Business Rules: A Case Study - High-Performance Web Databases: Design, Development, and Deployment - page 134

Exhibit 9-3. Erwin entity-relationship model showing entity definition rules.
(LDM3). Similarly, without primary identifier rules, step LDM4, propagation
of foreign keys, cannot be completed. LDM5, identification of key insert and
delete rules, also requires additional business rules.
Step LDM6, add non-key attributes, requires more investigation, although
business rule 5 hints at some attributes for the CONTRACT entity. Each of
these attributes needs its own attribute definition and domain business rule.
Additional attributes also need to be discovered for this and other entities.
The business rule administrator begins to feel confident that classifica-
tion of these business rules into data modeling constructs validates the
business rules themselves. Business rule-driven data modeling is an excel-
lent approach to developing entity-relationship models.
Additional business rule acquisition sessions are needed to solicit more
business rules. A preliminary entity-relationship diagram, however, can be
presented to Mary, the nanny, the customer, and the neighbor. Exhibit 3
depicts a sample entity-relationship diagram produced using ERwin from
Logic Works, Inc. In this rendition of the diagram, the entity boxes display
the business rule that defines the entity, although this model does not spe-
cifically depict the 13 business rules.