อภิชาต ชมภูนุช

Sunday, June 27, 2010

BUSINESS INTELLIGENCE &

BUSINESS PERFORMANCE MGT.
AGENDA
 Overview of BI and its components
 Data warehouse
 Data Analytical and Visualization
 Business Performance Management
OVERVIEW OF BI

Meaning
 Business Intelligence is a conceptual framework

for decision support. It combines architecture,

data base (or data warehouse), analytical tools,
and application.
Important
 Right information, right time, and right place
OVERVIEW OF BI

Significant Characteristic
 Analytical process
 Real-time information
 Competitive advantage

 Tangible and Intangible Cost/Benefit

consideration
DATA WAREHOUSE (DW)
 Definition and concept
 DW Process
 DW Architecture
 Extracted, Transformation, Load (ETL)
 DW Development
 DW Administration and Security
DEFINITION AND CONCEPT
 DW is a pool of data produced to support
decision making; it is a repository of current and
historical data of potential interest to managers
throughout the organization.
 Data Mart is a pool of data consisting of single
subject.
 Dependent : direct to DW
 Independent: serve strategic business unit
DEFINITION
 Data Warehouse:
 A subject-oriented, integrated, time-variant, non-updatable
collection of data used in support of management decision-
making processes
 Subject-oriented: e.g. customers, patients, students, products
 Integrated: Consistent naming conventions, formats,
encoding structures; from multiple data sources
 Time-variant: Can study trends and changes
 Nonupdatable: Read-only, periodically refreshed
 Data Mart:
 A data warehouse that is limited in scope

7
DEFINITION AND CONCEPT
 Operating Data Storage (ODS) is a type of data
warehouse often used for as an interim staging
area of data warehouse or short-term memory for
storing only recent information.
 oper mart: data comes from ODS
 Enterprise Data Warehouse is a large-scale data
that is used across organization for decision
support.
DEFINITION AND CONCEPT
 Metadata are data about data.
 Primary purpose of metadata is provides enrich
information that leads to the creation knowledge.
 METADATA (DATA CATALOG)

 Identify subjects of the data mart

 Identify dimensions and facts
 Indicate how data is derived from enterprise data
warehouses, including derivation rules
 Indicate how data is derived from operational data store,
including derivation rules
 Identify available reports and predefined queries
 Identify data analysis techniques (e.g. drill-down)
 Identify responsible people

10
DW PROCESS
 As shown in Figure 5.1 p.213
DW ARCHITECTURE
 Three-tier architecture

Data Acquisition Data warehouse Application

• Data/software • EDW • DSS
• Data mining
Figure 11-4 Dependent data mart with ODS provides option for
operational data store: a three-level architecture obtaining current data

T
E Simpler data access
Single ETL for
enterprise data warehouse Dependent data marts
(EDW) loaded from EDW
13
DW ARCHITECTURE
 Two-tier architecture

Data
.
Acquisition
Data warehouse
EDW
Data/software
Application
DSS
Data mining
 Figure 11-2: Generic two-level data warehousing architecture

L
One,
company-
wide
T warehouse

Periodic extraction  data is not completely current in warehouse

15
Figure 11-3 Independent data mart Data marts:
data warehousing architecture Mini-warehouses, limited in scope

T
E

Separate ETL for each Data access complexity

independent data mart due to multiple data marts

16
DW OF WEB-BASED ARCHITECTURE

.
Web
Web Web
browser
pages Application

Web Web DW
browser server
ETL PROCESS
 The purpose of ETL process is to load the
warehouse with integrated and cleaned data.
THE ETL PROCESS

 Capture/Extract
 Scrub or data cleansing
 Transform
 Load and Index
ETL = Extract, transform, and load

19
 Capture/Extract…obtaining a snapshot of a chosen subset of the
source data for loading into the data warehouse
Figure 11-10:
Steps in data
reconciliation

Static extract = capturing a Incremental extract =

snapshot of the source data at capturing changes that have
a point in time occurred since the last static
extract
20
 Scrub/Cleanse…uses pattern recognition and AI techniques to
upgrade data quality
Figure 11-10:
Steps in data
reconciliation
(cont.)

Fixing errors: misspellings, Also: decoding, reformatting, time

erroneous dates, incorrect field usage, stamping, conversion, key
mismatched addresses, missing data, generation, merging, error
duplicate data, inconsistencies detection/logging, locating missing
data
21
 Transform = convert data from format of operational system
to format of data warehouse
Figure 11-10:
Steps in data
reconciliation
(cont.)

Record-level: Field-level:
Selection–data partitioning single-field–from one field to one field
Joining–data combining multi-field–from many fields to one, or one
Aggregation–data summarization field to many

22
 Load/Index= place transformed data into
the warehouse and create indexes
Figure 11-10:
Steps in data
reconciliation
(cont.)

Refresh mode: bulk rewriting of Update mode: only changes in

target data at periodic intervals source data are written to data
warehouse

23
 ETL
 After ETL, data should be:
 Detailed–not summarized yet
 Historical–periodic
 Normalized–3rd normal form or higher
 Comprehensive–enterprise-wide perspective
 Timely–data should be current enough to assist decision-making
 Quality controlled–accurate with full integrity

24
DW DEVELOPMENT APPROACH
 EDW approach
 Emphasize top-down, exclude data marts
 Data Mart approach
 Emphasize bottom-up, “plan big, build small”
DW IMPLEMENT ISSUES
 Best practices
 Fit with corporate strategy and objective
 Management support; executive, managers, users
 Commitment and user expectation
 Built incrementally
 Adaptability
 Incorporate of both IT and professional
 Quality data
 Political awareness
DW IMPLEMENT ISSUES
 Don’t
 Starting with wrong sponsorship
 Setting expectation that does meet users needed
 Engaging in political naïve behavior
 Loading information because of it available
 Focus on technology oriented rather than user
oriented
 Delivery data with overlap and confusing definitions
DW ADMINISTRATION AND SECURITY
 Establish effective corporate and security policies
and procedures
 Implement logical security procedure and
technique to restrict access
 Limit physical access to data center environment
 Establish internal control focusing on security
and privacy
BUSINESS ANALYTICAL AND VISUALIZATION

 Definition and concept

 Online analytical Process (OLAP)
 Reports and Query
 Multidimensionality
 Data visualization
 Usage, benefit, and success of BA
BA DEFINITION AND CONCEPT
 BA is abroad category of applications and
techniques for gathering, storing, analyzing, and
providing access to data to help user make better
business and strategic decisions.
 Report
 Query
 Alert
OLAP
 OLAP refers to a variety of activities usually
performed by end users in online process.
 OLAP provides modeling, analysis, and
visualization capabilities to large data sets.
OLAP: TOOLS
 The use of a set of graphical tools that provides users
with multidimensional views of their data and allows
them to analyze the data using simple windowing
techniques
 Relational OLAP (ROLAP)
 Traditional relational representation
 Multidimensional OLAP (MOLAP)
 Cube structure
 OLAP Operations
 Cube slicing–come up with 2-D view of data
 Drill-down–going from summary to more detailed views

32
OLAP CHARACTERISTICS
 Categorical analysis: allow to predict future
based on historical data
 Drill-down analysis: allows to determine detail
data
 Contemplative analysis: allows user to change a
single variable to see its impact
 Formulaic analysis: allows to change multiple
variables
REPORT AND QUERY

Reports
 Routine report: generated and distributes

necessary information periodically to subscribers.

 Ad hoc or on-demand report: created specific

report for user needed.

 Multilingual support: offer report translation to

several languages
REPORT AND QUERY

Query
 Ad hoc query: allows user to request information

that does not in the report.

 Figure 11-13 Components of a star schema
Fact tables contain factual or
quantitative data

1:N relationship between dimension Dimension tables are denormalized to

tables and fact tables maximize performance

Dimension tables contain descriptions about

the subjects of the business

Excellent for ad-hoc queries, but bad for online transaction processing

36
MULTIDIMENSIONALITY
 Data cube represents some attributes of database.
It can be two-dimensional, three-dimensional, or
higher-dimensional.
Figure 11-23 Slicing a data cube

38
 Summary report
Figure 11-24
Example of drill-down

Starting with summary

data, users can obtain Drill-down with
details for particular cells color added

39
DATA VISUALIZATION
 Spreadsheet
 Dashboard and scorecard
 Financial data visualization
 DATA MINING AND VISUALIZATION
 Knowledge discovery using a blend of statistical, AI, and computer graphics
techniques
 Goals:
 Explain observed events or conditions
 Confirm hypotheses
 Explore data for new or unexpected relationships
 Techniques
 Statistical regression
 Decision tree induction
 Clustering and signal processing
 Affinity
 Sequence association
 Case-based reasoning
 Rule discovery
 Neural nets
 Fractals
 Data visualization–representing data in graphical/multimedia formats for analysis

41
USAGE, BENEFIT, AND SUCCESS BA
 Organization need analytical tools
 System is useless if it does not provide useful
information
 Technology adoption involve culture, people,
procedures, and so on.
BUSINESS PERFORMANCE MANAGEMENT: BPM

 Overview
 Strategies-where do we want to go?
 Plan-how do we go there?
 Monitoring-how are we doing?
 Act and Adjust-what do we need to do
differently?
 Measurements
OVERVIEW
 BPM is a framework for organizing, analyzing
business methodology, process, and system to
drive the overall enterprise performance.
STRATEGY
 Long-term planning
 Directional
 Bible of all participants
PLAN
 Operational planning
 Financial planning and budgeting
MONITORING
 Diagnostic the system
 Variation of operation and performance from
planning
ACT & ADJUST
 Keep the good thing done
 Get rid of the bad thing
 Improve for the better
 Keep on development
MEASUREMENT
 Balanced scorecard
 Six sigma
 DAMIC
 Define
 Analyze
 Measure
 Improve
 Control
CONCLUSION
 Important of BI: strategy, process, performance
 The most component of BI is data warehouse.
 BA and visualization allows users to access,
utilize, and exploit data.
 BI concern both tangible and intangible cost and
benefit which is measured by business
performance.