Relationship (Conceptual Modelling)

Context: FIT2094_MOC · a logical association between entities · classified by degree (unary/binary/ternary) · labelled with a verb

Quick Revision

• 🎯 Objective: a logical association between entities ➔ the only way entities connect on a conceptual model.
• 📦 Core Components: degree unary/binary/ternary ➔ verb label ➔ min/max cardinalities.
• ⚡ Key Constraint: ternary resolved by a composite entity; two entities may share multiple distinct relationships.

📝 Core

1. The Relationship

• Definition ➔ a logical association (“a customer places many orders”).
• Only connector ➔ the sole way entities connect on a Conceptual Model.
• Degree ➔ number of entities involved.

2. Degree

• Unary (recursive) ➔ entity with itself (EMPLOYEE manages EMPLOYEE).
• Binary ➔ two entities (most common; CUSTOMER places ORDER).
• Ternary+ ➔ >2 entities ➔ resolved by a composite entity (PRESCRIPTION).

3. Multiplicity of Relationships

• Multiple links ➔ two entities can share more than one relationship, modelled separately.
• Example ➔ EMPLOYEE member of TEAM and leader of TEAM = two lines.
• Label direction ➔ name in the 1:M direction (places, has).

⚙️ Core Implementation

🔹 Binary relationship

Mermaid erDiagram

erDiagram
  CUSTOMER ||--o{ ORDER : places

💡 Common Mistake: Model only the brief’s relationships ➔ a derivable/redundant association (already implied by existing relationships) must not be added; the brief decides, not intuition.

⚖️ Core Decision Matrix

Degree	Entities	Resolution
unary	1 (recursive)	self-referencing FK
binary	2	direct line
ternary+	≥3	composite entity
M:N binary	2	Associative Entity (if attributes)

When It Flips: M:N is kept on the conceptual model (simpler/desirable) and resolved into an Associative Entity only when attributes must be stored or at the logical stage. A relationship here is the ER counterpart of a Binary Relation.

📊 Exam Execution Trace

Manual Execution Trace

Classifying by degree:

Step / State	Relationship	Degree	Note
0 (Init)	—	—	—
1	EMPLOYEE manages EMPLOYEE	unary	recursive
2	CUSTOMER places ORDER	binary	common
3	doctor–drug–patient	ternary	→ PRESCRIPTION

Applied Exercise

Problem: EMPLOYEE is a member and a leader of a TEAM. Model it; may a derivable relationship be added? Derivation Proof / Hand-Calculation Walkthrough:

$$\begin{array}{rl}\text{two distinct associations}& \Rightarrow \text{two separate lines (member, leader)}\\ \text{derivable relationship}& \Rightarrow \text{NOT added (brief rule)}\end{array}$$

Final Extracted Output: two labelled relationships; no redundant/derivable relationship added.

🧠 Active Recall

Define relationship degree and give unary/binary/ternary examples.

• Hint: Entity count.

Answer

• Short answer: Degree = entities involved; unary (EMPLOYEE manages EMPLOYEE), binary (CUSTOMER places ORDER), ternary (doctor–drug–patient).
• Why: Ternary resolution ➔ add a composite entity (PRESCRIPTION).

If two entities have two real-world associations, how are they modelled, and may you add a relationship not in the brief?

• Hint: Separate lines; brief-only.

Answer

• Short answer: Two separate relationships; distinct associations can’t merge; no unstated relationship added.
• Why: Exactly the brief ➔ even a derivable/redundant relationship is excluded.