In this research, a new data warehouse system for textile- and apparel-related econometric and demographic data was developed, and the system was utilized to demonstrate some of the key capabilities - in particular, those capabilities which were lacking in existing systems.
The system, called the Textile and Apparel Business Information System (TABIS), integrated data from over twenty public and private sources into a data warehouse using the relational database structure, and a code-generating interface was custom-written using a high-level programming language to provide maximum flexibility (see Section 4 and Section 5). The integrated data warehouse structure allows data from previously disparate sources to be combined, and analyzed together. Many new and innovative techniques were developed, both in designing the system infrastructure and providing data analysis capabilities, allowing users to perform analyses that were not possible with the previously existing systems.
The Results section contains several significant samples showing how TABIS was utilized. Many data visualization techniques were developed, including: dot maps which show the distribution and importance of textile and apparel employment by county (Section 6.2.3), pie chart maps which show how the percentage of apparel purchased by men and women vary by state (Section 6.2.4), and animations which show how apparel consumption and population have changed over time (Section 6.3). Also, several forecasting techniques were demonstrated in Section 6.4, with a detailed example of "interactive data exploration and analysis" presented in Section 6.4.3. Section 6.5 shows the flexibility with which TABIS can be accessed, ranging from ASCII access via dial-in and dumb-terminal sessions, through full X-Windows graphical access - the data can even be accessed over the Internet using the World Wide Web (WWW).
The research makes a valuable contribution to both academia and industry, utilizing current computer technology to provide data analysis capabilities that were not previously available. With the infrastructure in place, and the capabilities of the system demonstrated, the next step will be to invite others, from both academia and industry, to utilize the framework and infrastructure provided by the system to solve problems in their respective areas of expertise.