The systems analyst needs to make use of the conceptual freedom afforded by data flow diagrams, which graphically characterize data processes and flows in a business system. In their original state, data flow diagrams depict the broadest possible overview of system inputs, processes, and outputs, which correspond to those of the general systems model discussed in Chapter “Understanding and Modeling Organizational Systems“. A series of layered data flow diagrams may be used to represent and analyze detailed procedures in the larger system.
To better understand the logical movement of data throughout a business, the systems analyst draws data flow diagrams (DFDs). Data flow diagrams are structured analysis and design tools that allow the analyst to comprehend the system and subsystems visually as a set of interrelated data flows.
Graphical representations of data movement storage and transformation are drawn with the use of four symbols: a rounded rectangle to depict data processing or transformations, a double square to show an outside data entity (source or receiver of data), an arrow to depict data flow, and an open-ended rectangle to show a data store.
The systems analyst extracts data processes, sources, stores, and flows from early organizational narratives or stories told by users or revealed by data and uses a top-down approach to first draw a context-level data flow diagram of the system within the larger picture. Then a level 0 logical data flow diagram is drawn. Processes are shown and data stores are added. Next, the analyst creates a child diagram for each of the processes in Diagram 0. Inputs and outputs remain constant, but the data stores and sources change. Exploding the original data flow diagram allows the systems analyst to focus on ever more detailed depictions of data movement in the system. The analyst then develops a physical data flow diagram from the logical data flow diagram, partitioning it to facilitate programming. Each process is analyzed to determine whether it should be a manual or automated procedure.
Six considerations for partitioning data flow diagrams include whether processes are performed by different user groups, processes execute at the same times, processes perform similar tasks, batch processes can be combined for efficient processing, processes may be combined into one program for consistency of data, or processes may be partitioned into different programs for security reasons.
Once you have mastered the material in this chapter you will be able to:
- Comprehend the importance of using logical and physical data flow diagrams (DFDs) to graphically depict data movement for humans and systems in an organization.
- Create, use, and explode logical DFDs to capture and analyze the current system through parent and child levels.
- Develop and explode logical DFDs that illustrate the proposed system.
- Produce physical DFDs based on logical DFDs you have developed.
- Understand and apply the concept of partitioning of physical DFDs.