Management Science

Engr. Pedrito A. Salvador, PhD

SISC
 Learning Objectives
Students will be able to:
Þ Describe the quantitative analysis approach
Þ Understand the application of QAM in a real
situation
Þ Discuss the meaning of Models
Þ Describe the use of modeling in QAM
Þ Discuss possible problems in using
quantitative analysis
 Introduction
 Mathematical tools have been used
for thousands of years
 MS can be applied to a wide variety
of problems
 One must understand: the specific
applicability of the technique, its
limitations and its assumptions
 The Evolution of MS
1990 Expert Systems and Artificial Intelligence
Decision Support
1980 Information System
1970 Goal Programming
Decision Theory
1960
Network Models
1950 Dynamic Programming
1940 Game Theory
Transportation
1930 Assignment Technique
1920 Inventory Control
Queuing Theory
1910
Markov Analysis
1900
 The Decision-Making Process
Quantitative Analysis
Logic
Historic Data
Marketing Research
Problem Scientific Analysis Decision
Modeling

? Qualitative Analysis
Weather
State and federal
legislation
New technological
breakthroughs
Election outcome
 Overview of Management Science

 Scientific Approach to Managerial

Decision Making
 Consider both Quantitative and
Qualitative Factors

Quantitative Meaningful
Raw Data Analysis Information
 The Management Science
Approach
 Define the problem
 Develop a model
 Acquire data
 Develop a solution
 Test the solution
 Analyze the results and perform
sensitivity analysis
 Implement the results
The MS Approach
Define the Problem

Develop a Model

Acquire Input Data

Develop a Solution

Test the Solution

Analyze the Results

Implement the Results

 Define the Problem
 All else depends on this
 Clear and concise statement required
 May be the most difficult step
 Must go beyond symptoms to causes
 Problems are related to one another
 Must identify the “right” problem
 May require specific, measurable objectives
 Develop the Model
 Model: representation of a situation
 Models: physical, logical, scale, schematic or
mathematical
 Models: variables (controllable or uncontrollable)
and parameters
 Controllable variables  decision variables
 Models must be:
 solvable
 realistic
 easy to understand
 easy to modify
 Acquire Data

 Accurate data is essential

 Data from:
 company reports
 company documents
 interviews
 on-site direct measurement
 statistical sampling
 Develop a Solution
 Manipulate the model, find the “best” solution
 Solution:
 practical
 implementable
 Various methods:
 solution of equation(s)
 trial and error
 complete enumeration
 implementation of algorithm
 Test the Solution
 Must test both
 Input data
 Model
 Determine:
 Accuracy
 Completeness of input data
 collect data from a different sources and
compare

 Check results for consistency

 Do they make sense?
 Analyze the Results
 Understand the actions implied by the solution
 Determine the implications of the action
 Conduct sensitivity analysis - change input
value or model parameter and see what
happens
 Use sensitivity analysis to help gain
understanding of problem (as well as for
answers)
 Implement the Results

 Incorporate the solution into the

company
 Monitor the results
 Use the results of the model and
sensitivity analysis to help you sell
the solution to management
 Introduction To Modeling
MS is a systematic approach to problem
solving based upon quantitative models and
computer technology.
MS mainly focuses on quantitative factors
which are measurable ingredients such as cost
or profit.
MS was developed during WWII to study
military operations. It was later used in
production and operations in bus. org.
 Modeling in the Real World

 Models are complex

 Models can be expensive
 Models can be difficult to sell
 Models are used in the real world by
real organizations to solve real
problems
 Models
 Are less expensive and disruptive than
experimenting with real world systems
 Allow “What if” questions to be asked
 Are built for management problems and
encourage management input
 Enforce consistency in approach
 Require specific constraints and goals
 Models
 A model is also a simplified
representation or abstraction of reality.
 Reality is generally too complex to copy
exactly
 Much of the complexity is actually
irrelevant in problem solving
 Use models instead of experimenting on
the real system
 Benefits of Models
1. Time compression
2. Easy model manipulation
3. Low cost of construction
4. Low cost of execution (especially that of errors)
5. Can model risk and uncertainty
6. Can model large and extremely complex systems
with possibly infinite solutions
7. Enhance and reinforce learning, and enhance
training.

Computer graphics advances: more iconic and

analog models (visual simulation)
 How to Develop a MS Model
Profit
Profits = Revenue - Expenses
Profits =
Revenue
(Price per Unit)  (Number Sold)
Expenses
- Fixed Cost
- (Variable Cost/Unit)  (Number Sold)
Profits = $10Q - $1,000 - $5Q
 How to Develop a MS Model
Q = quantity sold
Breakeven Point
F = fixed cost
V = Set
variable cost/unit
Revenue =0
PQ - F – VQ = 0
Then
F = PQ – VQ
And:
Breakeven Quantity = F/(P-V)
Q = F/(P – V)
 Models Can Help Managers to

 Gain deeper insight into the nature of business

relationships

 Find better ways to assess values in such

relationships; and

 See a way of reducing, or at least understanding,

uncertainty that surrounds business plans and

actions
 Models: The Down Side
Models
 May be expensive and time-consuming to
develop and test
 Are often misused and misunderstood (and
feared) because of their mathematical
complexity
 Tend to downplay the role and value of
nonquantifiable information
 Often have assumptions that oversimplify the
variables of the real world
 Possible Problems in Using
Models
 Define the  Acquire Input Data
 Accounting Data
Problem  Validity of Data
 Conflicting  Develop a Solution
viewpoints  Complex
 Departmental Mathematics
impacts  Only One Answer
is Limiting
 Assumptions
 Solutions become
 Develop a Model quickly outdated
 Fitting
the Model
 Understanding the
Model
 Possible Problems - Continued
 Test the Solution  Implement the
 Identifying Solution
appropriate test  Selling
the
procedures solution to others
 Analyze the
Results
 Holding all other
conditions
constant
 Identifying cause
and effect
 Models in Business Decision
Making
A model is a representation of a real situation
or object.
An MS model is a mathematical model that
represents a relationship among several
variables. (also called symbolic model)
An MS model allows a decision maker to
determine the impact of changes in decisions.
 Decision Models
 Symbolic models in which some of the
variables represent decisions that must be
made.
 An objective must be defined.
 Some Examples
 Marketing Media Selection Model
 Investment Portfolio Selection Model
 Product Mix Model
 Sales force allocation Model
 Modeling and Decision making
 Model Formulation --
 Analysis (Generating Results)
 Interpretation and Validation
 Implementation
 Building Models
 Study the environment of the managerial
situation
 Formulate a selective representation of
the situation
 Construct and analyze a symbolic model
Major Steps in Model Formulation
 Identifying the major conceptual
ingredients.
 Inputs (exogenous variables)
 Decision Variables and Parameters
 Outputs (endogenous variables)
 performance measures variables (objective
functions)
 consequence variables (other quantities that
might be useful in decision making process)
 Symbolic construction:
 Developing math equations
 Mathematical Models
Characterized by Risk
 Deterministic models - we know all
values used in the model with
certainty
 Probabilistic models - we know the
probability that parameters in the
model will take on a specific value
 Deterministic and Probabilistic
Models
 Deterministic Models
 allof relevant data are assumed to be known
with certainty.
 Linear programming model is an example.
 Probabilistic Models
 Some of variables are random.
 Decision Models under uncertainty are
examples.
Results of Decisions are Determined
by the

 Decision
 Uncontrollable Factors
 Relationships among Variables
 Degrees of Model Abstraction
(Least to Most)

 Iconic (Scale) Model: Physical replica of a system

 Analog Model behaves like the real system but does

not look like it (symbolic representation)

 Mathematical (Quantitative) Models use

mathematical relationships to represent complexity
 The Modeling Process--
A Preview
 How Much to Order for the Super King Grocery?
 Bob and Jean: How much bread to stock each day?

Solution Approaches
 Trial-and-Error
 Simulation
 Optimization
 Heuristics
 Problem Decomposition: Divide a complex problem
into (easier to solve) subproblems

 Some seemingly poorly structured problems may have

some highly structured subproblems

 Problem Ownership

Outcome: Problem Statement

 The Design Phase
 Generating, developing, and analyzing
possible courses of action

Includes
 Understanding the problem
 Testing solutions for feasibility
 A model is constructed, tested, and validated

Modeling
 Conceptualization of the problem
 Abstraction to quantitative and/or qualitative forms
 Mathematical Model

 Identify variables
 Establish equations describing their relationships
 Simplifications through assumptions
 Balance model simplification and the accurate
representation of reality

Modeling: an art and science

 The Structure of Quantitative
Models

 Mathematical expressions (e.g., equations or

inequalities) connect the components

 Simple financial model

P=R- C

 Present-value model
P = F / (1+i)n
 Descriptive Models
 Describe things as they are, or as they are believed
to be
 Extremely useful in DSS for evaluating the
consequences of decisions and scenarios
 No guarantee a solution is optimal
 Often a solution will be good enough
 Simulation: Descriptive modeling technique
 Descriptive Models
 Information flow
 Scenario analysis
 Financial planning
 Complex inventory decisions
 Markov analysis (predictions)
 Environmental impact analysis
 Simulation
 Waiting line (queue) management
 Components of
Quantitative Models
 Decision Variables
 Uncontrollable Variables (and/or Parameters)
 Result (Outcome) Variables
 Mathematical Relationships
or
 Symbolic or Qualitative Relationships
Quantitative Modeling Topics
 Model Components
 Model Structure
 Selection of a Principle of Choice
(Criteria for Evaluation)
 Developing (Generating) Alternatives
 Predicting Outcomes
 Measuring Outcomes
 Scenarios
 End