NATURE OF MANAGEMENT SCIENCE

Management science is a discipline that attempts to aid

managerial decision making by applying a scientific approach to
managerial problems or opportunities that involve quantitative
factors.
Therefore, management science is closely associated with
decision making, that employs quantitative techniques in the
analysis of managerial problems.

Management scientist don’t make managerial decisions,

managers do.
Management scientist diagnoses a managerial problem, then
chooses the appropriate management science techniques to
apply in analysing the problem or opportunity and make
recommendations to managers based on the quantitative factors.
The managers then take into account various qualitative factors
that are outside the realm of management science and then use
their best judgement to make the final decision.
The traditional name of Management Science is Operation
Research and these two terms are used interchangeably. This
name was applied because the teams of scientists in World War 
were doing research on how to manage military operations.

Management science is based strongly on some scientific fields,

including mathematics and computer science. It also draws on the
social sciences, especially economics.
Definition of Management Science:
Management Science is the application of scientific methods to
the analysis and solution of managerial decision-making
problems.
The Characteristics of Management Science
1. A primary focus on managerial decision-making.
2. The application of the scientific approach to decision-making.
3. The examination of the decision situation from a broad perspective.
4. The use of methods and knowledge from several disciplines.
5. A reliance on mathematical and other quantitative models.
6. The extensive use of electronic computers.
 
It is clear that the main theme of management science is to help in
decision-making. To help the decision making process different model
or approaches are developed by many scholars. These are collectively
called Managerial Science.
Since the consequence of decision-making is unpredictable, that’s why
this science is developed to lessen the errors in decision-making.
Decision making process can range from Structured Process
(programmed) to Unstructured (non-programmed) Process.

Decision making process itself divided into three phases:

(i) Intelligence: that means searching for conditions that call for
decisions.
(ii) Design: refers to inventing, developing and analyzing possible
courses of action.
(iii) Choice: selecting a course of action from those alternatives.
Types of Decision Making:
(a)Structured Process: It is used for routine and repetitive
problems for which standard solution procedure exists. A fully
structured problem is one, where all the three phases that is,
intelligence, design and choice, are structured. In a structured
process, procedures are standardized, objectives are clear and
input and output are clearly specified.
(b) Unstructured Process: In an unstructured process, none of
the three phases, that is, intelligence, design and choice, are
structured. In an Unstructured Problem, human intuitions are still
the basis for decision making. Typical Unstructured Problem such
as planning of new services to be offered and hiring an executive.
(c) Semi-structured Process: In a Semi-structured process, some,
but not all, of the phases that is, intelligence, design and choice
are structured. The semi-structured problem fall between
structured and unstructured, involving a combination of both
standard solution procedures and individuals judgment.
Managerial activities are categorized into three broad
categories:
(a) Strategic Planning: this involves planning long-term goals
and policies for resource allocation.
(b) Management Control: this involves activities relating to
acquisition and efficient allocation / utilization of resources in
the accomplishment of organization goals
(c) Operation Control: this involves efficient and effective
execution of specific tasks or routine task.

Decision Making and Support

Several tools/ techniques were developed to support managerial
decision-making. The nature of support provided by the various
tools are shown in the following table (next slide):
This table illustrates that:
Structured decision or problem is supported by management
science models and other quantitative models such as, statistical,
financial and accounting models.

For semi-structured and unstructured decision or problem,

conventional management science models or other quantitative
models, are insufficient. Such situation requires, use of Supportive
Information System, such as, Decision Support System (DSS) and
Expert System (ES).

Therefore, management science models or techniques are applied

to only Structured problem or decision.
Types of Model: A system has three parts, input, process and
output. A system or a problem can be represented through
model. A model is a simplified representation or abstraction of
reality. It is simplified because reality is too complex to copy
exactly. Different types of models ,are as follows:
(i) Iconic (scale) Model: A physical replica of a system on a
different scale than the original. For example – car or bridge or
model.
 (ii) Analog Model: Not a real model but behaves like it. For
example – two dimensional charts or diagrams, organizational
chart, map etc.
 (iii) Mathematical Model: When the physical representation
becomes more complex, a more abstract model is used with the
aid of mathematics. Most of the Managerial Science analysis is
executed with the aid of mathematical models. It is a process that
translates observed phenomena into mathematical expressions.
Classification of Mathematical Models:
(a)Optimization Models - are used to find the best solution
under a set of restrictions and scarce resources.

(b) Prediction models -are usually provide important input

to optimization models to predict future event.

(c)Classification by type of data / information available:

Deterministic models are used in decision under certainty,
i.e. complete information is available regarding future
events. Stochastic models (probabilistic) are used in
decision making under risk where the chances of
occurrence of future events are known.
(iv) Static Model: A single snapshot of a situation. For example
– A monthly schedule of employees work hours prepared at the
beginning of the month

 (v) Dynamic Model: It is used to evaluate scenarios that

change over time. Dynamic models show trends and patterns of
overtime. Dynamic models are therefore time dependent. For
example – return on investment for next five years.
The Management Science Process
Management Science process can be described by the following
four step procedure.
Usually, parts of this process must be repeated and revised when
new data or information become available, and when assumptions
are revised .

(i)Observation - Identification of a problem or opportunity that

exists (or may occur soon) in a system or organization.

(i)Definition of the Problem - problem must be clearly and

consistently defined, showing its boundaries and noting what its
symptoms are. Often, what is described as a problem (e.g.
excessive cost) may only be a Symptom of a problem (e.g.
improper inventory levels). Because real life problem are
complicated by many interrelated factors thus sometimes it is
difficult to distinguish between symptoms and problems. While
defining a problem, it needs to be classified or conceptualized into
a definable category such as structured and unstructured are two
major category. Decision situation are also need to be classified
on the basis of data available about the situation.
If decision maker has complete information about future event,
this is known as, decision making under certainty or deterministic
situation.

When the decision maker knows the chances or probability of

occurrence of future event, this is known as, decision making
under risk or probabilistic or stochastic decision situation.

If decision maker does not know or cannot estimate the

probability of occurrence of future event, this is known as,
decision making under uncertainty.
(iii) Model Construction - Development of the functional
mathematical relationships that describe the decision variables,
objective function and constraints of the problem.
All mathematical models are comprises of three basic
components. They are: Decision variable, Result variable and
uncontrollable factors.
(i) Decision Variables –this are controllable or manipulable by
the decision maker. Such as quantities of products to produce
or the number of units to order etc.

 (ii) Result Variables: are dependent variables, depends on the

occurrence of the decision variables and the uncontrollable
factors. Such as output of the system, outcomes, payoffs etc.

 (iii) Uncontrollable Factors: In any decision situation, there are

factors (variables, constants or parameters) that affect the
result variables but are not under the control of decision maker.
For instance, prices of goods, raw material required to produce
After defining the variable and uncontrollable factors, identify the
Objective and Constraints:
= Establish the mathematical equations describing relationship
between variables.
= Simplify the model through a set of assumptions. Such as-
relationship between two variable is linear.
=Formulate the objective function to be optimized (profit,
cost).
=Formulate the requirements and/or restrictions or
constraints.
(iv) Model Solution - Models solved using management science
techniques.
Choose an Appropriate Solution Technique
An optimization algorithm- is step by step process of
searching for an optimal solution by gradually improving each
solution.
A heuristic algorithm- is a step-by-step procedure or rules,
in a finite number of steps, to arrive a optimum solution.

Generate Model Solutions

Test / Validate Model Results

=Is the solution reasonable?
=Are radical changes needed?
=Does it fit present and future plans?
=Unacceptable results? Return to modeling.

Perform “What--If” Analyses

(v) Model Implementation - Actual use of the model or its
solution.