Data Modeling: DFDs and ERDs

• The purpose of this is to introduce data modeling.
• It details data flow diagrams (DFDs) and their usefulness in representing the various processes that act on data.
• It also explores entity relationship diagrams (ERDs).
• This model is a graphical technique for organizing and analyzing business information.

The Purpose of Data Flow Diagrams

• Data Flow Diagrams (DFDs) emphasize processes that transform data within a system.
• They illustrate the relationship between information sources, data stores, and processes to show where information comes from and where it is stored.
• DFDs help discover redundancies in processes and information storage.
• They show what data are exchanged between processes, acting as a map of data pathways within the system.
• DFDs show the system at different levels of abstraction, from high-level diagrams grouping complex processes to more detailed diagrams breaking them down into components.
• Data structures can also be divided into portions to represent separate data streams, e.g., items ordered, shipping, and billing addresses.
• Analysis typically starts with a high-level diagram showing the data system's context relative to external data sources like customers and suppliers; processes are then broken down into more specific levels.

The Visual Language of Data Flow Diagrams

• Techniques available for data flow diagramming processes include Yourdon method and the Gane-Sarson.
• The focus will be on the Gane-Sarson.
• It has four basic symbols:
• a double box for entities
• a rounded box for processes
• an arrow for data flows
• an open-ended box for data stores
• An entity is an actor, organization, or system outside the boundaries of the system.
• Examples: Customer, Supplier, Shipping, or Bank.
• Processes indicate that data are being transformed or manipulated.
• Data flows represent information moving between the objects in the diagram.
• Specificity, consistency, completeness and clarity are stressed.
• A data store represents data that are kept for a period of time, often a database.

Refining a Process with Data Flow Diagrams

• DFDs can be used to refine a system.
• Start by sketching the process at a high level, with the entire system represented as a single process surrounded by external entities and data stores.
• The diagram should be abstract enough to have between 5 and 9 elements plus data flows, using single-word names.
• This helps to clearly think about and communicate the scope of your analysis.

Analyzing Data Flow Diagrams

• Creating a DFD reveals ambiguities and pieces of information that had been overlooked.
• Data flow diagrams should follow these rules:
• A series of data flows always starts or ends at either an external agent or a data store.
• A process must have both data inflows and outflows.
• All data flows must be labeled.
• Data flows are named as descriptive nouns.
• A data store must have at least one data inflow.
• A data flow cannot go between an external agent and a data store; a process must be in between.
• A data flow cannot go directly between two external agents; a process must be in between.
• A data flow cannot go between two data stores; a process must be in between.
• External agents and data flows can be repeated on a process model to avoid crossing lines, but do not repeat processes.
• Data stores must have processes for creating, retrieving, updating, and deleting data (CRUD).
• Consider the system from the point of view of events, making a list of all the events that initiate system processes.
• Compare this list with the data flow diagram to ensure that each event and response is integrated within the model.
• DFDs are a useful tool for analyzing complex systems and understanding how information is handled.
• They help document the current system, clarify inefficiencies, and design new processes.
• A data flow diagram shows how the system should work and what information needs to be collected and provided during different steps.