Thus, methodologies for developing operational and DSS systems
are different. For instance, in operational systems (a) the development cycle is process
driven, based on a stable data model; (b) data must be normalized in order to
support transaction processing; (c) hardware is defined in the planning phase, remaining
quite stable; and (d) there is no periodic data loading. In DSS, we have (a) the
development cycle is data driven; (b) data is generally denormalized; (c) hardware
changes dynamically; and (d) periodical data loading is a typical process.
In spite of the popularity gained by DSS in the last decade, a methodology for software
development has not been agreed upon. Thus, it is not surprising that most
contributions on requirements analysis for DSS came from consulting companies
and software vendors. The NCR methodology is aimed at developing and main-
Data Quality-Based Requirements Elicitation for Decision Support Systems 6
Copyright ?© 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
taining the data warehouse infrastructure, assuring data quality, and improving
performance encouraging the use of traditional database design techniques. The
SAS Institute Rapid Development methodology is based on the argument that the
two great sources of failure of data warehouse projects are the lack of experience
and the development of very large projects.
Pages:
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145