1
Decision Models & Analytics
1. Introduction
Professor Jiawei Zhang
Agenda 2
What is business analytics?
Three categories of analytics
Syllabus
What is Business Analytics 3
Business Analytics: A scientific process of transforming data into insights
for making better decisions (INFORMS)
-The Institute for Operations Research and the Management Sciences
is the largest professional society in the world in the field of
operations research, management science, and analytics (more than
12K members).
INFORMS recognizes that analytics is seen as both
-a complete business problem solving and decision making process,
and
-a broad set of analytical methodologies that enable the creation of
business value.
Three Categories of Analytics 4
Descriptive analytics
- Prepares and analyzes data
- Identifies patterns from samples for reporting of trends
- Data visualization, clustering of data
Predictive analytics
- Predicts future probabilities and trends
- Finds relationships in data that may not be readily apparent with
descriptive analysis
Prescriptive analytics
- Evaluates and determines new ways to operate
- Makes optimal decisions to achieve business objectives
- Balances all constraints
Analytics Methods and Models 5
Descriptive Analytics
– Basic Statistics
– Data Visualization
Predictive Analytics
– Regression Methods
– Time Series Analysis and Forecasting
– Traditional machine learning
– Deep learning
Prescriptive Analytics
– Deterministic Optimization (first module of this course)
– Simulation (second module of this course)
– Decision Analysis (DMU course)
– Stochastic Optimization (DMU course)
– Reinforcement learning (active area of research)
Models & Tools to be Covered 6
Deterministic Models and Optimization:
Linear Programming (LP). Basic features, formulation of optimization
models, Spreadsheet modeling: using Excel Solver. Applications.
LP-based Sensitivity Analysis. The Solver sensitivity report. Economic
interpretation of the shadow prices and reduced costs.
Network Optimization. The transportation problem: applications.
Minimum-cost flow problems: applications.
Mixed Integer Programming. Integer and binary variables, logical
relationships as linear constraints. Solving MIPs with Excel Solver.
Applications.
Non-linear Programming (NLP). Nonlinearities in the objective function
and constraints. Excel Solver: The GRG algorithm parameterization.
Evolutionary Solver. Dealing with problems with little structure.
Linearization of nonlinear models.
Probabilistic Models and Stochastic Optimization
Simulation. Basic concepts. Review of probability distributions. Crystal
Ball software.
Stochastic Optimization. Using the Optquest. Applications.
Finance Applications 7
Capital Budgeting
Portfolio Management
Financial Planning
Pension Fund Management
Currency Trading
Cash Balance
Option Pricing
Retirement Planning
Marketing Applications 8
Demand Estimation
Price Bundling
Advertising Planning
Customer Valuation
Online Advertising
Search Engine “Keywords” Bidding
Operations Applications 9
Product Mix
Production Planning and Scheduling
Inventory Management
Logistics and Transportation
Airline Flight Scheduling
Staff Scheduling
Supply Chain Design
Hotel Overbooking
Course Objectives
Deeper understanding of prescriptive analytics
– What is it
– Why it is important to business
Basic principle and techniques of business analytics
– Translate business problems into formal decision models
– Understand the strength and limits of different decision models
– Understand their assumptions and learn to challenge them
– Implement decision models in Excel spreadsheet
This is NOT a math course, NOT a data science course!
This course is about decision modeling through Excel, NOT about
becoming an Excel expert
Hands-On Class 11
This class involves very little theory, but instead involves lots of
practice on modeling and solving business problems.
– Lectures
– Mini cases: model building
– In-class exercises
– Homework assignments
Laptop
– Install the required applications (Excel add-in).
– Download template files in advance.
Course Materials 12
Textbooks (free online access for NYU students)
– Many examples
– Solved problems
Slides (available on NYU Brightspace)
Spreadsheet templates (available on NYU Brightspace)
Grading 13
Homework Assignments 30%
Five individual assignments
Exam: open-notes
Midterm 1 25%
Midterm 2 22%
Quiz: 8%
Class participation 15%
Attendance (mandatory): 5 points
In-class discussions
In-class exercises
In-class model submission (mandatory): 5 points
Homework
1. Reading: Chapter 1 of Optimization Modeling with Spreadsheets book
2. Load the Solver Add-in in Excel. Use the link below or google “Load
Solver Add-in”.
https://support.office.com/en-us/article/load-the-solver-add-in-in-excel-612926fc-
d53b-46b4-872c-e24772f078ca
3. Download the zipped folders of template files from NYU Brightspace.
