Transcript Document

DECISION MODELING WITH
MICROSOFT EXCEL
Chapter 1
Introduction to Modeling
Copyright 2001
Prentice Hall Publishers and
Ardith E. Baker
INTRODUCTION TO MODELING
Modeling Approach to Decision Making:
Involves spreadsheet based
management models
Uses spreadsheet software such as Excel®
This approach is easy for managers to use,
Results in better management decisions,
Provides important insights into problem.
THE MODELING PROCESS
Managerial Approach to Decision Making
Manager analyzes
situation (alternatives)
Makes decision to
resolve conflict
Decisions are
implemented
Consequences
of decision
These steps
Use
Spreadsheet
Modeling
THE MODELING PROCESS
as applied to the first two stages of decision making.
Real
World
Management
Situation
Results
Interpretation
Symbolic
World
Abstraction
Model
Analysis
Intuition
Decisions
THE MODELING PROCESS
The Role of Managerial Judgment in the Modeling Process:
Analysis
Real
World
Management
Situation
Results
Managerial
Judgment
Interpretation
Symbolic
World
Abstraction
Model
Decisions
Intuition
THE MODELING PROCESS
Decision Support Models force you to
1. be explicit about your objectives.
2. identify and record the types of decisions that influence
3.
4.
5.
6.
7.
those objectives.
identify and record interactions and trade-offs among
those decisions.
think carefully about which variables to include.
consider what data are pertinent and their interactions.
recognize constraints or limitations on the values.
Models allow communication of your ideas and
understanding to facilitate teamwork.
Models allow us to use the analytical power of spreadsheets hand in
hand with the data storage and computational speed of computers.
TYPES OF MODELS
Physical Model
Characteristics
Tangible
Easy to Comprehend
Difficult to Duplicate and Share
Difficult to Modify and Manipulate
Lowest Scope of Use
Examples
Model Airplane
Model House
Model City
TYPES OF MODELS
Analog Model
(A set of relationships through a different, but analogous, medium.)
Characteristics
Intangible
Harder to Comprehend
Easier to Duplicate and Share
Easier to Modify and Manipulate
Wider Scope of Use
Examples
Road Map
Speedometer
Pie Chart
TYPES OF MODELS
Symbolic Model
(Relationships are represented mathematically.)
Characteristics
Intangible
Hardest to Comprehend
Easiest to Duplicate and Share
Easiest to Modify and Manipulate
Widest Scope of Use
Examples
Simulation Model
Algebraic Model
Spreadsheet Model
MORE ON MODELS
A model is a carefully selected abstraction of reality.
Symbolic models
1. always simplify reality.
2. incorporate enough detail so that
• the result meets your needs,
• it is consistent with the data you have available,
• it can be quickly analyzed.
Decision models are symbolic models in which some of the
variables represent decisions that must or could be made.
Decision variables are variables whose values you can
control, change or set.
MORE ON DECISION MODELS
Decision models typically include an explicit performance
measure that gauges the attainment of that objective.
For example, the objective may be to maximize profit
or minimize cost in relation to a performance measure
(such as sales revenue, interest income, etc).
In summary, decision models
1. selectively describe the managerial situation.
2. designate decision variables.
3. designate performance measure(s) that reflect
objective(s).
BUILDING MODELS
To model a situation, you first have to frame it (i.e.,
develop an organized way of thinking about the situation).
A problem statement involves possible decisions and a
method for measuring their effectiveness.
Steps in modeling:
1. Study the Environment to Frame the Managerial Situation
2. Formulate a selective representation
3. Construct a symbolic (quantitative) model
BUILDING MODELS
1. Studying the Environment
Select those aspects of reality relevant to the situation
at hand.
2. Formulation
Specific assumptions and simplifications are made.
Decisions and objectives must be explicitly identified and
defined.
Identify the model’s major conceptual ingredients using
“Black Box” approach.
BUILDING MODELS
The “Black Box” View of a Model
Decisions
(Controllable)
Parameters
(Uncontrollable)
Model
Performance
Measure(s)
Consequence
Variables
BUILDING MODELS
3. Model Construction
The next step is to construct a symbolic model.
Var. Y
Mathematical relationships are developed. Graphing the
variables may help define the relationship.
Var. X
To do this, use “Modeling with Data” technique.
MODELING WITH DATA
Consider the following data.
Graphs are created to view any relationship(s) between the variables.
This is the first step in formulating the equations in the model.
CLASSIFICATIONS OF MODELS
Decision making models are classified by the business function
they address or by the discipline or industry involved.
Classification
Examples
Business Function
Finance, Marketing, Cost Accounting, Operations
Discipline
Science, Engineering, Economics
Industry
Military, Transportation, Telecommunications, Non-Profit
Time Frame
One Time Period, Multiple Time Periods
Organizational Level Strategic, Tactical, Operational
Mathematics
Linear Equations, Non-Linear Equations
Representation
Spreadsheet, Custom Software, Paper and Pencil
Uncertainty
Deterministic, Probabilistic
DETERMINISTIC AND
PROBABILISTIC MODELS
Deterministic Models
are models in which all relevant data are assumed to be known
with certainty.
can handle complex situations with many decisions and constraints.
are very useful when there are few uncontrolled model inputs
that are uncertain.
are useful for a variety of management problems.
are easy to incorporate constraints on variables.
software is available to optimize constrained models.
allows for managerial interpretation of results.
constrained optimization provides useful way to frame situations.
will help develop your ability to formulate models in general.
DETERMINISTIC AND
PROBABILISTIC MODELS
Probabilistic (Stochastic) Models
are models in which some inputs to the model are not known
with certainty.
uncertainty is incorporated via probabilities on these “random”
variables.
very useful when there are only a few uncertain model inputs and
few or no constraints.
often used for strategic decision making involving an organization’s
relationship to its environment.
ITERATIVE MODEL BUILDING
DEDUCTIVE MODELING
Models
Models
Model Building
PROBABILISTIC
MODELS
DETERMINISTIC
MODELS
Process
Models
Models
INFERENTIAL MODELING
ITERATIVE MODEL BUILDING
Deductive Modeling
focuses on the variables themselves before data are collected.
variables are interrelated based on assumptions about algebraic
relationships and values of the parameters.
places importance on modeler’s prior knowledge and judgments of
both mathematical relationships and data values.
tends to be “__data poor_” initially.
Inferential Modeling
focuses on the variables as reflected in existing data collections.
variables are interrelated based on an analysis of data to
determine relationships and to estimate values of parameters.
available data needs to be accurate and readily available.
tends to be “data rich” initially.
MODELING AND REAL
WORLD DECISION MAKING
Four Stages of applying modeling to real world
decision making:
Stage 1: Study the environment, formulate the model
and construct the model.
Stage 2: Analyze the model to generate results.
Stage 3: Interpret and validate model results.
Stage 4: Implement validated knowledge.
MODELING AND REAL
WORLD DECISION MAKING
Management
Modeling Term
Lingo
Formal Definition
Example
Decision Variable
Lever
Controllable Exogenous
Input Quantity
Parameter
Gauge
Uncontrollable Exogenous Interest Rate
Input Quantity
Consequence
Variable
Outcome
Performance
Measure
Yardstick
Endogenous Output
Variable
Endogenous Variable
Used for Evaluation
(Objective Function Value)
Investment
Amount
Commissions
Paid
Return on
Investment