Conceptual data models and entity relationship modelling study guide

A method for producing a model

Begin by extracting the nouns and important concepts from the requirements. These provide candidates for entities, but each candidate must be checked: does the scenario need to store information about it? If so, keep it as an entity. Next, extract the pieces of information required about each entity. These become attributes. Avoid placing an attribute under an entity if the requirement actually associates it with another entity.

For every entity, decide which attribute or combination of attributes forms its entity identifier. The identifier must distinguish the entity occurrences represented by that entity. Mark the identifier clearly in the entity description, for example Customer (__CustomerID__, Name) and Product (__ProductCode__, Description). If the requirements indicate that more than one attribute is needed to identify an entity, show the complete identifier rather than selecting an attribute without justification.

Connecting the model to the scenario

After listing the entities and attributes, identify statements in the requirements that connect entities. These statements provide the basis for relationships in the entity relationship diagram. For example, if the requirements describe customers placing orders, a relationship between Customer and Order is justified. The diagram is not a separate invention: it is a visual representation of the model produced from the requirements.

Worked reasoning pattern

Suppose a simple scenario requires information about customers and orders. The reasoning is: Customer and Order are entities because information is required about both; customer ID and order ID are possible identifiers; names and order dates are attributes; and the statement that a customer places an order supports a relationship between the entities. A suitable description could be Customer (__CustomerID__, Name) and Order (__OrderID__, OrderDate). The exact attributes must still be checked against the given requirements.

Exam application and self-check

When given a scenario, annotate each proposed entity with its attributes before drawing the diagram. Then compare the diagram with the descriptions. Ask: Is every required entity present? Is every listed attribute attached to the correct entity? Is an identifier shown for each entity? Is every relationship supported by the scenario? Finally, ensure that the diagram and the Entity (Attribute1, Attribute2, ...) descriptions express the same model. Do not add detail simply because it seems realistic; use the given data requirements as the basis for each modelling decision.