Analyzing Data › Data Modeling Terminology › Subject Area

Subject Area

A subject area is a way to group together entity types in the same general area of interest.

For example, business activities within the scope of analysis deal with SALES and INVENTORIES. The names of subject areas are plural nouns by convention.

A subject area itself be composed of other, smaller subject areas that contain entity types. A subject area that contains only entity types is named a primitive subject area.

In the following table, CUSTOMERS, ORDERS, and INVENTORIES are primitive subject areas.

At some point in analysis, a subject area contain both entity types and other subject areas. Try to resolve this unbalanced situation.

For example in the table, the SALES subject area includes the subject areas CUSTOMERS, which includes the entity types CUSTOMER and DELIVERY POINT, and ORDERS, which includes the entity types ORDER, ORDER ITEM, and SALESPERSON. SALESPERSON is included in another subject area within SALES, or does it belong in some other subject area such as HUMAN RESOURCES?

Decomposing subject areas is discussed further in the chapter Building the Analysis Model.

In analysis, you can use subject areas in two ways:

• Progressively decompose data starting with subject areas; then subdivide them into more focused subject areas, based on the highest-level business functions within the project scope. As each entity type is discovered, it is added to one of the subject areas. Data analysis proceeds in parallel with activity analysis, and the data and activity results are kept in step at each level of analysis.

  Note: For more information, see Building the Analysis Model.

• Begin analysis without subject areas and later assign each of the entity types that are discovered to one of a set of subject areas. This may be done based on common usage of entity types by processes or because entity types are closely related.

  Note: For more information, see Analyzing Interactions.

In either case, the use of subject areas permits project teams to compare what they have each discovered about a subject area, for example, many parts of the business uses PRODUCTS.

A project can be given a CA Gen starter model that already contains the set of subject areas of the enterprise and the relevant entity types that have already been defined, as part of an Information Strategy Plan or previous system development projects.

An Information Strategy Plan exists, a few subject areas are defined and used as a high-level structure for the set of entity types that are identified during planning.

A complete Information Architecture, developed progressively by many development projects, include hundreds of entity types, so subject areas are a convenient means of dealing with related entity types and coordinating the needs of separate business systems for the same types of data. For this reason, even without the existence of an Information Strategy Plan, development coordinators may have provided a set of subject areas as a starting point for the analysis work of each development project.

Note: For more information about the examples of Entity Relationship Diagrams structured in subject areas, see Drawing Entity Relationship Diagrams.