Business Process Engineering

data mining

In most organizations, critical business data is distributed in systems that are at best loosely integrated and at worst completely independent. Management often has the need to combine data from these separate systems to understand the overall health of the organization. A data warehouse bridges the gap between these systems by collecting the appropriate data from the operational systems of a business and storing it in a manner that is useful from a reporting perspective. Since business rules and data structures are unique to any given operational system, clearly understanding the business process is crucial to extracting and presenting the aggregate date.

For a recent client, CTA used standardized integration templates to meet this challenge. The first step in this process was to define a standardized, generic integration process for new or legacy data sources. To make this process repeatable, a new source template was constructed that specified a set of data to collect during the discovery phase for any new source. The information captured using this template became the foundation for the functional and physical requirements for that source. This information also enabled the end user to identify whether new or merged systems could be readily integrated with existing systems, and provided a guide to the necessary steps required for integration. Understanding the impact of a major change to a business operational system required careful analysis of both the operational system and the data warehouse. Capturing this information in the integration templates reduced the amount of time spent by the development team in manually analyzing the impact of changes. Used in conjunction with detailed report specifications and report format mockups, the likelihood of development errors was significantly reduced as all impacted programs were readily identifiable.

Standard templates were also created to provide a strategy for the addition of new reports. These templates identified the data elements and specific criteria used in the data warehouse configuration and were used to determine the extent of modifications or enhancements to existing extraction processes for new data entities. The report specification included all of the information needed to build the report, such as a list of data elements and report formatting considerations. This served as the documented requirements set for each report.