The use of data warehouse in decision making proce Essay

This essay has a total of 1770 words and 16 pages.

the use of data warehouse in decision making process


It is obvious that there is no organization running without data. The data can be viewed
as tangible assets of an organization just as any physical asset. So, they need to be
stored and made available to those who need them when they need them. However, the data by
themselves are useless. So, they must be put together to produce useful information. In
turn, information becomes the basis for relational decision making. To facilitate the
decision-making process, a new development of database systems was developed called
“data warehouse”.

The data warehouse can be generally described as a decision-support tool that collects its
data from operational databases and various external sources, transforms them into
information and making that information available to decision-makers (top managers) in a
consolidated and consistent manner. (2:64)(4:82)


The data warehouse is not more than a database but separated from other databases like the
operational database distributed database and text database. When did management start to
utilize this powerful tool and why they seek to use it.

The data warehouse has been developed at the beginning of 1980s. However, it was optimize
to transform non-organized and lightly summarized data from the operational database into
analytical tool that supports intelligent decision-making. (6:19)

The term DSS (Decision support system) database is used interchangeably with the data
warehouse. On the other hand, other names for the operational database are transactional
database and production database.


The data warehouse can be very simply defined as an integrated, subject-oriented, time
variant and non-volatile database that provides support for decision-making (5:39) (6:19).
The following four sections will explain what this definition means.


The data warehouse is a centralized database that integrates data from different sources
(6:19) with diverse formats. This integration of the data provides a unified view of the
overall organizational situation. Data integration enhances decision-making and helps the
manager to better understand the operations of the organization (6:19).


The data in DSS database are organized to provide answers to questions coming from
different areas within the organization. They are arranged by topic such as sales,
marketing, finance and so on. The DSS database contains specific subject for each topic
like customer, product, region and so on. This form of data organization is different that
of more process-oriented of the operational database system. (5:39,43)

Time Variant

The data warehouse contains historical data over a long time. Those data reflect what
happened last week, last month, the past five years and the like. (6:19)


Once the data enter the data warehouse, they are never removed or changed. Because the
data warehouse represents the entire history of the organization, the data from
operational database are always added to it. Since DSS data are never deleted and new data
are periodically added, the data warehouse is always growing. That’s why the data
warehouse must be able to have hardware that supports gigabytes and even terabytes size of
databases. (5:43)


The operational database and the DSS database differ in the roles the do as well as the
data characteristics for each one.

Main Role

The transactional database is optimized to support transactions that represent daily
operations (2:67). For example, during the registration period at KFUPM, each time a
student adds, drops courses, or changes sections, he must be accounted for by the
operational database system of the university. So, student data and course data are in
frequent update mode.

On the other hand, the data warehouse is optimized to support data analysis and
decision-making (2:64). Basically, it takes the summarized data from the operational
database, filters them for analysis and decision making processes (2:64). For instance,
the manager of the admission and registration department may ask for the number of
students at KFUPM taking ENGL-214 last summer. The data warehouse answers this query for
him. Then, he would take decision whether to increase number of sections of this
particular course or not.

Operational Data Vs. Warehoused Data

Transactional data and DSS data are different in the summarization level, transaction
type, query activities and dimensionality.

Summarization level

The degree to which DSS data are summarized is very high when contrasted with the
operational data (5:39). For example, rather than storing thousands of sales transactions
for a given store on a given day, the data warehouse might simply store the total number
of units sold and the total price during that day. Then, the store manager may decide
whether to continue or discontinue selling or producing such products.

Transaction type

The operational database and the data warehouse are different in terms of transaction
type. Whereas production data are characterized by update transactions, DSS data are
mainly characterized by query (read only) transactions (2:67). The DSS data also require
periodic update (2:67) to load new data that are summarized from the transactional
database as well as other external sources. Therefore, the warehoused data are historic
(2:64) while the operational data represent transactions as they happen.

Query activity

It is difficult, if not impossible, to optimize a single database for both processing
purposes as well as for decision-making needs. For that reason, the data warehouse is
optimized for ad hoc (on demand or as needed) complex queries needed by decision-makers

The production database, on the other hand, is optimized to allow more processing for the
repetitive update transactions (2:67). So, it is difficult to get ad hoc queries from that
operational database because of the continuously updated transactions.


Dimensionality is the most distinguishing characteristic of the DSS data. The data
warehouse is set to provide the larger picture (2:64). In other words, it includes many
data dimensions. For instance, a sale manager may ask how many units of product X were
sold to customer Y during the last T months (2:66). So, he or she can view the data from
three dimensions: product, customer and time. In fact, (s)he could view the data from many
dimensions. This multidimensional view of the data is different from the single view of
the operational database.

Continues for 8 more pages >>