MS5216

Decision Analytics
 About Me
• Bachelor of Science from Tsinghua University
• Mathematics

• Master of Science from UPMC

• Mathematics

• Ph.D. from UIUC

• Operations Research

• Postdoc from MIT

 Course Information
• Lecture hours: Thursday, 19:00-21:50
• Office hours: by appointment
• We use Canvas for course materials and course
announcements. Please enable your email notification on
Canvas to avoid missing any information.
• Email ([email protected]) is the best way to contact
me. Please start with “MS5216: ” in the email subject.
• Reference
• Cliff T. Ragsdale (2022) Spreadsheet Modeling
and Decision Analysis: A Practical Introduction to
Business Analytics. 9th edition
 Course Mode
• According to the latest university guidelines, our course is
delivered in Face-to-Face mode.

• However, I would plan to use Zoom to record the lecture

and post the video on Canvas.

• The class is 3 hours per week. We will have a 10 mins break

for each 50 mins lecture
 Assessment
• Three group assignments • 30%

• One individual assignment • 10%

• Final exam • 60%

• Please form your team of 7-8 persons and sign up via

https://docs.google.com/spreadsheets/d/1nbw9VCTfEznXg7r4XfTds4nV
1NA4EgqVc-QCsrxDRBM/edit?usp=sharing
• Your group name must be chosen from {A, B, C, D, E, F, G, 1, 2, 3, 4, 5,6}
on a first come, first served basis.
 Policy
• There is an automatic deduction of 10% for late
assignments. Late assignments can be submitted for up to
24 hours after the due date.

• If you have questions regarding the grading of exams or the

assignments, you should contact me as soon as possible. I
will only accept regrade requests within ONE WEEK of the
grades being posted.

• Academic dishonesty is regarded as a serious offense in the

University.
• For example, plagiarism, collusion, use of unauthorized material in
assessment, etc.
• More can be found on https://www.cityu.edu.hk/ah/
 (Tentative) Course Roadmap
• Basic spreadsheet skills
• Formatting, charting, referencing, what-if analysis, …
• Building spreadsheet model & use of solver
• Linear programming
• Integer programming
• Nonlinear programming
• Simulation
• Monte Carlo, random number generation, all kinds of applications
• Project management
• Logic in spreadsheets
• Organizing and retrieving data
Introduction
 Managerial Decision Making

Analysis
Model Results

Symbolic

Interpretation
Abstraction

World Managerial
Judgment
Real
World

Management Intuition
Decisions
Situation
 A Scientific Approach

Define the Problem

Identify the Alternatives

Determine the Criteria Modeling

Problem Analysis
Solving Evaluate the Alternatives

Recommend an Alternative

Implement the Decision Decision

Evaluate the Results

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

Source: T. H. Davenport, “Competing on Analytics,” HBR, Jan. 2006

 McDonald’s Diet problem
• You want your diet to meet some nutritious standards.
According to your daily diet plan, you need to have:
• at least 100 percent of the U.S. Recommended Dietary Allowance1
(RDA) of vitamin C & calcium,
• at least 55 grams of protein,
• and at most 2000 calories.
• You are wondering if this can be accomplished by eating at
McDonald’s. Can you design the least-cost McDonald’s daily
meal plan that meets your daily nutritious standards?
Price
Menu Item (USD) Calories Protein Fat Sodium Vit. A Vit. C Calcium Iron
(g) (mg) % U.S. RDA
Hamburger 0.69 260 13 9 530 2 2 15 15
Big Mac 1.99 560 25 30 1010 8 2 25 25
Filet-O-Fish 1.89 400 14 18 640 2 0 15 10
French Fries (S) 1.39 250 2 13 140 0 6 2 4
Chicken McNuggets (6pcs) 1.99 250 15 15 670 2 2 2 4
Garden Salad 2.05 35 2 0 20 120 40 4 6
Honey 0 45 0 0 0 0 0 0 0
Baked Apple Pie 0.79 260 3 13 200 0 40 2 6

1National Institutes of Health (https://ods.od.nih.gov/Health_Information/Dietary_Reference_Intakes.aspx)

 Model for the McDonald’s Diet Problem
• MIN Money Spent at McDonald’s
• such that:
• Vitamin C obtained ≥ 100 % of U.S. RDA
• Calcium obtained ≥ 100 % of U.S. RDA
• Protein obtained ≥ 55𝑔
• Calories obtained ≤ 2000

• See “McDonalds” workbook

13
 Mathematical Models
• Model
• Abstraction of a real thing or process; some elements must be omitted
• Contains decision variables and represents a system in mathematical terms
• Finds values of decision variables which will improve system performance
• Used as a guide for aiding decision making
• Uses of Models
• To improve the existing decision
• To increase the understanding of a system
• To explicitly make trade-offs for a decision
 Elements of Models
• Decisions: choices, possible actions, controllable variables;
decision variables
• Data: information, environmental conditions,
uncontrollable variables; parameters
• Structure: relations, causes and effects, logics; equations or
inequalities
• Outcomes: performance measures, criteria, eventual
consequences; objective function

15
 Course Specifics
• Focus:
• Introduce the basic principles and techniques of management
science
• Sharpen your ability to structure problems and to perform logical
analysis
• Expose you to settings in which models are helpful
• Reinforce your computer skills

• Approach
• Start with small problems ⇒ Introduce methodology ⇒ Practise on
computers

16
 Why Spreadsheet Modeling ?
• Examples of spreadsheets:
• Microsoft Excel – still the industry standard…
• Numbers
• Google Spreadsheets
• Open office spreadsheet
• Advantages:
• Ubiquity
• Availability of Add-ins
• Low cost alternative to more specialized tools (statistics,
optimization)
• Disadvantages:
• Data size limitations, slow calculation speeds (large models)
• Difficult to document and organize models
• Often insufficient for advanced analysis
 Common Use of Spreadsheets
• The spreadsheet as a calculator:
• Easy access to the standard mathematical functions, probability
distributions
• Easy “what-if” calculations
• The spreadsheet as a mathematical tool:
• Simulation
• Optimization
• The spreadsheet as a statistical tool:
• Regression
• Graphical analysis
• The spreadsheet as a database:
• Pivot tables
 Features in Excel Spreadsheet
• Charting: See “Charting” workbook
• Data source ⇒ Insert tab ⇒ Charts group
• Chart tools ⇒ Design & Format & Chart buttons
• Different types of chart
• Frequency table
• Relative reference
• Fill, Fill handle, Format painter

19
 Next Lecture
• Before next lecture
• Read “Advertising Budget” problem

• Next Lecture: Basics of spreadsheet modeling

• Management Situation ⇒ Algebraic Model ⇒ Spreadsheet Model
• “What-if” analysis - Data Table, Scenario Analysis
• Break-even analysis - Goal Seek

20