DSME 6620

Decision Models and Applications

MGT578

Lecture One

Course Introduction
Introduction to Decision Making
Using Spreadsheets

1
DSME 6620
Decision Models and Applications
MGT578

Agenda

• Course overview

• Introduction to scientific decision making

• Why scientific? Judgmental bias

• Human misjudgment

• Introduction to Decision Modeling

2
DSME 6620
Decision Models and Applications
MGT578

Instructor Information

• Ehsan Bolandifar
– Has joined PHBS in 2019
– Associate Professor of Operations
Management
– Assistant Professor of Operations Management
at CUHK since 2012
• Background
– Born in Iran
– B.Sc. in Industrial Engineering (Iran University of
Science and Technology)
– M.Sc. in Industrial Engineering (Sharif University)
– PhD. in Operations Management (Washington
University in St Louis)
– Research interest: Procurement in supply chain,
Operations management interfaces with
finance, marketing, and economics Multi-sided
platforms, Healthcare operations
management
– Teaching interest: Operations Management,
Operations Analytics, Decision Modeling
3
DSME 6620
Decision Models and Applications
MGT578

Contact Information and Office Hours

• Contact Info
– Office: PHBS Building, Room 646
– Phone: 86-755-2603-2780
– Email: [email protected]
– Office Hour: Monday-Thursday, 15:30pm-
17:30pm

• Teaching Assistant: Jeff Zan

– Office: Rm 213
– Phone: 13439372695
– Email: [email protected]
– Office hour: Thus/Fri 14:00-16：00

• Lectures: Monday-Thursday, 13:30pm-

15:20pm
– Room 331

4
DSME 6620
Decision Models and Applications
MGT578

Course Materials

• Textbooks and Reading Materials

– S. C. Albright and W. Winston, Management
Science Modeling, 4th edition.
– S. Powell and K. Baker. Management Science:
The Art of Modeling with Spreadsheets

• Laptops:
– Not mandatory, but highly encouraged

• The Excel Solver add‐in

– The basic Solver add‐in provided with your
Excel software is sufficient for our purpose.
– No need to install the book’s Premium Solver.
– We will study in class how to use Solver.
• Crystal Ball
– A Monte Carlo simulation software by Oracle.

5
DSME 6620
Decision Models and Applications
MGT578

Grading

• There will be several homework assignments

and a final examination. The breakdown
towards the final grade is as follows:
Class Participation – 10%
Homework Assignments – 30%
Midterm Exam – 20%
Final Exam – 40%

• Please note that no late homework will be

accepted without informing the instructor in
advance the reasons of delay. Same policy
applies to the final exam. No make-up exam
will be given unless an excuse formally written
by the student's parents (in case of family
emergency), by a school official (in case of
school related activity), or by a physician (in
case of illness) is received by the instructor.
Non-attendance without acceptable reasons
will affect your grade unfavorably.

6
DSME 6620
Decision Models and Applications
MGT578

Conduct

• Honor Code: Students enrolled in PHBS course

work are subject to the student instituted and
managed Honor Code regarding academic
integrity. I am very serious about this subject!
Of course, if you have any concerns or
questions related to academic integrity,
please contact me without delay.

• Classroom Ethics: Please be considerate of

others. You may bring cell phones, and
laptop computers to class, but they should
not be a distraction to the class. Please avoid
late arrival and early departure from
classroom.

7
DSME 6620
Decision Models and Applications
MGT578

Course Objectives

• Introduce you to the art and science of

modeling managerial problems as a way to
enhance decision‐making performance.

• Enable you to become intelligent users of

management science techniques and
methodologies.

• Develop your confidence in utilizing

spreadsheet technology.

8
DSME 6620
Decision Models and Applications
MGT578

Approach

• Focusing primarily on the process of

quantitative modeling, and avoiding (as
much as possible) the mathematical theory
behind it.
– In other words, we will use math, but it’s not a
course about math.
• Create an active learning environment where
students play a role in their learning
experience.

• Context: Keep it simple!

– We will demonstrate the use of different
techniques and approaches through business
problems from different disciplines, such as
manufacturing, services, finance, and
marketing. But the purpose is not to teach you
these subjects.

9
DSME 6620
Decision Models and Applications
MGT578

Required Background

• Critical! You must be familiar with the basics

of Excel.
– You are not expected to be familiar with Excel
Solver or Crystal Ball in advance. You will learn
how to use these software packages during
the course.
• Basic knowledge of probability and statistics.

• Attitude matters a lot in a course of this type!

– Don’t get scared of the subject! I will make
sure the concepts and method are accessible
to each one of you.
– You don’t need to be a mathematician is
order to succeed in this course.
• If the material is “easy”, help a friend!

10
DSME 6620
Decision Models and Applications
MGT578

About Scientific Decision Making

• Scientific decision making, which is also

widely termed “management science”, is a
discipline that attempts to solve decision
making problems in the business field by
applying mathematics, statistics, and
computers.

• The roots of scientific decision making trace

to a branch of applied mathematics called
“Operations Research” (OR).

• The beginning of OR has been attributed to

the military services in WWII: The British and
American military management called
upon a large number of scientists to help
with strategic and tactical problems.

• In the industrial boom that followed WWII,

many firms became interested in operations
research as business management
became increasingly complex – Scientific
Decision Making was born …
11
DSME 6620
Decision Models and Applications
MGT578

Well-Known OR Work in WWII

Aircraft Attacks Against German Submarines

Problem:  German Submarines (U-boat) were escaping
aircraft attacks
 2% to 3% kill probability in early 1941

Questions:  What is the “best” depth to set bomb to

explode?
 What is the “best” lethal radius?

Depth in Feet ?

Lethal radius?

Result: By 1945 the attack kill probability was risen

to over 40%
12
DSME 6620
Decision Models and Applications
MGT578

Successful Applications Today

Function Description Examples
Supply chain Simulate and optimize supply Apple, Dell,
chain flows; reduce inventory Wal-Mart,
and stock-outs Amazon
Customer Identify customers with the Harrah’s,
selection, greatest profit potential; Capital One,
loyalty, and increase likelihood that they will Barclays
service want the product or service
offering; retain their loyalty
Pricing Identify the price that will Progressive,
maximize yield, or profit. Marriot
Human capital Select the best employees for New England
particular tasks or jobs, at Patriots,
particular compensation levels Oakland A’s,
Red Sox
Product and Detect quality problems early Honda, Intel
service quality and minimize them
Financial Better understand the drivers of MCI, Verizon
performance financial performance and the
effects of nonfinancial factors
Research and Improve quality, efficacy, and Novartis,
development where applicable, safety of Amazon,
products and services Yahoo

13
DSME 6620
Decision Models and Applications
MGT578
Harmonizing Qualitative and
Quantitative Analysis

A story told in Cambridge, Massachusetts:

• A young man, obviously a college student,
is unloading his purchases at a supermarket
checkout counter. A large sign above the
counter indicates that it is for customers
with eight items or less, but the young man
has at least twenty. The clerk tells him, “You
either go to MIT and you can’t read, or you
go to Harvard and you can’t count.”

• Managers need both qualitative and

quantitative skills to achieve superior
decision-making performance.

14
DSME 6620
Decision Models and Applications
MGT578

Procedure of Decision Making

Real World

Experience,
Business Intuition
Decisions
Situation

Interpretation
Abstraction

Model
Improvement

Decision-Making
Results
Model

Computer
Application

Symbolic World
15
DSME 6620
Decision Models and Applications
MGT578

Descriptive Models
OMM5705 Quantitative Decision Making Prof. Aviv Olin Business School, Washington University

• Descriptive models: Models that describe

relationships between various system
parameters, between inputs and outputs, or
Descriptive Models
demonstrate the dynamics of processes.
• Descriptive models: Models that describe relationships
between various system parameters, between inputs and
outputs, or demonstrate the dynamics of processes.
Assumptions and structural properties (e.g.,
constraints, system dynamics, behavior,
procedures, cause-and-effect relationship,
topology)

Outcome
Parameters
Math (
(e.g.,
(known, or
Model performance
unknown)
measure)

Decision variables
OMM5705 Quantitative Decision Making Prof. Aviv Olin Business School, Washington University

16
DSME 6620
Decision Models and Applications
MGT578

Prescriptive Models

• Prescriptive models: Models that prescribe

(or “recommend”) actions. Typically,
optimization models and/or heuristic
methods.
• For example, think of the advertising
budget decision problem we mentioned
earlier.

17
DSME 6620
Decision Models and Applications
MGT578

A Policy (from a mathematical point of

view)

• A Policy is a set of rules that prescribe

actions to different states of information.
– Implications:
• The state of information needs to be well-
defined.
• The set of actions needs to be well-defined.
• The rules need to be well defined.
• Policy Evaluation is the process through
which a system’s performance is assessed,
assuming that the policy is acted upon.
– Implications:
• We need a model of the system.
• The policy needs to be well-defined with
respect to the model.
• We must have metrics on the basis of which
we measure performance.

18
DSME 6620
Decision Models and Applications
MGT578

Prescriptive Models: Identification of

Efficient Portfolios
OMM5705 Quantitative Decision Making Prof. Aviv Olin Business School, Washington Univers

• An efficient portfolio is the portfolio of risky

assets that gives the lowest variance of
return among
Prescriptiveall portfolios
Models: Examplethat
5 have the
same expected return.
Identification of Efficient Portfolios
• An efficient portfolio is the portfolio of risky assets that gives the
lowest variance of return among all portfolios that have the same
• Mathematically:
expected return.
• Mathematically:
h i ll

OMM5705 Quantitative Decision Making Prof. Aviv Olin Business School, Washington Univers

19
DSME 6620
Decision Models and Applications
MGT578

Heuristics

• A Heuristic is typically an informal

method* to help solve a problem. It
is particularly used to rapidly come
to a “reasonable” (or
“close‐to‐optimal”) solution to a
complex problem.

*e.g., rules-of-thumb, educated guesses,

intuitive judgments or simply common sense.
20
DSME 6620
Decision Models and Applications
MGT578

Dialogue on Decision Making

Interviewer: Do you make regular mathematical

programming runs for scheduling the refinery?
Analyst: Oh yes!
Interviewer: Do you implement the results?
Analyst: Oh no!
Interviewer: Well, that seems odd. If you don’t
implement the results, perhaps you should stop
making the runs.
Analyst: No. No. We wouldn’t want to do that!
Interviewer: Why not?
Analyst: Well, what happens is something like this: I
make several computer runs and take them to
the plant manager. He is responsible for this
whole multimillion-dollar plumber’s paradise.
The plant manager looks at the runs, thinks
about them for a while, and then sends me
back to make more runs. This process continues
until, finally, the plant manager screws up
enough courage to make a decision.

21
DSME 6620
Decision Models and Applications
MGT578

Popular Decision Making Tools

• Statistics
– Regression
– Time Series Forecasting

• Optimization
– Linear Programming
– Integer Programming
– Nonlinear Programming

• Decision analysis

• Simulation

22
DSME 6620
Decision Models and Applications
MGT578

tentative course schedule

Lecture Date Content

1 7-Sep Course introduction
2 11-Sep
3 13-Sep
4 14-Sep
5 18-Sep
Mathematical programming
6 21-Sep
7 25-Sep
8 28-Sep
9 9-Oct
10 12-Oct Midterm
11 16-Oct
12 19-Oct Decision analysis
13 23-Oct
14 26-Oct
15 30-Oct Business simulation
16 2-Nov
17 6-Nov
18 9-Nov Course summary

19 18-Nov Final Exam-6:30-8:30 pm

23
DSME 6620
Decision Models and Applications
MGT578
Why Scientific?
Judgmental Bias

• Managers oftentimes develop rules of

thumb to reduce the information-
processing demands of decision making.

• These rules of thumb provide managers with

efficient ways of dealing with complex
problems that would require a significant
proportion of the time for producing good
decisions.

• However, these rules of thumb may lead

managers to systematically biased
outcomes. Such judgmental bias occurs in
situations in which the rules of thumb are
inappropriately applied by an individual in
reaching a decision.

24