Data Modeling with the Entity-Relationship Model - Database Processing: Fundamentals, Design, and Implementation - page 198

Figure 5-21
Report Indicating an
Association Pattern

PartNumber is the identifier of PART and CompanyName is the identifier of COMPANY.
Hence, QUOTATION is ID-dependent on both PART and C OMPANY.
In Figure 5-22, then, the relationships between PART and QUOTATION and between
COMPANY and QUOTATION are both identifying. This fact is shown in Figure 5-22 by the solid,
nondashed line that represents these relationships.
As with all identifying relationships, the parent entities are required. Thus, the minimum
cardinality from QUOTATION to PART is one, and the minimum cardinality from QUOTATION
to COMPANY also is one. The minimum cardinality in the opposite direction is determined by
business requirements. Here a PART must have a QUOTATION, but a COMPANY need not
have a QUOTATION.

By ThE Way Consider the differences between the data models in Figure 5-20 and
Figure 5-22. The only difference between the two is that in the latter the
relationship between COMPANY and PART has an attribute, Price. Remember this
example whenever you model an N:M relationship. Is there a missing attribute that per-
tains to the combination and not just to one of the entities? If so, you are dealing with
an association, ID-dependent pattern and not an N:M, strong entity pattern.

PART COMPANY

Figure 5-22
Association Pattern Data
Model for the Report in
Figure 5-21

PartNumber CompanyName

PartName
SalesPrice
ReOrderQuantity
QuantityOnHand

City
Country
Volume

QUOTATION
PartNumber
CompanyName

Price