5
Introduction
oday, most organizations in all sectors of industry, commerce and
government are fundamentally dependent on their information
systems. In industries such as telecommunications, media,
entertainment and financial services, where the product is already or is
been increasingly digitized, the existence of an organization crucially
depends on the effective application of information technology.
Consequently, organizations are increasingly looking toward the
application of technology not only to underpin existing business
operations but also to create new opportunities that provide them with a
source of competitive advantages.
Let’s take a look about the evolution of Information systems over time
(Chow D. 2009)1. Until the 1960s, the role of most information system
was simple. They were mainly used for electronic data processing (EDP)
purposes, such as transactions processing, recording-keeping and
accounting. EDP is often defined as the use of computers in recording
classifying, manipulating, and summarizing data. It is also called
Transaction Processing Systems2 (TPS), automatic data processing, or
information processing. In the 1960s, another role was added to the use of
computers: the processing of data into useful informative reports. The
concept of Management Information Systems3 (MIS) was born. This new
role focused on developing business applications that provided managerial
end users with predefined management reports that would give managers
the information they needed for decision-making purposes. By the 1970s,
these pre-defined management reports were not sufficient to meet many of
the decision-making needs of managements. In order to satisfy such
1
David Chow.(2009) Principles of Auditing and Mangement Information Systems.
2
These process data resulting from business transactions, update operational databases, and
produce business documents. Examples: sales and inventory processing and accounting
systems.
3
MIS provide information in the form of prespecified reports and displays to support business
decision-making. Examples: sales analysis, production performance and cost trend reporting
systems.
T
6
needs, the concept of Decision Support System4 (DSS) was born. The new
role for information systems was to provide managerial end users with ad
hoc and interactive support of their decision-making processes.
In the 1980s, the introduction of microcomputers into the workplace
ushered in a new era, which led to a profound effect on organizations. The
rapid development of microcomputer processing power, application
packages, and telecommunication networks gave birth to the phenomenon
of end user computing. End users could now use their own computing
resources to support their job requirements instead of waiting for indirect
support of a centralized corporate information services department. It
became evident that most top executives did not directly use either the
MIS reports or the analytical modeling capabilities of DSS, so the concept
of Executive Information Systems5 (EIS) was developed. Moreover,
breakthroughs occurred in the development and application of Artificial
Intelligence (AI) techniques to business information systems. With a less
need for human intervention, knowledge workers could be freed up to
handle more complex tasks. Expert Systems6 (ES) and other Knowledge
Management Systems7 (KMS) also forged a new role for information
systems. ES can serve as consultants to users by providing expert advice
in limited subject areas. The mid- to late 1990s saw the revolutionary
emergence of Enterprise Resource Planning (ERP) systems. This
organization-specific form of a strategic information system integrates all
facets of a firm, including its planning, manufacturing, sales, resource
management etc. (virtually every business function). The primary
advantage of these ERP systems lies in their common interface for all
4
DSS provide interactive ad hoc support for the decisional-making processes of amnagers and
other business professionals. Examples: product pricing, profitability forecasting and risk
analysis systems.
5
EIS provide critical information from MIS, DSS and other sources, tailored to the information
needs of executives. Examples: systems for easy access to analysis of business performance,
actions of all competitors, and economic developments to support strategic planning.
6
ES: knowledge-based system that provide expert advice and act as expert consultants to
users. Examples: credit application advisor, process monitor, and diagnostic maintenance
systems.
7
KMS: knowledge-based systems that support the creation, organization and dissemination of
business knowledge within the enterprise. Examples: intranet access to best business practices,
sals porposal strategis and customer problem resolution systems.
7
computer-based organizational functions and their tight integration and
data sharing needed for flexible strategic decision making.
The rapid growth of the Internet, intranets, extranets and other
interconnected global networks in the 1990s dramatically changed the
capabilities of information systems in business. Internet-based (and web-
enabled) enterprise and global electronic business (and commerce)
systems are becoming commonplace in the operations and management of
today’s business enterprises. The internet and related technologies and
applications have changed the way businesses operate and people work,
and how information systems support business processes, decision-making
and competitive advantage.
Information system can support a variety of management decision-making
levels and decisions. These include the three levels of management
activity:
ξ Strategic management: It is typical for a board of directors and an
executive committee of the CEO and top executives to develop the
overall organization goals, strategies, policies and objectives as part
of a strategic planning process. They also monitor the strategic
performance of the organization and its overall direction in the
political, economic and competitive business environment.
ξ Tactical management: Increasingly, business professionals in self-
directed teams as well as business unit managers develop short and
medium range plans, schedules and budgets and specify the
policies, procedures and business objectives for their sub-units of
the company. They also allocate resources and monitor the
performance of their organizational sub-units, including
departments, divisions, process teams and other workgroups.
ξ Operational management: The members of self-directed teams or
operating managers develop short-range plans such as weekly
production schedules. They direct the use of resources and the
performance of tasks according to procedures, and within budgets
and schedules they establish for the teams and other workgroups of
the organization.
8
During the lasts years have been developed various tools and techniques,
belonging to the Information Systems that enhance value of information.
These are:
ξ Data warehouse (DW): a data warehouse stores data that have been
extracted from the various operational, external and other databases
of an organization. It is a central source of the data that have been
cleaned, transformed and catalogued so they can be used by
managers and other business professionals for data mining, online
analytical processing and other forms of business analysis, market
research and decision support.
ξ Data mining (DM): is a major use of DW database and the static
data they contain. In data mining, the data in a DW are analyzed to
reveal hidden patterns and trends in historical business activity.
They can be used to help managers make decisions about strategic
changes in business operations to gain competitive advantages in
the marketplace. DM software analyzes the vast stores of historical
business data that have been prepared for analysis in corporate DW
and tries to discover patterns, trends, and correlations hidden in the
data that can help a company improve its business performance.
ξ Online analytical processing (OLAP): enables managers and
analysts to interactively examine and manipulate large amounts of
detailed and consolidated data from many perspectives. OLAP
involves analyzing complex relationships among thousands or even
millions of data items stored in data marts, DW and other multi-
dimensional database to discover patterns, trends and exceptional
conditions. An OLAP session takes place online in real time, with
rapid responses to a manager’s or analyst’s queries, so that their
analytical or decision-making process is undisturbed.
So more we go ahead over time and more the systems (not only
information systems) are complex. But, leaning about complex systems
when you also live in them is difficult. We are all passengers on an
aircraft; we must not only fly but redesign in flight (Sterman, 2000). In
this situation we can take benefit from System Dynamics (SD). System
9
dynamics is a method to enhance leaning in complex systems. Just as an
airline uses flight simulators to help pilots learn, system dynamics is,
partly, a method for developing management flight simulators, often
computer simulation models, to help us learn about dynamic complexity,
understand the sources of policy resistance, and design more effective
policies. But learning about complex dynamics systems requires more
than technical tools to create mathematical models. System dynamics is
fundamentally interdisciplinary. System dynamics is grounded in the
theory of nonlinear dynamics and feedback control developed in
mathematics, physics, and engineering. Because we apply these tools to
the behavior of human as well as physical and technical systems, system
dynamics draws on cognitive an social psychology, economics, and others
social sciences.
So the aim of this thesis is try to make a wider use of data, patterns and
information, extracted directly from databases of existing information
systems, for build system dynamics models based on a balanced
scorecard. This work foresees first the building of a normal SD model
based on a balanced scorecard, and then of an SD model, also based on the
balanced scorecard, supported by information systems. In this case, a data
mining analysis on a data warehouse is exploited to extract information,
data and patterns, which will be used for the building of a “pumped” SD
model.
The ideas will be explained better in the chapter one, where I’ll give also
theoretical explanation of the why and the how.
In the second chapter you’ll find a description of the tools and the
techniques that are useful for the work, not all of the techniques and tools
will be used.
An explanation of which techniques and tools I will use is write in the
third chapter. In this chapter you will find also a technical explanation of
the work and all the steps used to reach the purpose. Moreover, in the
chapter three there is also the case study.
10
1. DATA MINING IN SUPPORT OF DYNAMIC
MODELS
1.1 Underlying ideas
ince the beginning of human economy, people used to save data
about their business in various ways. Initially the quantity of
recorded data was really meager because the only support was the
paper and they had to write everything by hand. Besides the difficult in
saving data there was also the difficulty in using data. Things began to
change with the birth of computers, and even more with the enlargement
and perfusion of storage media.
At the same time, market is continually evolving, and companies doing
business in a dynamic and complex world, that is why is important garner
a massive amount of data, and use them in the most correct and profitable
way.
Companies understood that data can be used like a leaning opportunity,
and every kind of recorded data is a potential knowledge source: each call
to customer support, each point of sale transaction, each catalog order,
each visit to company web site. But learning requires more than simply
gathering data. In fact, many companies gather hundreds of gigabytes or
terabytes (today also petabytes) of data from and about their customers,
their suppliers, their employees, and themselves without learning
anything! Data are gathered because it is needed for some operational
purpose, such as inventory control or billing. And, once it has served that
purpose, it languishes on disk is discarded. So a lot of potential
information is unused and a lot of precious data are lost.
S
11
For learning to take place, data from many sources must first be gathered
together and organized in a consistent and useful way. This is called data
warehousing, and allows the enterprise to remember what is noticed
about, for example, its customers. A good data warehouse provides access
to the information gleaned from transactional data in a format that is much
friendlier than the way it is stored in the operational systems where the
data originated. Ideally, data in the warehouse has been gathered from
many sources, cleaned, merged, tied and summarized in various useful
ways. Reality often falls short of this ideal. 8
The data warehouse provides the enterprise with a memory. But, memory
is of a little use without intelligence. Intelligence is the key to give an
information gains to the data stored in memory. It allow us to comb
though our memories, noticing patterns, devising rules, coming up with
new ideas, figuring out the right questions, and making prediction about
the future. So we need tools and techniques that add intelligence to the
data warehouse. The purpose of these techniques is to exploit the large
quantities of data generated by interaction with the external world (for
example, with customers) and prospects in order to get to know them
better. Data mining techniques, belonging to the family of Business
intelligence techniques, include the right tools to perform this task. So, the
goal of data mining is to find patterns in historical data that shed lights on
those needs, preferences, and propensities. The task is made difficult by
the fact that the patterns are not always strong, and the signals sent by
customers are noisy and confusing. Separating signal from noise is an
important role of data mining, but not the only one. In fact, data mining
comes in two flavors: directed and undirected. Directed data mining
attempts to explain or categorize some particular target field such as
income or response. Undirected data mining attempts to find patterns or
similarities among groups of records without the use of particular target
field or collection of predefined class. Both these flavors are discussed
more deeply in later chapter.
8
Berry M. J. A., Linoff G. S. (2004: from pag. 5,6 ) Give a most accurate definition about the
application in Customer Relationship Management.
12
At this point some questions come to mind: why is so important catch
signals from data? Which are the advantages in finding patterns in
historical data? In which way can be used information?
From the perspective of this thesis the answer is one: try to improve the
understanding and the dominance of the dynamic models and simplify the
complexity of systems. In fact, the greatest constant of modern times is
change. We live in a very dynamic era, where real systems are even more
complex and interrelationships are even thicker. Accelerating changes in
technology, population and economic activity are transforming our world,
starting from small changes to bigger ones. Build models about systemic
entity is a way to simplify the observed reality and allows to study things
in a more efficient way.
So, the field of study and the topic of this thesis is use the modeling tool
called System Dynamics (we will talk over about this argument in later
chapter), supported by data mining techniques which, through data
analysis, allow to find hidden information, hidden relationship and hidden
pattern. Include that further information in a system dynamics model
(should) enable to simplify and improve construction, use and
comprehension of dynamic models, generating, in this way, models that
simulate and fit as best as possible the behavior of a real system.
So, in order to overcome the shortcomings of traditional modeling
methods, I present a data mining-based approach to assist SD model
construction.
13
1.2 Holistic vision
“The whole is more than the sum of its parts”. 9
Holism (from holos, a Greek word meaning all, whole, entire, total)
is the idea that all the properties of a given system (physical, biological,
social, economic, etc.) cannot be determined or explained by its
components part alone. Instead, the system as a whole determines in an
important way how the parts behave.
The term Holistic vision comes in two flavors: holism like property of
systems and holism like property of integration among Balanced
Scorecard, System Dynamics and Data Mining. In this chapter we discuss
only about the second flavor.
1.2.1 Why integration?
Integration is the act of combining into an integral whole; and
considering the previous quotation (Aristotele) we can understand why
integration is better. Further this philosophical understanding real facts
often prove that integration is the way for get improvement in several
fields. Integration means combine together separate but similar system or
also completely different systems. System integration is even more an
emerging model of industrial organization. In fact, in the past decade or
so, a new kind of systems integration has become a key factor in the
operations strategy, and competitive advantage of major corporations in a
wide variety of sectors (e.g. computing, automotive, communication etc. ).
Due to the expansion in use of integration, it has gained even more
importance over time. In fact, in the past, systems integration was
confined to a technical operations task; but today, systems integration is a
strategic task, which pervades business management not only at the
engineering level but also in senior management decision-making.
9
Aristotele, in the Metaphysics, summarized in this way the general principle of holism.
Ricevi informazioni sui nostri servizi, sulle offerte e non perdere news e consigli su università e lavoro.
