Fuzzy Logic

CHAPTER- 5

Fuzzy Logic
 Learn the concept of fuzzy logic and its day-to-day utility
 Explain the functions of fuzzy arithmetic and operations
 Draw the structure of a fuzzy system
 List the applications of fuzzy logic systems
 Understand MatLab Implementation of Fuzzy Logic
 Recent trends in fuzzy technology
 Fuzzy Logic

5.1 Introduction: Fuzzy Logic

 Introduced in 1965 by Lofti A. Zadeh, a professor in the University of California,

Berkeley

 Got accepted as emerging technology since 1980s

 Having a fundamental trade-off between precision and cost which can be called
“principle of incompatibility”.

 Generalizes classical two valued logic for reasoning uncertainty.

 Helps us to tackle the uncertainty and vagueness associated with the event.

5.2 Human Learning Ability, Imprecision and Uncertainty

 The factors influencing humans learning ability are:

1. Generalization
2. Association (memory mapping)
3. Information loss (memory loss or forgetfulness).
 Fuzzy Logic

5.3 Undecidability

 Ambiguity that originates from the inability to distinguish between various states of
an event is termed as Undecidability

5.4 Probability Theory versus Possibility Theory

 Possibility measures the degree of ease for a variable to take a value, while probability
measures the likelihood for a variable to take a value.

 So they deal with two different types of uncertainty.

 Possibility theory handles imprecision and probability theory handles

likelihood of occurrence.

 fuzzy set theory can define set membership as a possibility distribution.

 Also, fuzzy logic measures the degree to which an outcome belongs to an event while
probability measures the likely hood of event to occur.
 Fuzzy Logic

5.5 Classical Sets and Fuzzy Set

 „crisp set‟ is a collection of distinct (precisely defined) elements. In classical set

theory, a crisp set can be a superset containing other crisp sets.

 Fuzziness is a property of language. Its main source is the imprecision involved in

defining and using symbols.

 Fuzziness is a property of language. Its main source is the imprecision involved in

defining and using symbols.

5.5.1 Representation of a Classical Set

 Classical set as already mentioned is a collection of objects of any kind. They can be
represented as list method, rule method and characteristic function method.
 Fuzzy Logic

5.5.1 Representation of a Classical Set

Method Examples

List method N={1,2,3,...100}

Listing of objects in a set parenthesis. A={Madras, Madurai, Roorkee, Mumbai}

Rule method N={x│f(x)}

The set is defined by property or rule satisfied by N= {f (3)} = {3, 6, 9.....}
every objects within the set parenthesis. A={Densely populated city}

Characteristic function method A= {Pigeon, Horse, Trichy, 5, 6}

The set is defined by a function termed as x A implies that x is a member of set A
characteristic function denoted as μ . The x A implies that x is not a member of A.
characteristic function declares whether the element
is a member of the set or not 1 for x  A
A ( x )  
 0 for x  A
 Fuzzy Logic

5.5.2 Representation of a Fuzzy Set

 Representation of a physical variable as a continuous curve is called fuzzy set or

membership function.

 The numerical value is termed as 'Degree of membership' or 'Truth function'.

 Fuzzy Logic

5.5.2 Representation of a Fuzzy Set

 Each pair ( x, μA(x)) is called a singleton. Here, x is followed by its membership in

A, i.e. μA(x))

 Then the set A may be given by the collection or union of all singletons
A = {(1, 1.0), (2, 1.0), (3, 0.75), (4, 0.5), (5, 0.3), (6, 0.3), (7, 0.1), (8, 0.1)}

 Alternative notation, the set of small integers mentioned above can be written as:
A = A(1)/1 + A(2)/2 + A(3)/3 + A(4)/4 + A(5)/5 + A(6)/6 + A(7)/7 + A(8)/8

5.5.3 Basic Properties of Fuzzy Sets

 The complete range of the model variable is called ‘universe of discourse’.

 Universe of discourse is associated with the system variable (score) and not with a
particular fuzzy set (Good score, Average score or Bad score).

 The range of values covered by a particular fuzzy set is termed as domain of fuzzy
set.

 The domain of fuzzy set is set of elements whose degree of membership in the fuzzy
set is greater than zero.
 Fuzzy Logic

Non-convex and convex fuzzy sets

 Non-convex fuzzy sets are fuzzy sets, in which membership grade alternately
increases and decreases on the domain.

 Fuzzy sets in which membership grades do not alternately increases and decreases

are called convex fuzzy set.

Convex fuzzy set Non-convex fuzzy set

 Fuzzy Logic

Some of the commonly used fuzzy sets are

1.0

μ (x)
 0 xa
x  a
  x, a , b    a xb
0.5

b  a
 1 xb
0
a b
x

Fig. 5.7a Function fuzzy set (Linearly expressed gamma membership function)

 xa
1.0
0
μ (x)   x  a 2
 2  axb
  ca 
0.5

Sx , a , b , c   
1  2 x  c 
2

bxc
0
a b c  ca 
x

 1 xc
Fig. 5.7b Function fuzzy set (Sigmoidal membership function)
 Fuzzy Logic

Some of the commonly used fuzzy sets are

1.0
μ (x)

0.5

G( x ,a , b )  e  ba  x 
2

0
a x

Fig. 5.7c Function fuzzy set (Gaussian membership function)

1.0

μ (x)  1 xa
b  x
0.5
L  x, a , b    a xb
b  a
 0 xb
0
a b x

Fig. 5.7d Function fuzzy set (L membership function)

 Fuzzy Logic

Some of the commonly used fuzzy sets are

1.0

μ (x)
 0 xa
x  a
0.5

 axb
b  a
x , a , b , c    c x
 bxc
c  b
0
a b c
x

 0 xc
Fig. 5.7e Function fuzzy set (Triangle membership function)

1.0

 0 xa
x  a
μ (x)

0.5
b  a axb
 bxc
 x , a , b , c , d   1
d  x
0
a b c d  cxd
d  c
x

Fig. 5.7f Function fuzzy set (Trapezoidal membership function)  0 xd

 Fuzzy Logic

5.6 Fuzzy Set Operations

 Some of the important set operations that can be carried out using fuzzy sets are :
1. Intersection of fuzzy sets
2. Union of fuzzy sets
3. Complement of fuzzy sets

Intersection of Fuzzy Sets

 Intersection of two fuzzy sets contains the elements that are common to both the
sets and is also equivalent to logical AND operation.
 In fuzzy logic, this operation is performed by taking the minimum of the truth
membership functions.

 A x  B x  min  A x, B x

 Fuzzy Logic

5.6 Fuzzy Set Operations

Union of Fuzzy Sets

 Union of two fuzzy sets contains all the elements belonging to both the sets and is
equivalent to logical OR operation.
 In fuzzy logic, this operation is performed by taking the maximum of the truth
membership functions.

 A x  B x  max  A x, B x

Complement of Fuzzy Sets

 The complement of a fuzzy set consists of the elements from universe which are not
present in the set. This is equivalent to logical NOT operation.
 In fuzzy logic, this operation is performed by subtracting the membership value from 1.

 A x   1   A x 
~
 Fuzzy Logic

5.6.4 Important Terminologies in Fuzzy Set Operations

a) Empty Fuzzy Set

 A fuzzy set A is said to be an empty set if it has no members and its membership
function is zero everywhere in its universe of discourse U.

  if A(x) = 0, x U

b) Normal Fuzzy Set

 A fuzzy set A is said to be normal if it has a membership function that includes at

least one singleton equal to unity in its universe of discourse U
A (xo) = 1
 Fuzzy Logic

5.6.4 Important Terminologies in Fuzzy Set Operations

c) Equal Fuzzy Sets

 If the membership functions of any two fuzzy sets A and B are equal everywhere in
the universe of discourse, then the two fuzzy sets are said to be equal thereby

satisfying the equation. A  B if A (x) = B (x)

d) Fuzzy Set Support

 A fuzzy set A is supported only if the crisp set of all x  U such that the membership

values are non-zero. A(x) > 0

d) Fuzzy Set Support

 The product of the fuzzy set A and B produces a new fuzzy set, with its membership
function value equal to the algebraic product of the membership function of A and B.
A.B (x)  A (x). B (x)
 Fuzzy Logic

5.6.4 Important Terminologies in Fuzzy Set Operations

f) Fuzzy Set Multiplication by a Crisp Number

 The membership function of a fuzzy set A is multiplied by the crisp number „b‟ to
obtain a new fuzzy set whose membership function bA (x).

b A (x)  b . A (x)

g) Power of a Fuzzy Set

 If  is set as the power of a fuzzy set A, then a new fuzzy set A  has a membership
function value

 A x    A x 
 Fuzzy Logic

5.6.5 Properties of Fuzzy Sets

Part A:

( A)  A
Double negation law
A A  A A A  A
Idempotency

Commutativity
A B  B  A A B  B  A

Associative Property ( A  B)  C  A  ( B  C ) ( A  B)  C  A  ( B  C )

Distributive Property A  ( B  C )  ( A  B)  ( A  C ) A  ( B  C )  ( A  B)  ( A  C )

Adsorption
A  ( A  B)  A A  ( A  B)  A
De Morgan‟s laws
A  B  A B A  B  A B
 Fuzzy Logic

5.6.5 Properties of Fuzzy Sets

Part B :

Law of Contradiction A B  

Law of Excluded Middle A A  U

Intersection of a Fuzzy Set with an Empty Set A    A ( x)  0  0

Union of a Fuzzy Set with an Empty Set A  A A ( x )  0  A ( x )

Intersection of a Fuzzy Set with Universe of

AU  A  A (x)  1   A (x)
Discourse

Union of a Fuzzy Set with Universe of Discourse AU  U  A (x)  1  1

Some of the properties that are not valid for fuzzy sets but valid for crisp sets are
 Fuzzy Logic

5.6.6 Fuzzy Arithmetics

1. Extension principle (discrete fuzzy set)

 The extension principle is a mathematical tool for extending crisp mathematical

notions and operations to the milieu of fuzziness. The extension principle suggests that
fuzzifying the parameters of a function yields fuzzy outputs.

(a) Addition c ( z)  V  A ( x)   B ( y)

z  x y

(b) Subtraction A  B (z)  V A (x)  B ( y)

zx y

(c) Multiplication  A.B ( z)  V  A ( x)   B ( y)

z  x. y

(d) Division  AB ( z)  V  A ( x)   B ( y)

z  x y
 Fuzzy Logic

5.6.6 Fuzzy Arithmetics

2. Alpha Cut Method (Discrete Fuzzy Set)

 The alpha cut method for discrete fuzzy set is carried out by performing arithmetic
operations within the boundary elements present between fuzzy sets and the solutions
that depend on the boundary values of the discrete fuzzy function

(a) Addition : A + B = [a1(),a2()] + [b1(),b2()], A + B = [a1()+b1(), a2()+b2()]

(b) Subtraction : A - B = [a1(), a2()] - [b1(), b2()], A - B = [a1() - b2(), a2() - b1()]

(c) Multiplication: A.B  [a1(),a2()] . [b1(),b1()], A.B = [a1() .b1(), a ()

2 .b2()]

 a1(  ) a (2 ) 
(d) Division: A B  a ( )
1 ,a ( )
2   b
( )
1
( )
,b
2  A  B   () , () 
 b 2 b1 
 Fuzzy Logic

5.6.6 Fuzzy Arithmetics

3. Alpha Cut Method (Continuous Fuzzy Set)

 The alpha cut method for discrete fuzzy set is carried out by performing arithmetic
operations within the boundary elements present between fuzzy sets and the solutions
that depend on the boundary values of the discrete fuzzy function

(a) Addition : C = A + B = [a1()+ b1(), a2()+ b2() ]

(b) Subtraction : C = A - B = [ a1() - b2(), a2() - b1() ]

(c) Multiplication: C = A1 . B1 = [a1() . b1() , a2() . b2() ]

 a1() a (2 ) 
C  A  B   () , () 
(d) Division:
 b2 b1 
 Fuzzy Logic

5.6.6 Fuzzy Arithmetics

4. Higher Order Arithmetic Operations

 If the membership functions are non-linear or for higher order arithmetic operations,
it is not easy to carry out arithmetic operations.

 In the first step, it has been assumed that the membership functions of the variables
are at least linear and if not, then it has to be expressed in terms of approximate linear
functions.

 In case of multiplication and division mathematical complexity i.e., the order of  in

a1() is proportional to the order of multiplication and division. Hence, generation of
suitable membership function became difficult.
 Fuzzy Logic

5.7 Fuzzy Relations

 A fuzzy relation generalizes the aim of binary classical relation into one that allows
partial membership. It is developed by allowing relationship between elements of two
or more sets to take an infinite number of degrees of relation between extreme of
"Completely Related" and "Non-Related ". It converts relation into matter of degree.

5.7.1 Operations on Fuzzy Relations

(a) Union  R S ( x, y)  max(  R ( x, y),  S ( x, y))

(b)Intersection  R S ( x, y)  min(  R ( x, y),  S ( x, y))

(c) Complement  R ( x, y)  1   R ( x, y)
(d) Containment R  S   R ( x, y)   S ( x , y)
 Fuzzy Logic

5.8 Fuzzy Composition

 Fuzzy compositions can combine fuzzy relations defined on different Cartesian

products with each other in a number of different ways. A more complex form of relation
is called the „composition of fuzzy relation‟, where the associations among fuzzy sets are
of concern in addition to associations among the elements of each fuzzy set.

Max-Min Composition:

R1 o R 2  
xxz

V  R1 (x, y)   R 2 ( y, z) / (x, z)
y
 R i o R2
y

(x, z)  V R1 (x, y)  R 2 ( y, z) 
Max-Star Composition:
R1 * R 2   V  R1 
( x , y ) *  R 2 ( y, z ) / ( x , z )  R1*R 2 ( x, z)  V  R ( x, y) *  R ( y, z)
1 2
y y
xxz

Max-Product Composition:

R1 . R 2   V  y
R1 
(x, y). R 2 ( y, z) / (x, y, )
y

R1 .R 2 (x, z)  V R1 (x, y) . R 2 ( y, z) 
X Z

Max-Average Composition:

1 
R1  R 2  V
2
 R 1 (x, y)  R 2 ( y, z)  / (x, z)

1
 

 R 1   R 2 ( x, z)  V   R 1 ( x , y)   R 2 ( y, z) 
2 
y
X Z y
 Fuzzy Logic

5.9 Natural Language and Fuzzy Interpretations

 A collection of these primary terms will form phrases, of our natural language.
Examples of some atomic terms are slow, medium, young, beautiful, angry, cold,
temperature etc.
 The collection of atomic terms can form compound terms. Examples of compound terms
are very cold, medium speed, young lady, fairly beautiful picture, etc .
Membership Functio n M odel fo r Atomic Values
1

0.9

0.8
Membership Value of Temperature

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0
0 10 20 30 40 50 60 70 80 90 100 110 120
Temperature(T) in Degrees
 Fuzzy Logic

5.9.1 Linguistic Modifiers

Primary Terms with

Modification Membership Function
Modifiers
[  ( x )] 2
Very  2
I
x
[  ( x )] 4
Very very 4  I
x 1
More or less 1 [  ( x )] 2
Slightly  2 
I
x

[  ( x )]1.25
Plus 1.25 I x
[  ( x )] 0.75
Minus  0.75 I x
 (x)
1  , x  x max 1   , x  x min
Over I
x
0, x  x max 0, x  x min
  (x)
1  , x  x min 1  , x  x max
Under I
x
0, x  x min 0, x  x max

  (x)    (x) 
2

2    , 0   0.5
2 , 0    0.5 x 
2
I 
x  I 
Indeed 1  2[1  ] , 2
0.5    1.0    ( x ) 
2
   (x) 
1  2 1       , 0.5    x   1.0
 I  x   I  
 Fuzzy Logic

5.9.1 Linguistic Modifiers

1. Fuzzy Concentration

 Concentrations tend to concentrate or reduces the fuzziness of the elements in a fuzzy

set by reducing the degree of membership of all elements that are only „„partly‟‟ in the set.
Concentration
1.2

1
 A ( Low , T ) 
1

1  0.0005(T  30) 2.4

0.8

Low Before
2
 
Concentration
Membership Value

1
 A ( Very Low , T )  
0.6

2.4 
1  0.0005(T  30)  0.4

0.2 After
Concentration
Very
Low
0

-0.2
0 20 40 60 80 100 120
Temperature(T) in Degrees
 Fuzzy Logic

5.9.1 Linguistic Modifiers

2. Fuzzy Dilation
 Dilation tend to dilate or increase the fuzziness of the elements in a fuzzy set by
increasing the degree of membership of all elements that are only „„partly‟‟ in the set.
Dilation
1

0.9

1
 A ( Low , T )  0.8 After

1  0.0005(T  30) 2.4

Dilation

0.7

Low

Membership Value
0.6
0. 5
 1  Slightly
 A (Slightly Low , T )   2. 4 
0.5 Low

1  0.0005(T  30) 
0.4 Before
Dilation

0.3

0.2

0.1

0
0 20 40 60 80 100 120
Temperature(T) in Degrees
 Fuzzy Logic

5.9.1 Linguistic Modifiers

3. Fuzzy Intensification

 This property of the linguistic modifier is found to have a property which is a

combination of both fuzzy concentration and fuzzy intensification.
Intensification
1

0.9
Befo re
Intensificatio n
1
 A ( Low , T )  0.8

1  0.0005(T  30) 2.4

Low

0.7 After
Intensificatio n

Membership Value
0.6
Indeed
  1 
2
 1  Low
 2 2.0 
, 0 2.0 
 0.5 0.5


 A ( Indeed Low , T )   1  0.005(T  30)  1  0.005(T  30) 
1  2   1 
2
 1  0.4
, 0.5   2.0 
 1.0
 1  0.005(T  30) 2.0  1  0.005(T  30) 
   0.3

0.2

0.1

0
0 20 40 60 80 100 120
Temperature(T) in Degrees
 Fuzzy Logic

5.9.2 Logical Operations using Linguistic Modifiers

 It is possible to perform logical operations using the linguistic modifiers, however

the order of precedence need to be followed to avoid confusion. This problem can be
resolved by properly using the parentheses by the rule that use “association to the
right”.

 The precedence of operation should be “NOT” in the First, “AND” in the second and
“OR” in the third.

 For example, if we have a compound terms “plus very minus very low”, then
following the rule “association to the right” while using the parentheses we get “Plus
(very (minus (very (low))))”.
 Fuzzy Logic

5.10 Structure of Fuzzy Inference System

 Fuzzy Logic

5.10.1 Fuzzification

 The process of converting the crisp values to fuzzy variable is called „fuzzification‟.
 Fuzzy Logic

5.10.2 Fuzzy Propositions

 A sentence having only “true (1)” and “false (0)” as its truth value is called propositions
which belongs to classical proportional logic.

 Fuzzy propositions can be classified into 4 types. They are:

(a) Unconditional and Unqualified Propositions:
The canonical form of fuzzy propositions of this type, P, is expressed by the sentence
P : V is F, E.g., Temperature of 45o is high
(b) Unconditional and Qualified Propositions:
Propositions P of this type are characterized by the canonical form
P : V is F is S, E.g., Temperature of 65o is high is very true
(c) Conditional and Unqualified Propositions:
Propositions P of this type are expressed by the canonical form
P : If X is A, then Y is B,
E.g. If the temperature is high, then it is hot
(d) Conditional and Qualified Propositions:
Propositions P of this type can be characterized by the canonical form
P : If X is A, then Y is B is S, E.g. If the temperature is high, then it is hot is true.
 Fuzzy Logic

Fuzzy connectives

 The linguistic variables are combined by various connectives such as negation,

disjunction, conjunction and implication etc

Symbol Connective Usage Definition

- Negation 1  T ( P)
P

 Disjunction PQ max( T( P), T( PP)

PQ min( T ( P), T (Q))

 Conjunction

 Implication PQ P  Q  max( 1  T( P), T(Q))

 Fuzzy Logic

5.10.3 Fuzzy Implication Relations

Implication Operator  R ( x, y)  ( A (x),  B ( x)) 

Usage Interpretation of “else”
( Ψ)

Zadeh Max-Min ( A (x)   B ( y))  (1   A (x)) AND( )

Mamdani Min  A (x)   B ( y)

OR( )

Larsen Product  A ( x )   B ( y) OR( )

Arithmetic 1  (1   A ( x)   B ( y)) AND( )

Boolean (1   A ( x))   B ( y)) AND( )

Bounded Product 0  ( A (x)   B ( y)  1) OR( )

 A ( x ), if  B ( y)  1
Drastic Product  B ( x ), if  A ( y)  1
OR( )
0, if  A ( y)  1,  B ( y)  1

1, if  A ( x )   B ( y)
Standard Sequence
0, if  A ( x )   B ( y) AND( )

1, if  A ( x )   B ( y)
Gougen  B ( y) AND( )
, if  A ( x )   B ( y)
 A (x)

Godelian 1, if  A ( x )   B ( y)
AND( )
 B ( y), if  A ( x )   B ( y)
 Fuzzy Logic

5.10.4 Fuzzy Inference Procedures

1. Generalized Modus Ponens (GMP)

 GMP means “the way that affirms by affirming”.

 GMP is generally stated as IF x is A THEN y is B

x is A ' ( Analytical ly Known )


y is B ' ( Analytical ly Unknown )

2. Generalized Modus Tollens (GMT)

 GMT means “the way that denies by denying”.

 GMT is generally stated as IF x is A THEN y is B

y is B' ' ( Analytical ly Known )


x is A' ' ( Analytical ly Unknown )
 Fuzzy Logic

5.10.5 Fuzzy Inference Algorithms

 Fuzzy inference algorithms are procedures that perform evaluation of fuzzy if/then rules
If x is A1 then y is B1 ELSE
If x is A2 then y is B2 ELSE
……
if x is An then y is Bn

Case I: IF/THEN rules with connective AND/OR

 If x1 is A1 AND x2 is A2 AND…..AND xn is An then y is B, can be represented by min or

product operator
min() preposition -(x1,x2,……,xn,y) =  [A1(x1)  A2(x2) ……  An(xn), B(y)] or
product (.) preposition - (x1,x2,……,xn,y) = [A1(x1). A2(x2) …… An(xn), B(y)]
 Fuzzy Logic

5.10.5 Fuzzy Inference Algorithms

Case II: Nested IF/THEN Rules

 If x1 is A1 then (if x2 is A2 then….. (if xn is An then y is B)….) can be represented by

repeated application of an implication operator
(x1,x2,……,xn,y) =  [A1(x1),  [A2(x2) ……  [An(xn), B(y)]]

Case III: IF/THEN rules with connective ELSE

 If x1 is A1p AND x2 is A2p AND…..AND x1n is Anp then y is Bq ELSE, can be analytically
described by an algorithmic relation of the form

R(x1i, x2i,……..xni,yi) =   ( x1i , x 2i ,....., x ni , y j ) /( x1i , x 2i ,....., x ni , y j

( x1 ,x 2 ,......,x ni,y j )
 Fuzzy Logic

5.10.6 Defuzzification

 Defuzzification is the process of converting the fuzzy set into a crisp value.

(a) Max-membership (Height method)

(b) Centre of Area (COA) or Center of Gravity or Centroid Method

(c) Weighted Average Method

(d) Mean of Maxima (Middle of maxima)

(e) Centre of Sums

(f) Centre of Largest Area (COLA)

(g) First of Maxima

(h) Last of Maxima

 Fuzzy Logic

5.10.6 Defuzzification

a) Max-membership (Height method) (b) Centre of Area (COA) Defuzzification

 OUT ( u i ) N
max( (u i ))  1
1.
u 
i 1
i OUT (u i )
1.
Peak 0
0.
u*  N


0 point
0. 9
0. OUT (u i )
(u i )
9 8 i 1
0. 0.
8 7
0. 0.
7 6
0. 0.
6 5
0. 0.
5 4
0. 0.
4 3
0. 0.
3 2
0. 0.
2 1
0.
1
0
1 2 3 4 5 6 7 8 9 10 1 12 13 14 15 16
0 1
u *  6.65 ui
u* ui
 Fuzzy Logic

5.10.6 Defuzzification

(c) Weighted Average Method Z

z
j1
j  OUT ( z j )
(d) Mean of Maxima (MOM)
u 
*
Z

 OUT ( u i )
z2
 OUT (z j )
j1
1 M

z1
1.0 u 
*
um
0.9 M m 1
0.8

0.7

0.6 (e) Centre of Sums (COS)

0.5

0.4
N n
0.3

0.2
u 
i 1
i
k 1
 C' ( u i )
k

0.1
u 
*
N n

0

i 1 k 1
 C' ( u i )
k

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 ui
u *  5.86
 Fuzzy Logic

5.10.6 Defuzzification

(f) Centre of Largest Area (COLA)

(g) First of Maxima

 OUT ( u i )
N max(  C m ( u ))  max(  C i ( u )), i  1....N C
1.
0 C1
C2
 u i  Cm (u i )
i 1
0.
9 u*  N u *  min( u i )

0.
8
0. Cm (u i )
i 1
7
0. (h) Last of Maxima
6
Cm = C1
0.
5
0.
4
0. u *  max( u i )
3
0.
2
0.
1

0
1 2 3 4
ui
5 6 7 8 9 10 11 12 13 14 15 16

u*= 4.06 u*= min(Ui)=9 u*= max(Ui)=10

 Fuzzy Logic

Assessment of defuzzification methods

 The ideal defuzzification technique should satisfy the following 5 criteria .

(a) Continuity

(b) Disambiguity

(c) Plausibility

(d) Computational Complexity

(e)Weighting Methods

 It should be noted that none of the defuzzification techniques satisfy all the 5 criteria.

 However, the right choice of the defuzzification techniques for a particular application
can be selected only by understanding the trade-off between these criteria.