You are on page 1of 42

PHUONG NGUYEN

BUSINESS ANALYTICS
SYSTEMS FOR DATA-DRIVEN DECISION-MAKING
CONTENTS
1. WHAT IS BUSINESS ANALYTICS?
THE NEW NAME FOR DECISION SUPPORT SYSTEMS?

2. TYPES OF BUSINESS ANALYTICS


DESCRIPTIVE, PREDICTIVE, OR PRESCRIPTIVE?

3. ANALYTICS MODELS
EXPLANATORY OR PREDICTIVE?

2
EVOLUTION OF
COMPUTERIZED DECISION SUPPORT

Source: Sharda et al. (2020)

3
• In general: Data Analysis
→ Data analytics ~ analytics → Data science
• In business: Decision support systems (DSS)
→ Business intelligence (BI)
→ Business analytics → Business data science
Source: Kelleher & Tierney (2018), Sharda et al. (2020)
Business Analytics (BA)

EIS
(Business
Intelligence)

Information systems inside the organization


Source: Based on Turban et al. (2008)
Business Analytics

Information technology inside the organization


Source: Rainer et al. (2014)
BUSINESS ANALYTICS
▪ Business Analytics
▪ The process of developing recommendations for
actions or actionable decisions based upon insights
generated from historical data.
▪ Used for fact-driven decision making, which is often
seen as more objective than other alternatives for
decision making.

Source: Sharda et al. (2020), Camm et al. (2019)

7
Example of Analytics Applications in a Retail Value Chain
Source: Abhishek Rathi, CEO, vCreaTek.com
Source: Asllani (2015)
BUSINESS ANALYTICS VS
BUSINESS ANALYSIS
▪ Analysis vs Analytics (Delen, 2015)
▪ Analysis refers to the process of separating a whole
problem into its parts so that the parts can be
critically examined at the granular level.
▪ Analytics is a variety of methods, technologies, and
associated tools for creating new knowledge/insight
to solve complex problems and make better and
faster decisions.
▪ Business Analytics vs Business Analysis?
▪ Focus of business analysis is functions and processes
–identifying business needs and recommending
solutions.
10
https://www.capella.edu/blogs/cublog/the-differences-between-business-
analytics-and-business-analysis/
BUSINESS ANALYTICS VS
DECISION-MAKING PROCESS
▪ Decision making: A process of choosing among two
or more alternative courses of action for the purpose of
attaining a goal(s).
▪ Humans consciously or subconsciously follow a
systematic decision-making process: (Simon, 1977)
Decision making

1. Intelligence Problem solving Descriptive analytics


2. Design Predictive analytics
3. Choice Prescriptive analytics
4. Implementation
Business analytics process

13
TWO BASIC KINDS OF DECISION MAKING

DANIEL KAHNEMAN
http://phuongacademy.com/thinking-fast-slow/
TYPES OF BUSINESS ANALYTICS
 Descriptive analytics
(reporting analytics, business intelligence)
▪ the use of data to understand past and current business
performance and make informed decisions
 Predictive analytics
(machine learning, data mining, data science)
▪ predict the future by examining historical data, detecting
Advanced analytics

patterns or relationships in these data, and then


extrapolating these relationships forward in time
 Prescriptive analytics
(decision analytics, decision science, normative analytics)
▪ identify the best alternatives to minimize or maximize
some objective
Proposed by INFORMS (the Institute of Operations 15
Research and Management Sciences)
TYPES OF BUSINESS ANALYTICS
Example: Retail Markdown Decisions
Most department stores clear seasonal inventory by
reducing prices.
The question is: When to reduce the price and by
how much?
 Descriptive analytics: examine historical data for
similar products (prices, units sold, advertising, …)
 Predictive analytics: predict sales based on price
 Prescriptive analytics: find the best sets of pricing
and advertising to maximize sales revenue

16
TYPES OF BUSINESS ANALYTICS

Source: Sharda et al. (2020) 17


TYPES OF BUSINESS ANALYTICS

Business Analytics Defined


https://www.facebook.com/watch/?v=889414154743044

18
TYPES OF BUSINESS ANALYTICS

Four Types of Analytics Capability


Source: Gartner (2015)

19
TYPES OF BUSINESS ANALYTICS

20
Source: SAS (2008)
Source: SAS (2008)
Cognitive
analytics

Diagnostic
analytics

Source: Davenport & Harris (2017)


TYPES OF BUSINESS ANALYTICS
▪ Descriptive analytics (aka business intelligence or
performance reporting) provides access to historical
and current data. It provides the ability to alert, explore,
and report using both internal and external data from a
variety of sources.
▪ Predictive analytics uses quantitative techniques (e.g.,
propensity, segmentation, network analysis and
econometric forecasting) and technologies (such as
models and rule-based systems) that use past data to
predict the future.

Source: Davenport & Harris (2017)

24
TYPES OF BUSINESS ANALYTICS
▪ Prescriptive analytics uses a variety of quantitative
techniques (such as optimization) and technologies
(e.g., models, machine learning and recommendation
engines) to specify optimal behaviors and actions.
▪ Autonomous analytics employs artificial intelligence
or cognitive technologies (such as machine learning) to
create and improve models and learn from data—all
without human hypotheses and with substantially less
involvement by human analysts.

Source: Davenport & Harris (2017)

25
Data Engineer/Architect BI/Data Analyst ML/AI Engineer

Source: Sharda et al. (2020)


ANALYTICS MODELS
 A model is a simplified representation or
abstraction of reality.
 It is usually simplified because reality is too
complex to describe exactly and because
much of the complexity is actually irrelevant in
solving a specific problem.
 Modeling involves conceptualizing a problem
and abstracting it to quantitative and/or
qualitative form.

27
ANALYTICS MODELS
 A (decision) model is a logical or mathematical
representation of a problem or business situation that
can be used to understand, analyze, or facilitate making
a decision.
 Building models is more of an art than a science.
Creating good models requires:
▪ solid understanding of business functional areas
▪ knowledge of business practice and research
▪ logical skills
 It is best to start simple and enrich models as necessary.

28
ANALYTICS MODELS

f(X 1, X 2,…, X p) = Y

Input variables Output variable


Independent variables Dependent variable
Predictors Predicted variable
Explanatory variables Response variable
Attributes Outcome variable
Features Target variable
… …

29
ANALYTICS MODELS f(X , X ,…, X ) 1 2 p
output variable
= Y
▪ Three types of input: input variables
▪ Data (or parameters), which are assumed to be constant for
purposes of the model.
▪ Uncontrollable variables, which are quantities that can change
but cannot be directly controlled by the decision maker.
▪ Decision variables (or controllable variables), which are
controllable and can be selected at the discretion of the
decision maker.

Source: Sharda et al. (2020)

30
ANALYTICS MODELS

Source: Sharda et al. (2020)

31
ANALYTICS MODELS

Categories of business analytics modeling techniques


Source: Ragsdale (2015)

32
ANALYTICS MODELS
▪ Strategies for Modeling:
▪ Logic-Driven Models: based on experience,
knowledge, and logical relationships of
variables and constants connected to the
desired business performance outcome
situation
▪ Data-Driven Models: use data collected
from many sources to quantitatively establish
model relationships
Source: Evans (2016)

33
ANALYTICS MODELS:
LOGIC-DRIVEN MODELS
Example: A Profit Model
▪ Develop a decision model for predicting profit in face
of uncertain demand.
P = profit
R = revenue
C = cost
p = unit price
c = unit cost
F = fixed cost
S = quantity sold
D = demand Influence Diagram
Q = quantity produced
34
ANALYTICS MODELS:
LOGIC-DRIVEN MODELS
Example: A Profit Model
▪ Cost = fixed cost + variable cost
C = F + c*Q
▪ Revenue = price * quantity sold
R = p*S
▪ Quantity sold =
Minimum{demand, quantity produced}
S = min{D, Q}
▪ Profit = Revenue − Cost
P = p*min{D, Q} − (F + cQ)

35
ANALYTICS MODELS:
DATA-DRIVEN MODELS
Example: A Sales-Promotion Model
▪ In the grocery industry, managers typically need to
know how best to use pricing, coupons and advertising
strategies to influence sales.
▪ Using business analytics, a grocer can develop a model
that predicts sales using price, coupons and
advertising.
Price

Coupons Sales

Advertising
Influence Diagram

36
ANALYTICS MODELS:
DATA-DRIVEN MODELS
Example: A Sales-Promotion Model

෣ = 500 – 0.05(price) + 30(coupons) +0.08(advertising) + 0.25(price)(advertising)


Sales

37
ANALYTICS MODELS

38
ANALYTICS MODELS

Timeline for building predictive models


Source: Abbott (2014)

39
ANALYTICS MODELS
▪ The two cultures for data-driven modeling:
▪ Statistical (explanatory) modeling: The analysis
in this culture starts with assuming a stochastic
model for the inside of the black box
→ The black box is filled (using linear regression)
▪ Algorithmic (predictive) modeling: The
analysis in this culture considers the inside of the
box complex and unknown → find a function f(X)
- an algorithm that operates on X to predict the
responses Y.
Source: Breiman (2001)

40
ANALYTICS MODELS
▪ Situation 1: The model function f is theoretically or
empirically known. Then the output quantity of interest
may be calculated directly.
▪ Situation 2: The structural form of the function f is known
but not the values of its parameters. Then these
parameter values may be statistically estimated on the
basis of empirical data by curve fitting methods.
▪ Situation 3: Even the structural form of the function f is
unknown. As an approximation the function f may be
modelled by a machine learning technique on the basis
of empirical data.
Source: Zielesny (2011)

41

You might also like