How machines quantify words like “hot,” “cold,” and “almost”

Why “Hot” Is Not a Number

When we say:

“The weather is hot”

What do we really mean?

• 30°C might feel hot to one person
• 35°C might be the threshold for another
• 28°C might still be “warm,” not “hot”

This ambiguity is exactly what traditional systems fail to capture.

Fuzzy logic solves this problem using one of its most important tools:

The Membership Function (MF)

Membership Functions: Turning Language into Math

A membership function defines:

How much a value belongs to a fuzzy set

Instead of assigning a strict label, it assigns a degree of belonging between 0 and 1.

Example: Temperature

Let’s define fuzzy categories:

• Too Cold
• Cold
• Warm
• Hot
• Too Hot

Now, instead of saying:

30°C = Hot (True/False)

We say:

• 30°C → 0.6 Hot
• 30°C → 0.3 Warm
• 30°C → 0.1 Too Hot

This value (between 0 and 1) is called:

Degree of Membership

The Core Idea: Partial Truth

Membership values introduce a powerful concept:

Truth is not binary — it’s gradual.

• 0 → completely not in the set
• 1 → fully in the set
• Between 0 and 1 → partially true

This is what allows fuzzy systems to model human reasoning.

Shapes of Membership Functions

Membership functions are not random — they follow specific mathematical shapes.

Each shape represents a different way of modeling uncertainty.

1. Triangular (Simple & Fast)

• Defined by three points
• Looks like a triangle
• Efficient and widely used

Best for: simple systems and real-time applications

2. Trapezoidal (More Flexible)

• Flat top → allows full membership over a range
• More realistic than triangular

Best for: stable ranges (e.g., “comfortable temperature”)

3. Gaussian (Smooth & Natural)

• Bell-shaped curve
• Smooth transitions between states

Best for: modeling natural phenomena

4. Sigmoid (Gradual Transitions)

• S-shaped curve
• Good for “increasing” or “decreasing” behavior

5. Piecewise Linear

• Built from straight-line segments
• Highly customizable

6. Singleton (Special Case)

A fuzzy singleton is a unique type of fuzzy set:

• Membership = 1 at exactly one point
• Membership = 0 everywhere else

It behaves almost like a crisp value — but still fits within fuzzy logic.

The Fuzzy Logic System (FLS): The Full Pipeline

Membership functions are just one part of a bigger system.

A Fuzzy Logic System (FLS) transforms inputs into outputs through four stages:

1. Fuzzification

Convert crisp inputs into fuzzy values using membership functions.

Example:

Temperature = 30°C → partially Hot, partially Warm

2. Rule Evaluation

Apply fuzzy rules:

IF temperature is Hot → fan speed is Fast

3. Inference Engine

Combine rules and determine output fuzzy sets.

4. Defuzzification

Convert fuzzy output back into a crisp value.

Example:

Fan speed = 72%