Decision Support Systems Minor information & supply chain management
Table of Contents
Lecture 1 Introduction & Data Warehousing 26-08-2019 .................................................................. 2
Lecture 2 Data warehousing, Reporting & OLAP databases 29-08-2019 ............................................ 4
Lecture 3 OLAP databases & dashboards, simple forecasting 05-09-2019 ......................................... 7
Lecture 4 Knowledge-based systems (KBS) (1) 23-09-2019 ............................................................. 10
Lecture 5 Knowledge-based systems (2) 07-10-2019....................................................................... 13
Lecture 6 PA using Python 1: Introduction 04-11-2019 ................................................................... 16
Lecture 7 PA using Python 2: Data preparation 11-11-2019............................................................. 20
Lecture 7 PA using Python 3: Data Analytics 18-11-2019 ................................................................. 22
1
, Lecture 1 Introduction & Data Warehousing 26-08-2019
Content: organizational issues, introduction to DSS, introduction to data warehousing
Data Information Knowledge
Raw symbols Items that are the most Formatted data Data relationships
elementary descriptions Organized data that Processed data or information
Either internal or external (for organization) has meaning and value that is applicable to a business
Structured or unstructured (data has a data decision problem
structuring model, yes/no)
Unstructured data accounts for >80% of all
data in organizations.
The difference between these three is also a bit in the eye of the beholder.
When unstructured data is turned into structured data;
it’s called data preparation.
Business analytics is used to tackle organizational issues;
Many decisions in professional and private life are taken based on data that come from all sorts of
information systems. DSS, Business Intelligence, and Business Analytics, are about the way we can
use data stored in those information systems to generate new and useful information, that can
support executive managers in taking business decisions.
Types of business analytics (BA) methods:
- Descriptive analytics: use data to understand past & present
What happened? What is happening?
Data warehousing framework
Business reporting, OLAP, performance dashboard
- Predictive analytics: predict future behavior based on past performance
What will happen? Why will it happen?
Data mining process
Fundamentals of data mining models
- Prescriptive analytics: make decisions or recommendations to achieve the best performance
What should I do? Why should I do it?
Introduction to AI Italic: will learn about these topics
Knowledge management during this course
Knowledge-based systems
You can also divide them into groups based on function:
- Relevant for marketing, accounting, etc.
Business intelligence (BI): data warehousing + descriptive analytics
Business analytics (BA): predictive + prescriptive analytics
Business intelligence & business analytics is regarded as the same in this course.
Decision support systems is an umbrella term to describe any computerized system that supports
decision-making in an organization. → data driven decision making.
It’s about transforming data into meaningful information/knowledge to support business decision-
making.
Two clear definitions for business intelligence:
- Process based definition: BI is an umbrella term that combines the processes technologies,
and tools needed to transform data into information, information into knowledge, and
knowledge into plans that drive profitable business action.
- Product based definition: BI is information and knowledge that enables business decision-
making.
2
, BI tools & techniques:
- Data warehousing
- Knowledge management
- Query & Reporting
- OnLine Analytical Processing
- Digital dashboards
- Data, Process, and Text mining
- Visualization
-…
Business intelligence/analytics architecture:
ERP: contains operational/transactional systems:
- Different platforms, different databases, is process/product oriented, contains
internal/external data, structured and unstructured/data, no integrated data, inconsistent
data/low quality, limited historic data, transforming processing performance is key, hampered
by BI activities.
ELT staging area: temporary storage of data
- Extraction: extract data from operational/other systems
- Transformation: transform data (unification, cleaning, conversion)
- Loading: data loaded into data warehouses
Data warehouse: subject oriented integrated data that is consistent (high quality, historic, query
performance is key). Has star/snowflake model (multidimensional model). Is pre-aggregated and has
no current data.
Data marts: subset of data warehouse (less dimensions, history and detail), user group/application
base oriented. Multi-dimensional cubes. Can be both dependent or independent. Dependent data
mart updates when data warehouse updates. Independent data mart has its own ETL connection.
Meta data/master data: very important, is data about data (location, meaning, origin, etc.) contains
data models, definitions etc. Often from different sources (high maintenance) solution is one shared
repository or interchange using standards between sources (XML).
Multidimensional structures: cubes (Microsoft power BI uses these)
OLAP cubes: idea of slicing and dicing to get your report
Data warehouse working definition: “a database that is maintained separately from the
organization’s operational databases for the sole purpose of managerial decision-making”
Data warehousing: the process of constructing & using data warehouses.
3
Table of Contents
Lecture 1 Introduction & Data Warehousing 26-08-2019 .................................................................. 2
Lecture 2 Data warehousing, Reporting & OLAP databases 29-08-2019 ............................................ 4
Lecture 3 OLAP databases & dashboards, simple forecasting 05-09-2019 ......................................... 7
Lecture 4 Knowledge-based systems (KBS) (1) 23-09-2019 ............................................................. 10
Lecture 5 Knowledge-based systems (2) 07-10-2019....................................................................... 13
Lecture 6 PA using Python 1: Introduction 04-11-2019 ................................................................... 16
Lecture 7 PA using Python 2: Data preparation 11-11-2019............................................................. 20
Lecture 7 PA using Python 3: Data Analytics 18-11-2019 ................................................................. 22
1
, Lecture 1 Introduction & Data Warehousing 26-08-2019
Content: organizational issues, introduction to DSS, introduction to data warehousing
Data Information Knowledge
Raw symbols Items that are the most Formatted data Data relationships
elementary descriptions Organized data that Processed data or information
Either internal or external (for organization) has meaning and value that is applicable to a business
Structured or unstructured (data has a data decision problem
structuring model, yes/no)
Unstructured data accounts for >80% of all
data in organizations.
The difference between these three is also a bit in the eye of the beholder.
When unstructured data is turned into structured data;
it’s called data preparation.
Business analytics is used to tackle organizational issues;
Many decisions in professional and private life are taken based on data that come from all sorts of
information systems. DSS, Business Intelligence, and Business Analytics, are about the way we can
use data stored in those information systems to generate new and useful information, that can
support executive managers in taking business decisions.
Types of business analytics (BA) methods:
- Descriptive analytics: use data to understand past & present
What happened? What is happening?
Data warehousing framework
Business reporting, OLAP, performance dashboard
- Predictive analytics: predict future behavior based on past performance
What will happen? Why will it happen?
Data mining process
Fundamentals of data mining models
- Prescriptive analytics: make decisions or recommendations to achieve the best performance
What should I do? Why should I do it?
Introduction to AI Italic: will learn about these topics
Knowledge management during this course
Knowledge-based systems
You can also divide them into groups based on function:
- Relevant for marketing, accounting, etc.
Business intelligence (BI): data warehousing + descriptive analytics
Business analytics (BA): predictive + prescriptive analytics
Business intelligence & business analytics is regarded as the same in this course.
Decision support systems is an umbrella term to describe any computerized system that supports
decision-making in an organization. → data driven decision making.
It’s about transforming data into meaningful information/knowledge to support business decision-
making.
Two clear definitions for business intelligence:
- Process based definition: BI is an umbrella term that combines the processes technologies,
and tools needed to transform data into information, information into knowledge, and
knowledge into plans that drive profitable business action.
- Product based definition: BI is information and knowledge that enables business decision-
making.
2
, BI tools & techniques:
- Data warehousing
- Knowledge management
- Query & Reporting
- OnLine Analytical Processing
- Digital dashboards
- Data, Process, and Text mining
- Visualization
-…
Business intelligence/analytics architecture:
ERP: contains operational/transactional systems:
- Different platforms, different databases, is process/product oriented, contains
internal/external data, structured and unstructured/data, no integrated data, inconsistent
data/low quality, limited historic data, transforming processing performance is key, hampered
by BI activities.
ELT staging area: temporary storage of data
- Extraction: extract data from operational/other systems
- Transformation: transform data (unification, cleaning, conversion)
- Loading: data loaded into data warehouses
Data warehouse: subject oriented integrated data that is consistent (high quality, historic, query
performance is key). Has star/snowflake model (multidimensional model). Is pre-aggregated and has
no current data.
Data marts: subset of data warehouse (less dimensions, history and detail), user group/application
base oriented. Multi-dimensional cubes. Can be both dependent or independent. Dependent data
mart updates when data warehouse updates. Independent data mart has its own ETL connection.
Meta data/master data: very important, is data about data (location, meaning, origin, etc.) contains
data models, definitions etc. Often from different sources (high maintenance) solution is one shared
repository or interchange using standards between sources (XML).
Multidimensional structures: cubes (Microsoft power BI uses these)
OLAP cubes: idea of slicing and dicing to get your report
Data warehouse working definition: “a database that is maintained separately from the
organization’s operational databases for the sole purpose of managerial decision-making”
Data warehousing: the process of constructing & using data warehouses.
3
