools that eliminate duplication of
methods applied to solve any specific problem.
• Models help in finding avenues for new research and development in a
system.
• Models are often most economical and safest way to test alternative
actions, since models provide a means of predicting future.
• Models, if properly constructed, can suggest where a decision maker’s
information is in sufficient though the decision maker himself may not be
aware of it.
 Advantages of Models-2
• Models can be manipulated easily.
• Models take a complicated situation and make it simpler by eliminating
all those factors which in the decision maker’s judgment, are irrelevant
to the problem to be considered.
• Models provide a framework for decision maker. Thus, a number of
diverse considerations can be brought together in an organized fashion.
• It determines the extent of risk and uncertainties involved among
alternatives.
• Models help to formulate policy and programmes by keeping balance
with changing conditions.
• Models provide different information by evaluating alternative course
of action.
• It facilitates in short-term profit planning
 Disadvantages of Models-1
• As models are abstractions and simplifications, they
may inaccurately reflect the real situation and one
may fail to take into account all variables and
expectations.
• It is often difficult to define all the important elements
of a model in mathematical terms and set them down
on paper.
• At the end of the process of abstraction, the model
can be so complex that it becomes very difficult to
distinguish between different elements accurately.
• Models may sometimes be very expensive in relation
to the expected returns from their use.
• There is on many occasions difficultly of
communication with management and personnel who
do not understand the models and the result
obtained.
 Disadvantages of Models-2
• There is no guarantee that an investment of time and
effort in constructing the model will play dividends in the
form of satisfactory prediction.
• It may be beyond the ability of a mathematician to
manipulate the symbolic language so as to obtain useful
results.
• It is very difficult to build representative models. Some
scientists become so devoted to their model (especially if
it is a brain child) that they will insist that this model is the
real world.
• Models do not consider the qualitative aspects and it gives
no solution to the problem.
• It requires adequate knowledge and experience in model
building as well as in using models and this versatility is
not always available.
 Principles of Applying Models
The following principles are to be considered in using models:
➢To determine the pattern of decision in which model will be used.
➢The objectives of using models must clearly be expressed.
➢To decide about the person to whom the task of model building will be
entrusted.
➢To appraise the success of model in attaining the objectives.
➢Models used in an organization should be acceptable to all.
➢The model builder should have sufficient knowledge about the
strengths and limitations of model.
➢The task of building model and the function of collecting data should be
started at the same time.
➢In order to construct or validate a model, the model maker must
typically have access to information.
➢Development in model is needed to keep balance with the changing
situation.
 Modeling Process

There are five basic elements in the modeling process,

namely
➢Abstraction,
➢Validation
➢Prediction
➢Evaluation and
➢Revision.
Properties of a Good Model

1. A good model should be capable of taking into account new

formulations without having any significant changes in its frame.
2. Assumptions made in the model should be as small as possible.
3. It should be simple and coherent. Number of variables used should
be less.
4. It should be open to parametric type of treatments.
 Limitations of QBA
• Quantification
• High costs are involved in the use of QBA
• Huge time involvement in using QBA.
• QBA are just the tools of analysis and not the complete decision making
process.