3.

Crisp and Fuzzy Relations

2012 1
3.1. Introduction
A relation is of fundamental importance in all engineering fields.
Relations can be also be used to represent similarity. Relations are
involved in logic, classification, pattern recognition, and control
Some relations concern elements within the same universe: one
measurement is larger than another, one event occurred earlier than
another, one element resembles another, etc.
Other relations concern elements from disjoint universes: the
measurement is large and its rate of change is positive, the x-
coordinate is large and the y-coordinate is small, for example.
These examples are relationships between two objects, but in
principle we can have relationships which hold for any number of
objects.

2012 2
 3.2. Crisp Relation
A Crisp Relation R from a set A to a set B assigns to each
ordered pair exactly one of the following statements:
(i)’’a is related to b’’ or (ii) ’’a is not related to b’’

The Cartesian Product AxB is the set of all possible

combinations of the items of A and B. For example when:
A = {a1,a2,a3} and B = {b1,b2}

The Cartesian product yields

the shown figure.

Which means:
AxB = {(a1,b1),(a1,b2),(a2,b1),(a2,b2),(a3,b1),(a3,b2)}
2012 3
 example 1: “Owning Cars”-Crisp Relation
X = {Aly, Baher, Kamel}
Y = {BMW, Chrysler, Ford, Mazda, Fiat}

BMW Chrysler Ford Mazda Fiat

Aly 1 0 0 0 1
Baher 1 0 1 1 0
Kamel 0 1 0 0 0

2012 4
 example 2: “Close to”-Crisp Relation
X = Y = {1, .......,8}

1 2 3 4 5 6 7 8
1 1 1 0 0 0 0 0 0
2 1 1 1 0 0 0 0 0
3 0 1 1 1 0 0 0 0
4 0 0 1 1 1 0 0 0
5 0 0 0 1 1 1 0 0
6 0 0 0 0 1 1 1 0
7 0 0 0 0 0 1 1 1
8 0 0 0 0 0 0 1 1

2012 5
3.3. Fuzzy Relation:
Fuzzy relations map elements of one universe, say U, to those of
another universe, say V, through the Cartesian product of the two
universes. However, the ‘‘strength’’ of the relation between ordered
pairs of the two universes is measured with a membership function
expressing various ‘‘degrees’’ of strength of the relation on the unit
interval [0,1].
As an example a fuzzy relation “Friend” describes the degree of friendship
between two persons (in contrast to either being friend or not being friend in
classical relation!)

2012 6
example 3: Fuzzy relation “Similarity” U = V = {1, . . . , 8}

1 2 3 4 5 6 7 8
1 1 0.5 0 0 0 0 0 0
2 0.5 1 0.5 0 0 0 0 0
3 0 0.5 1 0.5 0 0 0 0
4 0 0 0.5 1 0.5 0 0 0
5 0 0 0 0.5 1 0.5 0 0
6 0 0 0 0 0.5 1 0.5 0
7 0 0 0 0 0 0.5 1 0.5
8 0 0 0 0 0 0 0.5 1

2012 7
example 4: Let A be a fuzzy set defined on a universe of three discrete
emperatures, X = {x1,x2,x3}, and B be a fuzzy set defined on a universe
of two discrete pressures, Y = {y1,y2} Fuzzy set A represents the
“ambient” temperature and fuzzy set B the “near optimum” pressure for
a certain heat exchanger, and the Cartesian product might represent
the conditions (temperature-pressure pairs) of the exchanger that are
associated with “efficient” operations. For example, let:

A = 0.2/x1 + 0.5/x2 + 1/x3 and B = 0.3/y1 + 0.9/y2

y1 y2
Then AxB = R = x1 0.2 0.2
x2 0.3 0.5
x3 0.3 0.9

2012 8
 Because the fuzzy relation is one kind of fuzzy sets. Therefore we
can apply operations of fuzzy set to the relations (e.g. Union,
Intersection, Complement,..).

example 4:For the given 2 fuzzy relations:

We get

and

2012 9
 Also -cut can be calculated, but the result is a crisp
relation. So for example

2012 10
 3.4. Composition of Fuzzy Relations:
A fuzzy relation R is defined on sets A, B and another
fuzzy relations S is defined on sets B,C.
That is, R ⊆ A x B, S ⊆ B x C.
The composition S • R = SR of the two relations R and S
expresses the relation from A to C.
This composition is defined by an inner product. The inner
product is similar to an ordinary matrix (dot) product,
except Multiplication is replaced by Minimum and
Summation by Maximum. Thus this composition is defined by
the following

µ S•R (a, c) = max [ min (µR (a, b), µS (b, c))]

2012 Hany Selim 11

 example:
Consider the fuzzy sets A, B and C to represent sets of
events.
The relation R ⊆ AxB, gives the possibility of occurrence of
B after A, and the relation S ⊆ BxC gives the possibility of
occurrence of C after B.

R b1 b2 b3 b4 S c1 c2 c3
a1 0.1 0.2 0.0 1.0 b1 0.9 0.0 0.3
a2 0.3 0.3 0.0 0.2 b2 0.2 1.0 0.8
a3 0.8 0.9 1.0 0.4 b3 0.8 0.0 0.7
b4 0.4 0.2 0.3

For example, by the relation R, the possibility of b1 to occur after a1

is 0.1. And by the relation S, the possibility of occurrence of c1 after
b1 is 0.9.
2012 12
Now we want to get the possibility of occurrence of C when A has occurred.
So our main job now will be the obtaining the composition S • R ⊆ AxC.
The following matrix M S • R represents this composition and is obtained
from the product of MR and MS..

R b1 b2 b3 b4 S c1 c2 c3
a1 0.1 0.2 0.0 1.0 b1 0.9 0.0 0.3
a2 0.3 0.3 0.0 0.2 b2 0.2 1.0 0.8
a3 0.8 0.9 1.0 0.4 b3 0.8 0.0 0.7
b4 0.4 0.2 0.3
S•R c1 c2 c3
a1 0.4 0.2 0.3
a2 0.3 0.3 0.3
a3 0.8 0.9 0.8

and it is also depicted in the following figure:

2012 13
 So, the possibility of occurrence of C when A has occurred
has been induced from the composition rule S • R . This
manner is named as an “inference” which is a process
producing new information.

2012 14
Application: Computer Engineering
In computer engineering, different logic families are often compared on the
basis of their power-delay product. Consider the fuzzy set F of logic
families, the fuzzy set D of delay times (ns), and the fuzzy set P of power
dissipations (mw).
If F = {NMOS,CMOS,TTL,ECL,JJ}, D = {0.1,1,10,100},
P = {0.01,0.1,1,10,100}
Suppose R1 = D x F and R2 = F x P
then we can compute
R3= R1 o R2 R3= R1 o R2

2012 Hany Selim 15

 There is a very interesting physical analogy for the max–min
composition operator, the figure below illustrates several chains
placed together in parallel. If we were to take one of these chains
out of the system, we would find that the chain would break at its
weakest link. Hence, the strength of one chain is equal to the
strength of its weakest link; in other words, the minimum (∩)
strength of all the links in the chain governs the strength of the
overall chain.
Now, if we take the entire chain system, we would find that the
weaker chains would break at first until the strongest chain was
left alone; in other words, the maximum (∪) strength of all the
chains in the chain system would govern the overall strength of
the chain system. Each chain in the system is analogous to the
min operation in the max–min composition, and the overall chain
system strength is analogous to the max operation in the max–
min composition.

2012 16
 Assume

Which can be represented by the graph

2012 17
from which

Ending with the result

2012 18
 3.5. Properties of Fuzzy Relations:
the properties of commutativity, associativity, distributivity,
involution, idempotency and De Morgan’s principles all hold for
fuzzy relations. But since there is overlap between a relation and
its complement:

Where O= the null relation, and E=the complete relation e.g.

2012 19
 3.5. Fuzzy Tolerance and Equivalence Relations:
Let R be the fuzzy relation defined on the set of cities and
representing the concept very near. We may assume that a city is
certainly (i.e., to a degree of 1) very near to itself. The relation is
therefore reflexive. Furthermore, if city A is very near to city B,
then B is certainly very near to A. Therefore, the relation is also
symmetric. Finally, if city A is very near to city B to some
degree, say .7, and city B is very near to city C to some degree,
say .8, it is possible (although not necessary) that city A is very
near to city C to a smaller degree, say 0.5. Therefore, the relation
is nontransitive.
A fuzzy relation is a fuzzy equivalence relation if all three of the
following properties for matrix relations define it:
Reflexivity μR(xi, yi) = 1
Symmetry μR(xi, yj ) = μR(xj, yi)
Transitivity μR(xi, yj ) = λ1 and μR(xj, yk) = λ2
→ μR(xi, yk) =λ where λ ≥ min[λ1, λ2].
2012 Hany Selim 20
 Fuzzy Tolerance and Equivalence Relations (Contd.)
Tolerance relation (Aehnlichkeitsrelation), has only the properties of
reflexivity and symmetry.
A tolerance relation, R, can be reformed into an equivalence relation by at
most (n − 1) compositions with itself, where n is is the number of rows
or columns of R.

Example:
Consider the relation

is reflexive and symmetric. However, it is not transitive, e.g.,

μR(x1, y2) = 0.8, and μR(x2, y5) = 0.9 ≥ 0.8
but μR(x1, y5) = 0.2 ≤ min(0.8, 0.9)
2012 Hany Selim 21
 One composition results in the following relation:

where transitivity still does not result; for example,

μR2(x1, y2) = 0.8 ≥ 0.5 , and μR2(x2, y4) = 0.5
But μR2(x1, y4) = 0.2 ≤ min(0.8, 0.5)
Finally, after one or two more compositions, transitivity results:

R3(x1, y2) = 0.8 , and R3(x2, y4) = 0.5 , and R3(x1, y4) = 0.5 ≥ min(0.8, 0.5)
2012 22