The Operation of

Fuzzy Set
 Objectives
• Introduces various operations of fuzzy sets
• Introduces the concepts of disjunctive sum,
distance, difference, conorm and t – conorm
operators

Material sources is taken from First Course on Fuzzy Theory and Applications, Lee, K.H.
 Classic/Crisp/Boolean Logic

• Logical AND (∩) • Logical OR (U)

Truth Table Truth Table
A B A∩B A B AUB
0 0 0
0 0 0
0 1 0
1 0 0 0 1 1
1 1 1 1 0 1
1 1 1

A
B A
B

Crisp Intersection Crisp Union

 Stndard Fuzzy Operations
(Fuzzy Union, Intersection, and Complement)

• In Fuzzy Logic, intersection, union and

complement are defined in terms of their
membership functions
• Fuzzy intersection and union correspond to
‘AND’ and ‘OR’, respectively, in
classic/crisp/Boolean logic
• These two operators will become important
later as they are the building blocks for us to
be able to compute with fuzzy if-then rules
 Standard Operations of Fuzzy Set
• Complement set A

• Union

• Intersection
Example 1:
Suppose we have the following (discrete) fuzzy sets:
A = 0.4/1+0.6/2+0.7/3+0.8/4
B = 0.3/1+0.65/2+0.4/3+0.1/4

The union of the fuzzy sets A and B

= 0.4/1+0.65/2+0.7/3+0.8/4
The intersection of the fuzzy sets A and B
= 0.3/1+0.6/2+0.4/3+0.1/4
The complement of the fuzzy set A
= 0.6/1+0.4/2+0.3/3+0.2/4
Example 1: (cont.)
Let’s show the fuzzy sets A and B graphically

1 1

0.8 0.8

0.6 0.6

0.4 0.4

0.2 0.2

0 0
1 2 3 4 5 1 2 3 4 5

A B
Example 2

Given two fuzzy sets A and B

a. Represent A and B fuzzy sets graphically

b. Calculate the of union of the set A and set B
c. Calculate the intersection of the set A and set B
d. Calculate the complement of the union of A and B

A  0.0 /  2  0.3 /  1  0.6 / 0  1.0 / 1  0.6 / 2  0.3 / 3  0.0 / 4

B  0.1 /  2  0.4 /  1  0.7 / 0  1.0 / 1  0.5 / 2  0.2 / 3  0.0 / 4

Example 2 (cont)

a
Example 2 (cont)

Union  max (A, B)  0.1 /  2  0.4 /  1  0.7 / 0  1.0 / 1  0.6 / 2  0.3 / 3  0.0 / 4

Intersection  min (A, B)  0.0 /  2  0.3 /  1  0.6 / 0  1.0 / 1  0.5 / 2  0.2 / 3  0.0 / 4

Complement of (b)  1 - max(A, B)  0.9 /  2  0.6 /  1  0.3 / 0  0.0 / 1  0.4 / 2  0.7 / 3  1.0 / 4
Operations on Fuzzy Sets: Intersection
(figure from Klir&Yuan)
Operations on Fuzzy Sets: Union and
Complement (figure from Klir&Yuan)
Characteristics of standard fuzzy set
operators
Characteristics of standard fuzzy set
operators
• Note that the two characteristics of crisp set
operators does not hold in fuzzy set.

• The reason for this occurrence is that the

boundary of complement of A is ambiguous.
 Example 4: Graphical representation of the Fuzzy
operations (taken from J.M. Mendel, Uncertain Rule-
Based Fuzzy Logic Systems, 2001)
Consider the fuzzy sets A = damping ratio x considerably
larger than 0.5, and B = damping ratio x approximately
equal to 0.707. Note that damping ratio is a positive real
number, i.e., its universe of discourse, X, is the positive
real numbers 0  x  1
Consequently, A  ( x,  A ( x)) xX  and B  ( x,  B ( x)) xX 
where, for example, µA(x) and µB(x) are specified, as:

0 if 0  x  0.5

 A ( x)   1
 1

if 0.5  x  1   B ( x)  0  x 1
1  ( x  0.5) 2
 
1  ( x  0.707) 4
Example 4: (cont.)
Figure (a): µA(x), µB(x) Figure (b): µAUB(x)
Figure (c): µA∩B(x) Figure (d): µB(x), µB(x)
Example 4: (cont.)
• This example demonstrates that for fuzzy sets, the Law of
Excluded Middle and Concentration are broken, i.e., for
fuzzy sets A and B:
 
A  A  X and A  A  O
• In fact, one of the ways to describe the difference
between crisp set theory and fuzzy set theory is to explain
that these two laws do not hold in fuzzy set theory
Standard Fuzzy Operation
(Continued ...)
 Fuzzy Complement
(Requirement for complement function)
• Complement set of A ( ) carries the sense of
negation.
• Complement function C maps the
membership function of fuzzy set A to
[0,1] (written by ).
• Fuzzy complement function should satisfy two
axioms:
 Fuzzy Complement
(Requirement for complement function)
• Axioms C1 and C2 are fundamental requisites to be a
complement function.
• These two axioms (C1 &C2) are called “axiomatic
skeleton”.
• Additional axioms are:
 Fuzzy Complement
(Requirement for complement function)

• The four axioms for complement hold in standard

complement operator
 Example of Complement Function
The standard function and and it’s visual representation
is shown in figure below:

Standard complement set function Illustration of standard complement set function

Example of Complement Function
• The following is a complement function satisfying
only the axiomatic skeleton.

Example of complement set function

Example of Complement Function
• The following complement function is
continuous (C3) but not holds C4

When a = 0.33, C(0.33) = 0.75 in this function. However since C(0.75)

= 0.15 ≠ 0.33, C4 does not hold .
 Popular Complement Function
• Yager’s Function

• When w = 1, the Yager’s function becomes the

standard complement function C(a) = 1 – a.
Yager Complement Function
Fuzzy Partition
 Fuzzy Union

• In general, union of A and B is specified by a function

• This union function calculates the membership degree of union A and

B
 Fuzzy Union
(Axioms for Union Function)

Union function should obey next axioms

 Fuzzy Union
(Axioms for Union Function)
• It is often desirable to restrict the class of fuzzy union by considering various
additional such as:

• The standard operator Max is treading on those six axioms

 Fuzzy Union
(Axioms for Union Function)

Visualization of standard union operation

 Yager’s union function
• Yager’s union function holds all axioms except U6

• The shape of Yager function varies with parameter w.

• For instance,
 Yager’s union function
• What if w increases?
• Supposing w  , Yager union function is transformed
into the standard union function
The example of Yager’s union function
The example of Yager’s union function

Example : let’s reconsider the example: If set A be “young”

and B “senior”, the union of A and B is “young or senior”. In
the sense of meaning, the union and OR logic are
completely identical
 Other Union Operations
(1) Probabilistic sum A B (Algebraic sum)

• It follows commutativity, associativity, identity, and De

Morgan’s law. This operator holds also the following :
 Other Union Operations
(2) Bounded sum A  B (Bold union)

• This operator is identical to Yager function at w = 1.

• Commutativity, associativity, identity, and De Morgan’s Law
are perfected, and it has relations,

• But it does not idempotency and distributivity at absorption

Other Union Operations
 Fuzzy Intersection
• In general, intersection A  B is defined by the function I.

• The argument of this function shows possibility for element

x to be involved in both fuzzy sets A and B.
 Fuzzy Intersection
(Axioms for Intersection Function)

The most important additional requirements:

 Fuzzy Intersection
(Axioms for Intersection Function)

• Standard fuzzy intersection completes the above 6 axioms

 Fuzzy Intersection
(Axioms for Intersection Function)

Visualization of standard fuzzy intersection set

Yager’s intersection function
Yager’s intersection function
 Example of Yager’s intersection
Take Yager function at w = 1 for example. Providing a = 0.4
and b = 0.6, then
Example of Yager’s intersection
Example of Yager’s intersection
 Other Intersection Operations
1. Algebraic product A ● B (probabilistic product)

• Operator ● is obedient to rules of commutativity,

associativity, identity, and De Morgan’s law
 Other Intersection Operations
2. Bounded product A B (Bold intersection)

• Identical to Yager intersection function with w = 1

• commutativity, associativity, identity, and De Morgan’s

law hold in this operator . The following relations

• are also satisfied, but not idempotency,

distributability, and absorption
 Other Intersection Operations
3. Drastic product A B

4. Hamacher’s intersection A  B
Other Operations in Fuzzy Set
 Disjunctive Sum
• Disjunctive sum is the name of operation corresponding
“exclusive OR” logic.

Disjunctive sum of two sets

 Disjunctive Sum
1. Simple disjunctive sum
– disjunctive sum in fuzzy set is allowed just like in crisp set.
 Disjunctive Sum
1. Simple disjunctive sum (cntd..)
– Here goes procedures obtaining disjunctive sum of A and B
Example of simple disjunctive sum
 Disjunctive Sum
2. Disjoint sum
– The key idea of “exclusive OR” is elimination of
common area from the union of A and B. An operator Δ
for the exclusive OR disjoint sum as follows.

– If we reconsider the previous example, we have

Example of Disjoint Sum

Example of disjoint sum (exclusive OR sum)

 Difference
The difference in crisp set is defined as follows
 Difference in Fuzzy Set
1. Simple difference
– If we reconsider the previews example, A – B would be,
 Difference in Fuzzy Set
1. Simple difference
– If we reconsider the previews example, A – B would be,
 Difference in Fuzzy Set
2. Bounded difference
– For novice-operator , define the membership function

– bounded difference of preceeding two fuzzy sets is as

follows
 Difference in Fuzzy Set
2. Bounded difference
 Distance in Fuzzy Set

• The concept ‘distance’ is designated to describe the

difference. But it has different mathematical
measure from the ‘difference’ introduced in the
previews section
• Measures for distance in fuzzy setare defined in:
1. Hamming Distance
2. Euclidean Distance
3. Minkowski distance
 Distance in Fuzzy Set

1. Hamming Distance

– Example : Following A and B for instance,

A = {(x1, 0.4), (x2, 0.8), (x3, 1), (x4, 0)}
B = {(x1, 0.4), (x2, 0.3), (x3, 0), (x4, 0)}
– Hamming distance; d(A, B),
 Distance in Fuzzy Set

1. Hamming Distance (cont...)

– Hamming distance contains usual mathematical
senses of ‘distance’.
(1) d(A, B)  0
(2) d(A, B) = d(B, A) commutativity
(3) d(A, C)  d(A, B) + d(B, C) transitivity
(4) d(A, A) = 0
 Distance in Fuzzy Set

1. Hamming Distance (cont...)

– Assuming n elements in universal set X; i.e., |X| = n, the
relative Hamming distance is,

– might rename Hamming distance as ‘symmetrical distance’

and written as below by using operator ,

– This operator doesn’t hold distributivity. In addition,

‘disjoint sum’ using operator  introduced before may be
applied to this symmetrical distance.
 Distance in Fuzzy Set

1. Hamming Distance (cont...)

 Distance in Fuzzy Set

2. Euclidean Distance

– Euclidean distance between sets A and B used for the previous Hamming distance is

– and relative Euclidean distance is

 Distance in Fuzzy Set

3. Minkowski Distance

– Generalizing Hamming distance and Euclidean distance

results in Minkowski distance. It becomes the Hamming
distance for w = 1 while the Euclidean distance for w = 2
 Cartesian Product of Fuzzy Set
• Power of fuzzy set
– Second power of fuzzy set A is defined as follows :

– Similarly mth power of fuzzy set Am may be computed as,

– This operator is frequently applied when dealing with the

linguistic hedge in expression of fuzzy set
 Cartesian Product of Fuzzy Set
• Cartesian product
– Cartesian product applied to multiple fuzzy sets can
be defined as follows.

– the probability for n-tuple (x1,x2,...,xn) to be involved

in fuzzy set A1 x A2 x ... x An is,
t-norms and t-conorms
 t-norms and t-conorms
• There are two types of operators in fuzzy sets:
– t-norms (triangular-norm)
– t-conorms (triangular-conorm) or also called as s-
norms
 Definitions for t-norms

• Now can easily recognize that the following operators

hold conditions for t-norm.
Definitions for t-conorm (s-norm)

• There are examples of t-conorm operators

 t-norms and t-conorms
• When computing the t-norm and t-conorm, operands in
most cases are values of membership functions . An
alternative symbol for these functions is *. And in other
cases symbol * is used for t-norm, while symbol for t-
conorm,
 t-norms and t-conorms
• t-norm is sometimes called s-norm.
• In practical, sometimes could not determined which
operator is appropriate to the application  In that case,
put the symbol * as a tentative operator.
• When we know that a t-norm operator can be used
without selecting a specific operator, we put the symbol T.
• All t-norm and t-conorm functions follow these relations.
 t-norms and t-conorms
• Some operators following the above properties
 Duality of t-norms and t-conorms
• Duality exists between t-norm and t-conorm. Let function
T represent a t-norm operator. If we define T’s as

• it becomes a t-conorm. That is,

• and again for x, y  [0,1], presume complements of x and y as

• and complement of result from operation as,

• then following relations are held, and they can be apprehended by

De Morgan’s law.
Example Case 1
Example Case 1
 Example Case 1
1. The set of loadings for which one expects that either material B or
material D will be ‘‘safe.’’
2. The set of loadings for which one expects that both material B and
material D are ‘‘safe.’’
3. The set of loadings for which material D and material B are unsafe,
respectively.
4. The set of loadings for which the ductile material is safe but the brittle
material is in jeopardy.
5. The set of loadings for which the brittle material is safe but the ductile
material is in jeopardy.
6. Asserts that the loadings that are not safe with respect to both
materials
7. Asserts that the loads that are safe for neither material D nor material
B
Example Case 1
Example Case 1
Exercise
 Exercise
Continuing from the chemical engineering case described in Problem
above. Suppose the selection of an appropriate analyzer to monitor the
‘‘sales gas’’ sour gas concentration is important. This selection process
can be complicated by the fact that one type of analyzer, say A, does not
provide an average suitable pressure range but it does give a borderline
value of instrument dead time; in contrast another analyzer, say B, may
give a good value of process dead time but a poor pressure range.
Suppose for this problem we consider three analyzers: A, B and C.

represent the fuzzy set showing the pressure range suitability of

analyzers A, B, and C (a membership of 0 is not suitable, a value of 1
is excellent).
 Exercise

represent the fuzzy set showing the instrument dead time suitability of
analyzers A, B, and C (again, 0 is not suitable and 1 is excellent).

1. Show the analyzers that are not suitable for pressure range and
instrument dead time, respectively
2. Show which analyzer is most suitable in either category
3. Show which analyzer is suitable in both categories
Exercise