Introduction to Business Analytics

ADM2302: Business Analytics

Presented by
Danial Khorasanian
University of Ottawa, Telfer School of
Management

Spring/Summer 2021
 2

Business Analytics (or Analytics)

• Analytics includes all quantitative decision
sciences, statistics, date science, etc.
• The main goal is to provide useful
information and/or decisions using
quantitative methods.

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 3

Analytics categories
• Descriptive analytics: To analyze the available data
to describe what has happened.
• Predictive analytics: To predict the future by
examining historical data. It includes “Machine
Learning” techniques.
• Prescriptive analytics: To identify the best decision
to minimize or maximize some objectives in
optimization models. It includes “Operations
Research” methods.

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 4

Analytics categories …

• Prescriptive analytics suggesting actions to benefit

from the predictions and showing the implications
of each decision option.
• “Operations Research”, “Management Science”,
and “Prescriptive Analytics” are usually used
interchangeably in the literature.

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 5

History of Operations Research

• The beginning of the activity called operations research has generally
been attributed to the military services early in World War II.
• Because of the war effort, there was an urgent need to allocate
scarce resources to the various military operations and to the
activities within each operation in an effective manner.
• Therefore, the British and then the U.S. military management called
upon a large number of scientists to apply a scientific approach to
dealing with this and other strategic and tactical problems. In effect,
they were asked to do research on (military) operations.
• Operations Research is commonly referred to as OR.

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 6

Applications of OR
OR has been applied extensively in such diverse areas as:
• manufacturing,
• transportation,
• construction,
• telecommunications,
• financial planning,
• health care,
• military, and
• public services,
to name just a few.

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 7

Applications of OR …

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 8

Main steps of an OR study

1. Define the problem and collect data
2. Mathematically formulate the problem
3. Develop a (usually computer-based) method to
solve the mathematical model
4. Test the model and refine it as needed
5. Prepare for the ongoing application of the model
as prescribed by management.
6. Implement

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 9

An example of an OR study: Watch project

• A company plans to start producing a new

expensive watch.
• They want to study about profitability of this
project.

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 10

Step 1: Define the problem and collect data

• Most practical problems encountered by OR teams

are initially described to them in a vague, imprecise
way.
• Therefore, the first order of business is to
study the relevant system and develop a well-
defined statement of the problem to be
considered.

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 11

Step 1: Define the problem and collect data …

This step includes determining

• Input parameters: Constants in the mathematical model
• Decision variables: Quantifiable decisions to be made
• Objective function: It is a function of decision variables
• Constraints: Each constraint is an inequality or equation that
expresses some restrictions on the values that can be assigned
to the decision variables.
• interrelationships between the area to be studied and other
areas of the organization,
• time limits for making a decision, and so on.

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 12

Step 1: Define the problem and collect data …

• OR teams typically spend a surprisingly large amount of time gathering

relevant data about the problem.
• Much data usually are needed both to gain an accurate understanding
of the problem and to provide the needed input for the mathematical
model being formulated in the next phase of study.
• Frequently, much of the needed data will not be available when the
study begins, either because the information never has been kept or
because what was kept is outdated or in the wrong form.
• Therefore, it often is necessary to install a new computer-based
management information system to collect the necessary data on an
ongoing basis and in the needed form.

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 13

Watch project: Step 1

• Decision: To produce this watch or not and if yes,

how many watches they should produce?
• Objective: Maximize profit
• Constraints: number of watches produced is (i)
nonnegative, (ii) integer, and (iii) less than or equal
to an estimated demand (sales forecast)

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 14

Watch project: Step 1 …

Input parameters:

Unit revenue 2,000

Fixed cost 10,000,000
Variable cost 1,000
Sales forecast 30,000

• A cost that remains the same regardless of the production

volume (provided that it is nonzero) is referred to as a fixed
cost, whereas a cost that varies with the production volume is
called a variable cost.

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 15

Step 2: Mathematically formulate the problem

After the problem is defined, the next phase is to

mathematically formulate it in a form that is convenient
for analysis.
The key elements of a mathematical model are:
• Input parameters
• Decision variables
• Objective function
• Constraints

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 16

Step 2: Mathematically formulate the problem …

The mathematical model might then say that the

problem is
• to choose the values of the decision variables
• so as to maximize (or minimize) the objective
function,
• subject to the specified constraints and
• for a set of known input parameters.

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 17

Watch project: Step 2

• Decision variable: There is one decision variable

showed by 𝑄 that is the number watches
produced.

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 18

Watch project: Step 2 …

• Input parameters:
Unit revenue (𝑟 ) 2,000
Fixed cost ( 𝑐 𝑓 ) 10,000,000
Variable cost ( 𝑐 𝑣 ) 1,000
Sales forecast (s) 30,000

• Objective function: To maximize the profit

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 19

Watch project: Step 2 …

• Constraints:
1. Number of watches produced cannot be less
than 0, and should be integer: 𝑄 ≥ 0,
𝑖 𝑛 𝑡 𝑒 𝑔 𝑒 𝑟
2. Number of watches produced should not be
more than sales forecast 𝑠 : 𝑄 ≤ 𝑠

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 20

Watch project: Step 2 …

• Therefore, the mathematical model is:

, integer

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 21

Step 3: Solve the mathematical model

• After deriving the mathematical model in an OR

study, the next step is to solve this model to get a
good solution for the decision variables.
• There are many different ways to solve a
mathematical model that we will discuss about
some of them in this course.

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 22

Step 3: Solve the mathematical model …

• Some methods called “exact methods” guarantee

the best solution (“optimal solution”).
• Some other methods called “heuristics” only try
to find a good solution (suboptimal solution) due
to time limitation or complexity of the model.

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 23

Step 3: Solve the mathematical model …

• In addition to obtain the (sub)optimal solution for the model, we need to do
a what-if analysis in this step as well.
• There are too many uncertainties associated with real problems,
and assumptions made for the model can be imprecise.
• Therefore, even if we obtain the optimal solution for a model, there is no
guarantee that it will be the best possible solution that could have been
implemented for the real problem.
• What-if analysis is used to check the effect of changing model assumptions
on the (sub)optimal solution.

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 24

Watch project: Step 3

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 25

Watch project: Step 3 …

• The main assumption for this model are about the values of its
input parameters, i.e. fixed cost, etc.
• A what-if analysis for this model is to investigate what happens to
the recommendations of the model if any of the input parameters
turn out to be considerably different from the current ones.

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 26

Useful software for business analytics

• Optimizing small-scale problems: Excel
• Optimizing large-scale problems: Cplex,
Gurobi, etc.
• Date science/statistics: R, Python, SAS
• Visualization: Excel, Tableau
• Communicate and interact with databases:
SQL

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 27

Some other
OR problems

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 28

Example 1: Assembly line balancing

Assembly line balancing is
the problem of assigning
tasks to stations,
considering line and tasks
restrictions, for optimizing
one or more performance
criteria, like:
• Minimize number of
stations
• Minimize cycle time

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 29

Example 2: Operating room scheduling

• The recovery room
(PACU) is an immediate
unit after surgery
room, in which patients
are recovered after
their surgeries.
• When the capacity of
recovery room is low,
how to schedule the
patients in the
operating rooms to
minimize the number
of cancellations.

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 30

Example 3: Home care routing and scheduling

• The home care problem
is about serving
patients at their home
by some nurses.
• There are multiple
nurses and patients.
• How to assign patients
to nurses to minimize
the rejection, travel,
and overtime costs?

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
 31

Reference
• Hillier, Business Analytics, 1st CUSTOM
edition: Chapter 1

Telfer.uOttawa.ca
Business Analytics, spring/summer 2021
Thanks for Your
Attention