Download Module 5 foundations of analytics

MODULE 5
FOUNDATIONS OF
ANALYTICS
OBJECTIVES
To understand the fundamentals of business
analytics.
 To know the evolution of business analytics.
 To study the scope of business analytics.
 To evaluate the DATA for business analytics.
 To describe Decision Models.
 To understand fundamentals of data
warehousing.
 To prepare dashboard and reporting.

DEFINING BUSINESS ANALYTICS
Analytics is the use of data, information
technology, statistical analysis, quantitative
methods, and mathematical or computer –based
models to help managers gain improved insight
about their business operations and make better,
fact-based decisions.
BUSINESS ANALYTICS APPLICATIONS
Management of customer relationships
 Financial and marketing activities
 Supply chain management
 Human resource planning
 Pricing decisions
 Sport team game strategies

IMPORTANCE OF BUSINESS ANALYTICS
There is a strong relationship of BA with:
Profitability of business
 Revenue of business
 Shareholder return
BA enhances understanding of data
BA is vital for business to remain competitive
BA enables creation of informative reports

EVOLUTION OF BUSINESS ANALYTICS
Operations research
 Management science
 Business intelligence
 Decisions support systems
 Personal computer software

TYPES OF BUSINESS ANALYTICS

Descriptive analytics

Uses data to understand past and present.
Predictive analytics- Analyzes past performance

 Predictive
 Data
analysis techniques
mining
 Simulation

Prescriptive analytics- Uses optimization techniques
 Prescriptive analytics techniques
Simulation optimization
 Decision analysis

SHOPPERS STOP-RETAIL MARKDOWN
Shoppers stop clears 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 prices
 Prescriptive analytics: find the best sets of
pricing and advertising to maximize sales
revenue.

SCOPE OF BUSINESS ANALYTICS
Analytics in Practice:
Ginger Hotel from TATAs
Ginger has owns numerous hotels.
Uses analytics to:


Forecast demand for rooms
Segment customers to chose right destination
Uses prescriptive models to:
Set room rates
 Allocate rooms

TOOLS OF BUSINESS ANALYTICS
MS Excel
SAS
SPSS
Modeler
• Excel is an excellent reporting tool. We may use
different analytic software to do analytical work but
at the end we will use Excel for reporting and
presentation of results.
• This software has wide range of capabilities from
data management to advanced analytics.
• It’s a data mining software. This tool has an
intuitive GUI and its point-and-click modelling
capabilities are very comprehensive.
TOOLS OF BUSINESS ANALYTICS
Salford
Systems
• It provides a host of predictive analytics and data
mining tools for businesses. This software is easy
to use.
KXEN
• Its one of the few companies that are driving
automated analytics. This software can run huge
amount of data. But its difficult to understand and
explain the results.
MATLAB
• It’s a statistical computing software. It allows
matrix manipulations, plotting of functions and
data, implementation of algorithms and creations
of user interfaces.
TOOLS OF BUSINESS ANALYTICS
R:
WEKA
• R is a programming language and software
environment for statistical computing and
graphics. It is used hardly for any analysis.
• Waikato Environment for Knowledge Analysis
(WEKA), it’s a machine learning software. It’s a
open source software most popular among business
peoples.
CATEGORIES OF BUSINESS ANALYTICS
1.
Information and Knowledge Discovery
1.
2.
3.
4.
5.
6.
2.
Online Analytical Processing(OLAP)
Ad-hoc Queries and Reports
Data Mining
Text Mining
Web Mining
Search Engines
Decision Support and Intelligence Systems
1.
2.
3.
4.
5.
Decision Support System(DSS)
Group DSS Virtual Groups
Executive Support
Automated Decision Support
Web Analytics
CATEGORIES OF BUSINESS ANALYTICS
2.
Decision Support and Intelligence Systems
7.
8.
9.
3.
Management Science and Statistical Analysis
Applied Artificial Intelligence
Business Performance Management(BPM)
Visualization
1.
2.
3.
Visual Analysis
Dashboards and Scorecards
Virtual Reality
DATA FOR BUSINESS ANALYTICS

DATA


DATABASE


Collected facts and figures
Collection of computer files containing data
INFORMATION

Comes from analyzing data
DATA FOR BUSINESS ANALYTICS
EXAMPLES OF USING DATA IN BUSINESS:
 Annual reports
 Accounting audits
 Financial profitability analysis
 Economic trends
 Marketing research
 Operations management performance
 Human resource measurements
DATA FOR BUSINESS ANALYTICS
Metrics are used to quantify performance.
 Measures are numerical values of metrics.
 Discrete metrics involve counting

On time or not on time
 Number or proportion of on time deliveries


Continuous metrics are measured on a
continuum
Delivery time
 Package weight
 Purchase price

DATA FOR BUSINESS ANALYTICS

Excel sheet example
DATA FOR BUSINESS ANALYTICS
Four Types Data Based on Measurement Scale:
 Categorical (nominal) data
 Ordinal data
 Interval data
 Ratio data
DATA FOR BUSINESS ANALYTICS
Example
Classifying Data elements in Purchasing database
DATA FOR BUSINESS ANALYTICS
Classifying Data elements in Purchasing database
categorical
Ratio
Interval
DATA FOR BUSINESS ANALYTICS
Categorical (nominal) Data
 Data placed in categories according to a specified
characteristic
 Categories bear no quantitative relationship to
one another
 Examples:
Customer’s location (America, Europe, Asia)
 Employees classification (manager, supervisor,
associate)

DATA FOR BUSINESS ANALYTICS
Ordinal Data
 Data is ranked or ordered according to some
relationship with one another
 No fixed units of measurement
 Examples:
College football rankings
 Survey responses
(poor, average, good, very good, excellent)

DATA FOR BUSINESS ANALYTICS
Interval Data
 Ordinal data but with constant differences
between observations
 No true zero point
 Ratios are not meaningful
 Examples;
Temperature readings
 SAT scores

DATA FOR BUSINESS ANALYTICS
Ratio Data
 Continuous values and have a natural zero point
 Ratios are meaningful
 Examples:


Monthly sales
Delivery times
DECISION MODELS
Model:
 An abstraction or representation of a real system,
idea, or object
 Captures the most important features
 Can be a written or verbal description, a visual
display, a mathematical formula, or a
spreadsheet representation
DECISION MODELS
Examples Three Forms Of a Model- Samsung
Galaxy
The sales of a Samsung Galaxy. Often follow a
common pattern.
 Sales might grow at an increasing rate over time
as positive customer feedback spreads.
(See the S-shaped curve on the following slide.)
 A mathematical model of the S-curve can be
ct
identified; for example, S=a℮be , where S is
sales, t is time, e is the base of natural
logarithms, and a, b and c are constants.
DECISION MODELS
DECISION MODELS
A decision model is a model used to understand,
analyze, or facilitate decision making.
 Types of model input

Data
 Uncontrollable variables
 Decision variables (controllable)


Types of model output
Performance measures
 Behavioral measures

DECISION MODELS
Nature of Decision Models
Output
Input
Data,
Uncontrollable
Variables, and
Decision
Variables
Decision
Model
Measures of
Performance
or Behavior
DECISION MODELS
Example A Sales-Promotion Model of Big Bazaar
In the big bazaar, managers typically need to
know how best to use pricing, coupons and
advertising strategies to influence sales.
Using Business Analytics Big Bazaar can develop
a model that predicts sales using price, coupons
and advertising.
DECISION MODELS
Sales=5000.05(price)+30(coupons)+0.08(advertising)
DECISION MODELS
Descriptive Decision Models
•Simply tell “what is” and describe relationships
•Do not tell managers what to do
Example An Influence Diagram for Total Cost
Influence Diagrams
visually show how
Various model elements
relate to one another.
Fixed
cost
Total
cost
Variable
cost
DECISION MODELS
Example A Mathematical Model for Total Cost
TC = F+VQ
TC is Total Cost
F is Fixed Cost
V is Variable Unit Cost
Q is Quantity Produced
DECISION MODELS
Example A Break – Even Decision Model
TC(Manufacturing) = Rs50,000 + Rs125*Q
TC(Outsourcing) = Rs175*Q
Breakeven Point:
Set TC(Manufacturing)
= TC(Outsourcing)
Solve for Q = 1000 unit
DECISION MODELS
Examples A Linear Demand Prediction Model
As price increases, demand falls.
DECISION MODELS
Example A Nonlinear Demand Prediction Model
Assumes price elasticity (constant ratio of %
change in demand)
DECISION MODELS
Predictive Decision Models often incorporate
uncertainty to help managers analyze risk.
Aim to predict what will happen in the future.
Uncertainty is imperfect knowledge of what will
happen in the future.
Risk is associated with the consequences of what
actually happens.
DECISION MODELS
Prescriptive Decision Models help decision makers
indentify the best solution.
 Optimization – finding values of decision
variables that minimize (or maximize) something
such as cost (or profit).
 Objective function – the equation that minimizes
(or maximizes) the quantity of interest.
 Constraints – limitations or restrictions.
 Optimal solution – values of the decision
variables at the minimum (or maximum) point.
DECISION MODELS
Example A Pricing Model
 A firm wishes to determine the best pricing for
one of its products in order to maximize revenue.
 Analysts determined the following model:
Sales = -2.5698(price) + 5200.6
Total revenue = (price) (sales)
 Identify the price that maximizes total revenue,
subject to any constraints that might exist.
DECISION MODELS
Deterministic prescriptive models have inputs
that are known with certainty.
 Stochastic prescriptive models have one or more
inputs that are not known with certainty.
 Algorithms are systematic procedures used to
find optimal solutions to decision models.
 Search algorithms are used for complex problems
to find a good solution without guaranteeing an
optimal solution.

PROBLEM SOLVING AND DECISION MAKING
BA represents only a portion of the overall
problem solving and decision making process.
 SIX STEPS IN PROBLEM SOLVING PROCESS

1.
2.
3.
4.
5.
6.
Recognizing the problem
Defining the problem
Structuring the problem
Analyzing the problem
Interpreting results and making a decision
Implementing the solution
PROBLEM SOLVING AND DECISION MAKING
1.


Recognizing the problem
Problems exists when there is a gap between
what is happening and what we think should be
happening.
For example: Cost are too high compared with
competitors.
PROBLEM SOLVING AND DECISION MAKING
2.


Defining the problem
Clearly defining the problem is not a trivial
task.
Complexity increases when the following occur:





Large number of courses of action
Several competing objectives
External groups are affected
Problem owner and problem solver are not the same
person
Time constraints exist
PROBLEM SOLVING AND DECISION MAKING
3.



Structuring the Problem
Stating goals and objectives
Characterizing the possible decisions
Identifying any constraints or restrictions
PROBLEM SOLVING AND DECISION MAKING
4.


Analyzing the problem
Identifying and applying appropriate Business
Analytics techniques
Typically involves experimentation, statistical
analysis, or a solution process
Much of this course is devoted to learning BA
techniques for use in step 4.
PROBLEM SOLVING AND DECISION MAKING
5.




Interpreting Results and Making a Decision
Managers interpret the results from the
analysis phase.
Incorporate subjective judgment as needed.
Understand limitations and model assumptions.
Make a decision utilizing the above information.
PROBLEM SOLVING AND DECISION MAKING
6.


Implementing the Solution
Translate the results of the model back to the
real world.
Make the solution work in the organization by
providing adequate training and resources.
DATA WAREHOUSING

What is DATA WAREHOUSING?
It’s a subject oriented integrated non- volatile,
time varying collection of data in support of its
decision making process.
INTRODUCTION-CONT’D.

Where is it used?
It is used for evaluating future strategy.

It needs a successful technician:
Flexible.
 Team player.
 Good balance of business and technical understanding.

DATA WAREHOUSE
 Subject
oriented
 Data integrated
 Time variant
 Nonvolatile
CHARACTERISTICS OF DATA WAREHOUSE
 Subject
oriented. Data are organized based on
how the users refer to them.
 Integrated. All inconsistencies regarding
naming convention and value representations
are removed.
 Nonvolatile. Data are stored in read-only
format and do not change over time.
 Time variant. Data are not current but
normally time series.
DATA WAREHOUSING ARCHITECTURE
DATA WAREHOUSING ARCHITECTURE
It’s a structure that brings all the components of
a data warehouse together is known as
architecture.
 Architecture is a comprehensive blueprint.
 It defines the standards, measurements, general
design, and support techniques.

DATA WAREHOUSING ARCHITECTURE

1.
It includes
Warehouse Database Server:
The bottom tire is a warehouse database server.
It is a relational database system, Data from
operational databases and external sources(such as
customer profile information provided by
external consultants) are extracted using
application program interfaces known as
gateways.


2.
OLAP Server:
Middle tire one is an OLAP server.
Which is implemented using




A relational OLAP (ROLAP)
A multidimensional OLAP (MOLAP)
DATA WAREHOUSING ARCHITECTURE
3.
Client:


Top tire is a client.
Which contains query and reporting tools, analysis
tools and data mining tools( ex – Trend analysis,
prediction)
ADVANTAGES OF DW





More Cost Effective Decision Making
Better Enterprise Intelligence
Enhanced Customer Services
Business Reengineering
Information System Reengineering
DISADVANTAGES OF DW
Installation cost
 Time – Taking
 Change Resistance
 Specific Skills Required
 Complex
 Management Acceptance
 Security Issues

APPLICATIONS OF DW
Standard Reports and Queries
 Queries against Summarized Data
 Data Mining
 Interface with Other Data Warehouses

DASHBOARD
It’s an executive system UI that is designed to be
easy to read.
 It provides decision makers the input necessary
to “drive” the business.
 It displays tables, graphics, gauges (colour
differences)
 It’s a combined information holder which
provides multiple views to user i.e he can access
the information in any devices.

TYPES OF DASHBOARDS
1.
2.
3.
1.
Strategic Dashboard
Analytical Dashboard
Operational Dashboard
PRINCIPLES OF EFFECTIVE DASHBOARDS
It should provide timely summary information that
are important to the user.

2.
3.
Example- A CAR dashboard which provides all
information like speed, oil indicator, heat level, etc.
It should provide all information on one single
screen, with multiple windows in it.
The Key Performance Indicators(KPI) is displayed
in the data dashboard should convey meaning to its
end user and be related to the decisions the user
makes
PRINCIPLES OF EFFECTIVE
DASHBOARDS
4.
5.
A data dashboard should call attention to
unusual measures that may require attention,
but not in an overwhelming way.
Color should be used to call attention of specific
values.
BENEFITS OF DASHBOARD
Visual presentation of performance measures.
 Ability to identify and correct negative trends.
 Measures efficiencies/inefficiencies.
 Ability to generate detailed reports showing new
trends.
 Ability to make more informed decisions based on
collected data.
 Align strategies and organizational goals.
 Save time over running multiple reports.
 Gain total visibility of all systems instantly.

REPORTING
These are often used to display the results of an
experiment, investigation or inquiry.
 Reports provide thus some static snapshots in
time of the performance/status of the entity one
is examination.
TYPES OF REPORTS
1. Routine Reports


2.
Example- weekly sales figures, units produced.
Ad-hoc(or On Demand) Reports

Example- list of all customers who purchased a
company’s products more than Rs5000/- each during
October 2005.
MASTER DATA MANAGEMENT(MDM)
Definition:
It is a comprehensive method of enabling an
enterprise to link all of its critical data to one file,
called a master file, that provides a common
point of reference. When properly done, MDM
streamlines data sharing among personnel and
departments. In addition, MDM can facilitate
computing in multiple system architectures,
platforms and applications.
Categories of Data
Meta Data
Reference
Data
Master
Data
Transaction
Data
Historical
Data
ADVANTAGES OF MDM
Enhances efficiency
 Optimise outcome
 Spot and Act on Insights Faster
 Accelerate Time to Market
 Elevate Customer Satisfaction

DISADVANTAGES OF MDM
Lack of Functional sponsorship
 Failure to Adjust Business Processes Accordingly
 Lack Of Validation
 Taking an “ALL at Once” Approach to
Deployment
 Failure to Create And Enforce Data Governance
Procedures
