No Slide Title

Download Report

Transcript No Slide Title

ISQS 5242—Decision Theory and
Management Science





1
Instructor: Dr. Burns
Telephone: 742-1547
Email: [email protected]
Off hrs: 9:00-10:35 MW
Website: burns.ba.ttu.edu
ISQS 5242







2
Prerequisites
Requirements (for completion)
Syllabus
Homework
Exams
Term Projects?
Final Grade
Prerequisites

Completion of Tool Core Courses in MBA Program
–
–



3
ISQS 5237
Economics
ISQS 5345
Accounting
Be able to solve systems of linear algebraic equations
Understand probability basics--conditional probabilities,
joint probabilities, probability trees
DROP THE COURSE if you don’t have the
prerequisites (especially since we have such a short
semester)
Text

APPLIED MANAGEMENT SCIENCE
–
–

Some material from former texts co-authored by me
–
4
Lawrence and Pasternack, 2nd edition
will cover chs 1-4, 6, part of 9 and 10 in that order
will be furnished as handouts
Software



WINQSB (for math programming, decision trees and waiting-lines
homework)
VENSIM (free--shareware)
PROMODEL, MEDMODEL OR SERVICE MODEL
–
–
–
–
–
5
you may purchase these from Promodel Corporation (optional)
cost is $30.00
call 1-801-223-4600
use a portion of your video game budget to pay
Purchase from www.promodel.com
Requirements

3 homework sets
–
–


6
Collected the day before we do our review
TO GET US READY TO TAKE THE EXAMS
3 exams
That’s it
Assigned problems: Turn in…



Hand Formulated model
Computer-generated solution
If using solver, that would consist of
–
–
–
7
Formulation worksheet
Solution report
Sensitivity report
1.1 What Is Decision Theory/Management
Science?


8
Decision Theory is the discipline that adapts the scientific
approach for problem solving to help managers make
informed decisions.
The goal of management science is to recommend the
course of action that is expected to yield the best
outcome with what is available.
1.1 What Is Management Science?

The basic steps in the management science problem
solving process involves
–
–
–
9
Analyzing business situations and building mathematical
models to describe them;
Solving the mathematical models;
Communicating/implementing recommendations based on the
models and their solutions.
The Management Science Approach



A scientific method of providing executive departments with a
quantitative basis for decisions regarding operations (Philip
McCord Morse).
Logic and common sense are basic components in supporting
the decision making process.
The use of techniques such as (US army pamphlet 660-3):
–
–
–
–
10
Statistical inference
Mathematical programming
Probabilistic models
Network and computer science
Management Science Applications


Linear Programming was used by Burger King to find how
to best blend cuts of meat to minimize costs.
Integer Linear Programming model was used by American Air
Lines to determine an optimal flight schedule.
The Shortest Route Algorithm was implemented by the Sony
Corporation to developed an onboard car navigation
 system.

11
Management Science Applications
12

Project Scheduling Techniques were used by a contractor to
rebuild Interstate 10 damaged in the 1994 earthquake in the Los
Angeles area.

Decision Analysis approach was the basis for the development of
a comprehensive framework for planning environmental policy in
Finland.

Queuing models are incorporated into the overall design plans for
Disneyland and Disney World, which lead to the development of
‘waiting line entertainment’ in order to improve customer
satisfaction.
Requirements (for completion)

Homework is worth 10% of total
–

13
Three installments each worth .0333 or 3.33%
Three exams (including FINAL), each worth 30%
Homework



14
Will be assigned every day (two problems)
Will be turned in the day before we do our review
We will do reviews just prior to each exam
Syllabus

DECISION MAKING under Certainty
–

DECISION MAKING under Uncertainty and Risk
–

Exam 2
DECISION MAKING under Change and Complexity
–
15
Exam I
Final
Exams



16
Multiple choice and some discussion problems
Bring your own orange scantron sheets
May use other formats as well
GRADING



90-100 -- A
80-89.9999 -70-79.9999 --
B
C
ISQS 5242—Summer II 2003
My Expectations




Attend class
Perform reading assignments before coming to class
Tech policy for academic honesty enforced
Assistance for Disabled students provided
ISQS 5242—Summer II 2003
How to study for exams


Read chapters before coming to class
After each class day, go over your notes
–
–


19
As soon as possible after class
Preferably with someone else
Make sure you understand everything discussed
As soon as possible do homework
Term Projects?


20
Any 3-hour 5342 students?
Any BA-7000 students?
Final Grade


21
Once your course grade is in, University Policy forbids
changing a grade for any reason other than professorial
mistake
It is impossible to do extra work to improve your final or
course grades
Main Thesis of Course:



Management science models can make us better
decision makers and problem solvers
Do we have to use models to solve problems and make
decisions!!!???
What is a model?
–
–
22
A paradigm, a description and an explanation of the
relationship of the parts of the problem to each other
Can be self-fulfilling--a caveat
Just how do you make decisions?






23
Emotional direction
Intuition
Analytic thinking
Are you an intuit, an analytic, what???
Gates: is an NT—an intuitive thinker—according to
Myers Briggs
What are you?
How about your ability to learn?


What is learning?
What is the best way to learn?
–
–

Experience??
Off-line simulation – Flight Simulators
Are there biases in your judgment?
–
–
Anchoring?
Overconfidence?

24
Commonplace in situations where there is too little information
How’s your judgment?



Enron/Global Crossing/Margaret ??/MCI WorldCom/Arthur
Andersen
What can we learn from these scenarios?
What has happened to moral judgment?
–
We have lost all explicit moral knowledge


–
Recent ruling by Supreme Court regarding Sodomy

25
We do not teach the moral knowledge base anywhere except Sunday School
Media refrains from making any moral judgments
Why remembers the story of Sodom and Gomorrah?
Problems
Arise whenever there is a perceived difference between
what is desired and what is in actuality.
 Problems serve as motivators for doing something
 Problems lead to decisions

ISQS 5242—Summer II 2003
42
27
Biases

Personally, I have a cognitive bias against teaching
cognitive biases
–
–
–
28
Availability heuristic
Anchoring heuristic
Representativeness heuristic
Precepts to Problem-solving





29
Get total picture
Withhold judgment
Change representation
Ask the right questions
Have the will to doubt
Problem-solving approaches





30
Work backwards
Generalize or specialize
Explore directions that appear plausible
Use stable substructures
Use Analogies and metaphors
A problem: Many years from now two MBA
students from Texas Tech meet on the
street. The following is part of their
discussion:
–
–
–
–
–
–
–
–

31
Man 1: Yes, I’m married and have three fine sons
Man 2: That’s wonderful! How old are they?
Man 1: Well, the product of their ages is equal to 36
Man 2: Hmm. That doesn’t tell me enough. Give me another clue.
Man 1: O.K. The sum of their ages is the number on that building
across the street
Man 2: (after a few minutes of thinking): Ah ha! I’ve almost got the
answer, but I still need another clue.
Man 1: Very well: The oldest one has red hair.
Man 2: I’ve got it!
WHAT WERE THE AGES OF THE THREE SONS, ASSUMING
AGES ARE INTEGERS
Model Classification Criteria





32
Purpose
Perspective
Degree of Abstraction
Content and Form
Decision Environment
Purpose




33
Planning
Forecasting
Training
Behavioral research
Perspective

Descriptive
–
–

Prescriptive
–
–
34
“Telling it like it is”
Most simulation models are of this type
“Telling it like it should be”
Most optimization models are of this type
Degree of Abstraction

Isomorphic
–

Homomorphic
–
35
One-to-one
One-to-many
Content and Form





36
verbal descriptions
mathematical constructs
simulations
mental models
physical prototypes
Decision Environment

Decision Making Under Certainty
–

Decision Making under Risk and Uncertainty
–

Decision analysis--tables, trees, Bayesian revision
Decision Making Under Change and Complexity
–
37
all of mathematical programming
Structural models, simulation models
Mathematical Programming


Linear programming
Integer linear programming
–

Network programming (produce all integer solutions)

Nonlinear programming
Dynamic programming
Goal programming
The list goes on and on



–
38
some or all of the variables are integers
Geometric Programming
A Model of this class

39
What would we include in it?
Management Science Models

40
A QUANTITATIVE REPRESENTATION OF A PROCESS
THAT CONSISTS OF THOSE COMPONENTS THAT
ARE SIGNIFICANT FOR THE PURPOSE BEING
CONSIDERED
Summary







41
This course: models for improved decision making
3 parts: DM under certainty
DM under risk and uncertainty
DM under change and complexity
3 exams, each worth 30%
Types of models: planning, execution, control
Models can help us substantially improve the bottom line
1.3 Mathematical Modeling


Many managerial decision situations lend themselves to
quantitative analyses.
A constrained mathematical model consists of
–
–
42
An objective
One or more constraints
1.3 Mathematical Modeling
Example

NewOffice Furniture produces three products



–
–


$50 per desk
$30 per chair
$6 per pound of molded steel
sold
Raw material required

Net profit is

43
Desks (D)
Chairs (C)
Molded steel (M)


–
7 pounds of per desk
3 pounds of per chair
1.5 pounds per one pound of
molded steel produced.
Raw material available
2000 pounds
1.3 Mathematical Modeling


44
Objective: Determine production mix that maximizes the
profit under the raw material constraint and other
production requirements (detailed next).
Maximize 50D + 30C + 6 M
Subject to 7D + 3C + 1.5M 2000 (raw steel)
D
100 (contract )
C
500 (cushions available)
D, C, M 0
(Non-negativity)
D and C are integers
Classification of Mathematical Models

Classification by the model purpose
–
–

Classification by the degree of certainty of the data in the
model
–
–
45
Optimization models
Prediction models
Deterministic models
Probabilistic (stochastic) models
The Management Science Process


46
Management Science is a discipline that adopts the
scientific method to provide management with key
information needed in making informed decisions.
The team concept calls for the formation of (consulting)
teams consisting of members who come from various
areas of expertise.
The Management Science Process

The four-step management science process (for details
click on each button)
Problem definition
Mathematical modeling
Solution of the model
47
Communication/implementation
of results
The following slides will not be
covered in class
 It
is assumed that you have good working
knowledge of spreadsheets
48
1.6 Using Spreadsheets in Management
Science Models


Spreadsheets have become a powerful tool in
management science modeling.
Several reasons for the popularity of spreadsheets:
–
–
–
49
Data are submitted to the modeler in spreadsheets
Data can be analyzed easily using statistical and mathematical
tools readily available in the spreadsheet.
Data and information can easily be displayed using graphical
tools.
Basic Excel functions and operators

Arithmetic Operations
–
–
–
–
–
50
Addition of cells A1and B1:
Subtracting cell B1 from A1:
Multiplication of cell A1 by B1:
Division of cell A1 by B1:
Cell A1xraised to the power in cell B1:
= A1 + B1
= A1 - B1
= A1 * B1
= A1 / B1
= A1^ B1
Basic Excel functions and operators

Relative and absolute addresses
–
–
All row and column references are considered relative unless
preceded by a “$” sign
When copied, ‘relative addresses’ change relative to the
original cell position.
Example:
Cell E5 =A1+B$3+$C4+$D$6
Cell G9 = C5+D$3+$C8+$D$6
51
Basic Excel functions and operators

The F4 key
–
–
Pressing F4 will automatically put a $ sign in highlighted
portions of formulas.
Specific operations:




52
Press the F4 key once: The sign “$” appears in front of all rows and
columns of the highlighted area of the formula.
Press the F4 key twice: The “$” sign appears in front of only the row
references of the highlighted area of the formula.
Press the F4 key third time: The “$” sign appears in front of only the
column references of the highlighted area of the formula.
Press the F4 key forth time: All the “$” signs are eliminated.
Basic Excel functions and operators

Arithmetic functions
–
Sum

–
–
53
=SUMPRODUCT(A1:A3,B1:B3)
Returns the sum of products A1B1+A2B2+A3B3
ABS

=Average(A1:A3)
Returns the arithmetic average of cells A1, A2, A3
SUMPRODUCT

–
Returns the sum A1+A2+A3
Average

=SUM(A1:A3)
=ABS(A3)
Returns the absolute value of the entry in cell A3.
Basic Excel functions and operators

Arithmetic functions – continued
–
SQRT

–
–
=MAX(A1:A9)
Returns the Maximum of the entries in cells A1 through A9.
MIN

54
Returns A3
MAX

=SQRT(A3)
=MIN(A1:A9)
Returns the Minimum of the entries in cells A1 through A9.
Basic Excel functions and operators

Statistical functions
–
RAND()

–
Generate a random number between 0 and 1 from a uniform distribution.
Probabilities and variable values under the normal distribution


55
=RAND()
NORMDIST
=NORMDIST(25,20,3,TRUE)
Returns P(X<25) when m = 20
and s = 3
NORMSDIST
=NORMSDIST(1.78)
Returns P(Z<1.78)
NORMINV
=NORMINV(.55,20,3)
Returns x0,, such that P(X<x0)=.55
when m = 20 and s = 3
NORMSMINV
=NORMSINV(.55)
Returns z0, such that P(Z<z0)=.55
Basic Excel functions and operators

Statistical functions
–
Probabilities and variable values under the t- distribution

TDIST
=TDIST(1.5,12,1)
Returns P(t>1.5) when n=12
Note:
=TDIST(1.5,12,2)
returns P(t<-1.5) + P(t>1.5)
when n=12.
56
TINV
=TINV(.05,15)
Returns t0,, such that
P(t<-t0)=.025 and P(t>t0)=.025
when n=15.
Basic Excel functions and operators

Statistical functions – Other probability distributions
–
Poisson

–
57
Returns P(X<7) for Poisson with l = 5.
Note: false returns the probability density P(X = 7)
EXPONDIST

=POISSON(7,5,TRUE)
=EXPONDIST(40,1/20,TRUE)
Returns P(X<40) for the exponential distribution with 1/m=20
Note: false returns the probability density f(40)=20exp(-20(40))
Basic Excel functions and operators

Conditional functions:
–
IF

–
Returns B1+B2 if A4>4, and B1 – B2 if A4
SUMIF

58
=IF(A4>4,B1+B2, B1 – B2)
=SUMIF(F1:F12, “>60”,G1:G12)
Returns G1+G2+…+G12 only if F1+F2+…+F12>60
Basic Excel functions and operators
–
VLOOKUP

=VLOOKUP(6.6,A1:E6,4)
If the values in column A of a given table [A1:E6] are sorted (in an
ascending order), VLOOKUP finds the largest value in column A that is
less than or equal to 6.6, identifies the row it belongs to, and returns the
value in the fourth column that correspond to this row.
Note: If the values in column A are not sorted,
=VLOOKUP(6.6,A1:E6,4,FALSE) finds the value 6.6 in column A,
identifies the row it belongs to, and returns the value in the fourth
column that corresponds to this row.
59
Basic Excel functions and operators

Statistical/Optimization
–
Data Analysis [Selected from the Tools menu]. Useful entries:




60
Descriptive Statistics
Regression
Exponential Smoothing
Anov