Fuzzy Sets and Systems 161 (2010) 1406 – 1419

www.elsevier.com/locate/fss

On the distributivity of fuzzy implications over continuous and

Archimedean triangular conorms
Michał Baczyński
Institute of Mathematics, University of Silesia, 40-007 Katowice, ul. Bankowa 14, Poland

Received 14 March 2009; received in revised form 29 November 2009; accepted 5 December 2009
Available online 16 December 2009

Abstract
Recently, we have examined the solutions of the following distributive functional equation I (x, S1 (y, z)) = S2 (I (x, y), I (x, z)),
when S1 , S2 are either both strict or nilpotent t-conorms and I is an unknown function. In particular, between these solutions, we
have presented functions which are fuzzy implications. In this paper we continue these investigations for the situation when S1 , S2
are continuous and Archimedean t-conorms, i.e., we consider in detail the situation when S1 is a strict t-conorm and S2 is a nilpotent
t-conorm and vice versa. Towards this end, we firstly present solutions of two functional equations related to the additive Cauchy
functional equation. Using obtained results we show that the above distributive equation does not hold when S1 , S2 are continuous
and Archimedean t-conorms and I is a continuous fuzzy implication. Further, we present the solutions I which are non-continuous
fuzzy implications. Obtained results are not only theoretical but also useful for the practical problems, since such equations have an
important role to play in efficient inferencing in approximate reasoning, especially in fuzzy control systems.
© 2009 Elsevier B.V. All rights reserved.

MSC: 03B52; 03E72; 39B52

Keywords: Fuzzy connectives; Fuzzy implication; Functional equations; t-Conorm

1. Introduction

Distributivity of fuzzy implications over different fuzzy logic connectives, like t-norms, t-conorms and uninorms,
has been studied in the recent past by many authors (see [2,3,6,9,26,27,30]). The importance of such equations has
been introduced by Combs and Andrews [13], wherein they exploit the following classical tautology:

( p ∧ q) → r ≡ ( p → r ) ∨ (q → r )

in their inference mechanism towards reduction in the complexity of fuzzy “IF–THEN” rules. Subsequently, there
were many discussions [14,15,18,25], most of them pointing out the need for a theoretical investigation required for
employing such equations, as concluded by Dick and Kandel [18], “Future work on this issue will require an examination
of the properties of various combinations of fuzzy unions, intersections and implications” or by Mendel and Liang [25],
“We think that what this all means is that we have to look past the mathematics of IRC ⇔ URC and inquire whether
what we are doing when we replace IRC by URC makes sense.”

E-mail address: michal.baczynski@us.edu.pl.

0165-0114/$ - see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.fss.2009.12.001
 M. Baczyński / Fuzzy Sets and Systems 161 (2010) 1406 – 1419 1407

Recently (see [6]), we have examined the solutions of the following distributive functional equation

I (x, S1 (y, z)) = S2 (I (x, y), I (x, z)), x, y, z ∈ [0, 1], (1)

when S1 , S2 are either both strict or nilpotent t-conorms and I is any function. The above equation is a generalization
of the following classical tautology:

p → (q ∨ r ) ≡ ( p → q) ∨ ( p → r ),

so between all solutions we have indicated fuzzy implications. In this paper we continue these investigations for
continuous and Archimedean t-conorms S1 , S2 . We believe that obtained results are not theoretical—as we mentioned
earlier, such distributive equations have an important role to play in inference invariant rule reduction in fuzzy inference
systems (see also [7,8,29]). It is also worth to note that such theoretical developments connected with solutions of
different functional equations can be also useful in other topics like fuzzy mathematical morphology (see [16,17]) or
similarity measures (cf. [12]).
The paper is organized as follows. In Section 2 we present some results concerning basic fuzzy logic connectives.
In Section 3 we show some new and important results connected with functional equations that will be employed
extensively in the sequel. In next four sections we study Eq. (1), when I is a binary operation on [0,1] (in particular
fuzzy implication), while S1 , S2 are both continuous and Archimedean t-conorms.

2. Basic notations

Firstly we recall some basic notations and results that will be useful in the sequel. If F is an associative binary
operation on [a, b] with the neutral element e, then the power notation x F[n] , where x ∈ [a, b] and n ∈ N0 , is defined by

e if n = 0,
x F[n] :=
F(x, x F[n−1] ) if n > 0.

Definition 2.1 (see Schweizer and Sklar [28], Klement et al. [22]). An associative, commutative and increasing oper-
ation S : [0, 1]2 → [0, 1] is called a t-conorm if it has the neutral element 0.

Definition 2.2 (cf. Klement et al. [22, Definitions 2.9, 2.13]). A t-conorm S is said to be
(i) Archimedean, if for every x, y ∈ (0, 1) there is an n ∈ N such that x S[n] > y,
(ii) strict, if S is continuous and strictly monotone, i.e., S(x, y) < S(x, z) whenever x < 1 and y < z,
(iii) nilpotent, if S is continuous and if for each x ∈ (0, 1) there exists n ∈ N such that x S[n] = 1.

Remark 2.3.
(i) For a continuous t-conorm S the Archimedean property is given by the following simpler condition S(x, x) > x,
for x ∈ (0, 1) (cf. [20, Proposition 5.1.2]).
(ii) If a t-conorm S is strict or nilpotent, then it is Archimedean. Conversely, every continuous and Archimedean
t-conorm is either strict or nilpotent (cf. [22, Theorem 2.18]).

We shall use the following characterizations of continuous and Archimedean t-conorms.

Theorem 2.4 (Ling [24], cf. Klement et al. [22, Corollary 5.5]). For a function S : [0, 1]2 → [0, 1] the following
statements are equivalent:
(i) S is a continuous and Archimedean t-conorm.
(ii) S has a continuous additive generator, i.e., there exists a continuous, strictly increasing function s : [0, 1] → [0, ∞]
with s(0) = 0, which is uniquely determined up to a positive multiplicative constant, such that

S(x, y) = s −1 (min(s(x) + s(y), s(1))), x, y ∈ [0, 1]. (2)

1408 M. Baczyński / Fuzzy Sets and Systems 161 (2010) 1406 – 1419

Remark 2.5 (see Klement et al. [22, Section 3.2]). (i) S is a strict t-conorm if and only if each continuous additive
generator s of S satisfies s(1) = ∞.
(ii) S is a nilpotent t-conorm if and only if each continuous additive generator s of S satisfies s(1) < ∞.

In this paper we are interested in finding the solutions of the functional equation (1) for continuous and Archimedean
t-conorms, so observe that by Remark 2.3(ii) it is enough to consider the following four cases:
• both t-conorms S1 , S2 are strict,
• both t-conorms S1 , S2 are nilpotent,
• t-conorm S1 is strict and t-conorm S2 is nilpotent,
• t-conorm S1 is nilpotent and t-conorm S2 is strict.
According to results obtained in [6], it rests to solve our main equation (1) for the last two cases.
Finally, let us recall the definition of a fuzzy implication. In the literature we can find several diverse definitions of
fuzzy implications (see [19,11]). In this article we will use the following one, which is equivalent to the definition used
by Fodor and Roubens [19, Definition 1.15] (see also Kitainik [21, p. 50]).

Definition 2.6. A function I : [0, 1]2 → [0, 1] is called a fuzzy implication if it satisfies the following conditions:

I is decreasing in the first variable, (I1)

I is increasing in the second variable, (I2)
I (0, 0) = 1, I (1, 1) = 1, I (1, 0) = 0. (I3)

From the above definition we can deduce that for each fuzzy implication I (0, x) = I (x, 1) = 1 for x ∈ [0, 1]. Thus,
I satisfies also the normality condition I (0, 1) = 1, and, consequently, every fuzzy implication restricted to the set
{0, 1}2 coincides with the classical implication.

3. Some new results pertaining to functional equations

In this section we present some results related to the additive Cauchy functional equation

f (x + y) = f (x) + f (y). (3)

First results are well-known in the literature.

Theorem 3.1 (Aczél [1], cf. Kuczma [23, Theorem 5.2.1]). For a continuous function f : R → R the following state-
ments are equivalent:
(i) f satisfies the additive Cauchy functional equation (3) for all x, y ∈ R.
(ii) There exists a unique constant c ∈ R such that f (x) = cx for all x ∈ R.

Theorem 3.2 (Kuczma [23, Theorem 13.5.3]). Let A ⊂ R be an interval such that 0 ∈ cl A, where cl A denotes the
closure of the set A and let B = A + A = {a1 + a2 |a1 ∈ A, a2 ∈ A}. If a function f : B → R satisfies the additive
Cauchy functional equation (3) for all x, y ∈ A, then f can be uniquely extended onto R to an additive function g such
that g(x) = f (x) for all x ∈ B.

Theorem 3.3 (Kuczma [23, Theorem 13.6.2]). Fix a real a > 0. Let A = [0, a], and let H ⊂ R2 be the set

H = {(x, y) ∈ R2 |x ∈ A, y ∈ A and x + y ∈ A}. (4)

If f : A → R is a function satisfying equation (3) on H, then there exists a unique additive function g : R → R such
that g(x) = f (x) for all x ∈ A. Moreover, the closed interval [0, a] may be replaced by any one of these intervals
(0, a), [0, a) and (0, a].
 M. Baczyński / Fuzzy Sets and Systems 161 (2010) 1406 – 1419 1409

Theorem of Bernstein–Doetsch (Bernstein and Doetsch [10], see also Kuczma [23, Corollary 6.4.1]). Let D ⊂ Rn
be a convex and open set, and let f : D → R be a convex function, i.e., it satisfies the Jensen’s functional inequality
 
x+y f (x) + f (y)
f ≤ , x, y ∈ D.
2 2
If f is bounded above on a non-empty open set U ⊂ D, then it is continuous in D.
Since every additive function is convex as well as concave (observe that the converse is not true), the above result
can be used for the additive functions, i.e., functions which satisfy the additive Cauchy functional equation.
By virtue of the above theorems we can find all solutions for the following two functional equations:
• the additive Cauchy functional equation (3) when f : [0, ∞] → [0, ∞] and x, y ∈ [0, ∞],
• f (min(x + y, a)) = min( f (x) + f (y), b), where a, b > 0 are fixed, f : [0, a] → [0, b] and x, y ∈ [0, a].
These results, also for the continuous functions, have been firstly presented by Baczyński and Jayaram [6]. In this paper
we show next consequences of the presented earlier general solutions of the additive Cauchy functional equation for
the following two functional equations:
• f (x + y) = min( f (x) + f (y), b), where b > 0 is fixed, f : [0, ∞] → [0, b] and x, y ∈ [0, ∞],
• f (min(x + y, a)) = f (x) + f (y), where a > 0 is fixed, f : [0, a] → [0, ∞] and x, y ∈ [0, a].
Obtained results will be crucial in the proofs of new facts related to distributive equation (1) in fuzzy logic.
Proposition 3.4. Fix real b > 0. For a function f : [0, ∞] → [0, b] the following statements are equivalent:
(i) f satisfies the functional equation
f (x + y) = min( f (x) + f (y), b), (5)
for all x, y ∈ [0, ∞].
(ii) Either f = b, or f = 0, or

0 if x = 0,
f (x) = x ∈ [0, ∞], (6)
b if x > 0,
or

0 if x < ∞,
f (x) = x ∈ [0, ∞], (7)
b if x = ∞,
or there exists a unique constant c ∈ (0, ∞) such that
f (x) = min(cx, b), x ∈ [0, ∞]. (8)

Proof. (ii) ⇒ (i) It is obvious that f = b and f = 0 satisfy the functional equation (5).
Let f have the form (6). If x = y = 0, then the left side of (5) is equal to f (0 + 0) = f (0) = 0 and the right side of
(5) is min( f (0) + f (0), b) = min(0 + 0, b) = 0. If x  0 or y  0, then the both sides of (6) are equal to b.
Let f have the form (7). If x < ∞ and y < ∞, then x + y < ∞, so the left side of (5) is equal to f (x + y) = 0,
Also the right side of (5) is min( f (x) + f (y), b) = min(0 + 0, b) = 0. If x = ∞, then the left side of (5) is equal to
f (∞ + y) = f (∞) = b and the right side of (5) is equal to min( f (∞) + f (y), b) = min(b + f (y), b) = b. If y = ∞,
then, similarly, both sides of (5) are equal to b.
Finally, if f has the form (8) with some c ∈ (0, ∞), then the left side of (5) is equal to
f (x + y) = min(c(x + y), b), x, y ∈ [0, ∞].
Now, the right side of (5) is equal, for all x, y ∈ [0, ∞], to
min( f (x) + f (y), b) = min(min(cx, b) + min(cy, b), b) = min(cx + cy, cx + b, cy + b, b + b, b)
= min(c(x + y), b),
since b ≤ cx + b and b ≤ cy + b for all x, y ∈ [0, ∞] and b < b + b, which ends the proof in this direction.
1410 M. Baczyński / Fuzzy Sets and Systems 161 (2010) 1406 – 1419

(i) ⇒ (ii) Let f satisfy (5). Setting x = y = 0 in (5) we get

f (0) = min( f (0) + f (0), b). (9)
If f (0) = b, then for any x ∈ [0, ∞] we have
f (x) = f (x + 0) = min( f (x) + f (0), b) = min( f (x) + b, b) = b,
thus we obtain the first possible solution f = b.
Now, let us substitute x = y = ∞ in (5). We have
f (∞) = min( f (∞) + f (∞), b). (10)
If f (∞) = f (∞) + f (∞), then f (∞) = 0 and for every x ∈ [0, ∞] we get
0 = f (∞) = f (∞ + x) = min( f (∞) + f (x), b) = min(0 + f (x), b) = min( f (x), b),
thus f (x) = 0 and we obtain the second possible solution f = 0.
Let us assume that f  0 and f  b. Considering the alternate cases in (9) and (10) we get that f (0) = 0 and
f (∞) = b. Let us define
x 0 = inf{x ∈ [0, ∞]| f (x) = b}.
First we will show that if x0 < ∞, then
f (x) = b, x ∈ (x0 , ∞]. (11)
Indeed, let us take any x ∈ (x0 , ∞]. From the definition of the element x0 there exists x1 ∈ [x0 , x] such that f (x1 ) = b.
Further, from our Eq. (5), we obtain
f (x) = f (x1 + (x − x1 )) = min( f (x1 ) + f (x − x1 ), b) = min(b + f (x − x 1 ), b) = b.
Therefore, if x0 = 0, then we get the third possible solution (6).
Let us assume that x0 > 0. Now we show that for any x, y ∈ [0, x0 ) such that x + y ∈ [0, x0 ), the function f is
additive, so it satisfies (3). Suppose that this does not hold, i.e., there exist x1 , y1 ∈ [0, x0 ) such that x1 + y1 ∈ [0, x0 )
and f (x1 + y1 )  f (x1 ) + f (y1 ). Setting x = x1 and y = y1 in (5) we get
f (x1 + y1 ) = min( f (x1 ) + f (y1 ), b) = b.
But x 1 + y1 < x0 , a contradiction to the definition of the element x0 . We proved that f satisfies the additive Cauchy
functional Eq. (3) on the set H defined by (4) for A = [0, x 0 ). By Theorem 3.3 the function f can be uniquely extended
to an additive function g : R → R, such that g(x) = f (x) for all x ∈ [0, x0 ). Consequently, g is bounded on [0, x0 ),
and in virtue of Theorem of Bernstein–Doetsch g is continuous. Because of Theorem 3.1 there exists a unique constant
c ∈ R such that g(x) = cx for every x ∈ R, i.e., f (x) = cx for every x ∈ [0, x0 ). Since the domain and the range of f
is non-negative, we get that c ≥ 0. Moreover,
lim f (x) = lim cx = cx0 ≤ b.
x→x0− x→x0−

Thus, if x0 = ∞, then c = 0, and we obtain the fourth possible solution (7). If x 0 ∈ (0, ∞), then we obtain that
c ∈ [0, b/x0 ]. Let us assume that c ∈ [0, b/x0 ). We get
x x0   x  x    x x0 
0 0 0 0
f (x0 ) = f + = min f + f , b = min c + c , b = min(cx0 , b) = cx0 < b,
2 2 2 2 2 2
since c < b/x 0 . Therefore there exists x1 ∈ (0, x0 ) such that cx0 + cx1 < b. Setting x = x0 and y = x1 in (5) we get,
by x0 + x1 > x0 and (11), that
b = f (x0 + x1 ) = min( f (x0 ) + f (x1 ), b) = min(cx0 + cx1 , b) = cx0 + cx1 ,
which contradicts the above assumption.
 M. Baczyński / Fuzzy Sets and Systems 161 (2010) 1406 – 1419 1411

Consequently, we have showed that if f  0 and f  b and x0 ∈ (0, ∞), then there exists a unique c = b/x0 ∈ (0, ∞)
such that

cx if x ≤ x 0 ,
f (x) = x ∈ [0, ∞].
b if x > x0 ,
Easily calculations show that
⎧
⎪ b
⎪ cx if x ≤ b
⎨ 
x0 cx if cx ≤ b
f (x) = = = min(cx, b), x ∈ [0, ∞],
⎪
⎪ b b if cx > b
⎩ b if x > b
x0
i.e., f has the last possible representation (8). 

Corollary 3.5. Fix real b > 0. For a continuous function f : [0, ∞] → [0, b] the following statements are
equivalent:
(i) f satisfies the functional equation (5) for all x, y ∈ [0, ∞].
(ii) Either f = 0, or f = b, or there exists a unique constant c ∈ (0, ∞) such that f has the form (8).

Proposition 3.6. Fix real a > 0. For a function f : [0, a] → [0, ∞] the following statements are equivalent:
(i) f satisfies the functional equation
f (min(x + y, a)) = f (x) + f (y), (12)
for all x, y ∈ [0, a].
(ii) Either f = 0, or f = ∞, or

0 if x = 0,
f (x) = x ∈ [0, a]. (13)
∞ if x > 0,

Proof. (ii) ⇒ (i) It is obvious that f = 0 and f = ∞ satisfy the functional equation (12). Let f have the form
(13). If x = y = 0, then the left side of (12) is equal to f (min(0 + 0, a)) = f (0) = 0 and the right side of (12) is
f (0) + f (0) = 0 + 0 = 0. If x  0 or y  0, then the both sides of (12) are equal to ∞.
(i) ⇒ (ii) Setting x = y = 0 in (12) we get f (0) = f (0) + f (0), so f (0) = 0 or f (0) = ∞. If f (0) = ∞, then
for any x ∈ [0, a] we have
f (x) = f (min(x + 0, a)) = f (x) + f (0) = f (x) + ∞ = ∞,
thus we obtain the first possible solution f = ∞.
Now, let us substitute x = y = a in (12). We get f (a) = f (a) + f (a). Hence f (a) = 0 or f (a) = ∞. If f (a) = 0,
then for every x ∈ [0, a] we have
0 = f (a) = f (min(a + x, a)) = f (a) + f (x) = f (x),
thus we obtain the second possible solution f = 0.
Let us assume that f  0 and f  ∞. Considering the alternate cases above we get that f (0) = 0 and f (a) = ∞. Let
us define
x 0 = inf{x ∈ [0, a]| f (x) = ∞}.
Firstly observe that x0 < a. Indeed if we assume that x0 = a, then putting x = y = 43 a < x0 in (12) we get that the left
side is equal to f (min( 43 a + 43 a, a)) = f (a) = ∞, while the right is equal to f ( 43 a) + f ( 43 a) < ∞, a contradiction.
Therefore we can assume that x0 < a. Now we show that f (x) = ∞ for any x ∈ (x0 , a]. Indeed, let us take any
x ∈ (x0 , a]. From the definition of the element x0 there exists x1 ∈ [x0 , x] such that f (x1 ) = ∞. Now, we get
f (x) = f (x1 + (x − x1 )) = f (min(x1 + (x − x1 ), a)) = f (x1 ) + f (x − x1 ) = ∞ + f (x − x1 ) = ∞.
1412 M. Baczyński / Fuzzy Sets and Systems 161 (2010) 1406 – 1419

In the case when x0 ∈ (0, a), putting x = y = 43 x0 in (12), we get, from the above properties, that the left side of (12) is
equal to f (min( 43 x0 + 43 x0 , a)) = f (min( 23 x0 , a)) = ∞, while the right side of (12) is equal to f ( 43 x0 ) + f ( 43 x0 ) < ∞,
a contradiction.
If x0 = 0, then since f (0) = 0 we get the third possible solution (13). 

Corollary 3.7. Fix real a > 0. For a continuous function f : [0, a] → [0, ∞] the following statements are equivalent:

(i) f satisfies the functional equation (12) for all x, y ∈ [0, ∞].
(ii) Either f = 0 or f = ∞.

4. On Eq. (1) when S1 , S2 are strict t-conorms

The case when both t-conorms S1 , S2 in (1) are strict has been considered by Baczyński and Jayaram [6]. We would like
to underline that in this case the solutions of the Cauchy functional equation (3) for the function f : [0, ∞] → [0, ∞]
are used in the proofs of main theorems. We present here only main results connected with this case.

Theorem 4.1 (Baczyński and Jayaram [6, Theorem 10]). Let S1 , S2 be strict t-conorms. For a function I : [0, 1]2 →
[0, 1] the following statements are equivalent:
(i) The triple of functions S1 , S2 , I satisfies the functional equation (1) for all x, y, z ∈ [0, 1].
(ii) There exist continuous and strictly increasing functions s1 , s2 : [0, 1] → [0, ∞] with s1 (0) = s2 (0) = 0, s1 (1) =
s2 (1) = ∞, which are uniquely determined up to positive multiplicative constants, such that S1 , S2 admit the
representation (2) with s1 , s2 , respectively, and for every fixed x ∈ [0, 1], the vertical section I (x, ·) has one of
the following forms:
I (x, y) = 0, y ∈ [0, 1],
I (x, y) = 1, y ∈ [0, 1],

0 if y = 0,
I (x, y) = y ∈ [0, 1],
1 if y > 0,

0 if y < 1,
I (x, y) = y ∈ [0, 1],
1 if y = 1,
I (x, y) = s2−1 (cx s1 (y)), y ∈ [0, 1],
with a certain cx ∈ (0, ∞), uniquely determined up to a positive multiplicative constant depending on constants
for s1 and s2 , i.e., if s1 (y) = as1 (y), s2 (y) = bs2 (y) for all y ∈ [0, 1] and some a, b ∈ (0, ∞), and we assume
that

s2−1 (cx s1 (y)) = s2−1 (cx s1 (y)), y ∈ [0, 1],

then cx = (b/a)cx .

Example 4.2 (See Baczyński and Jayaram [6, Example 1]).

(i) If S1 , S2 are both strict t-conorms, then the least solution of (1) which is a fuzzy implication is the least fuzzy
implication (see [4]):

1 if x = 0 or y = 1,
I0 (x, y) =
0 otherwise.

(ii) If S1 , S2 are both strict t-conorms, then the greatest solution of (1) which is a fuzzy implication is the greatest
fuzzy implication (see [4]):

0 if x = 1 and y = 0,
I1 (x, y) =
1 otherwise.
 M. Baczyński / Fuzzy Sets and Systems 161 (2010) 1406 – 1419 1413

Corollary 4.3 (Baczyński and Jayaram [6, Corollary 5]). If S1 , S2 are strict t-conorms, then there are no continuous
solutions I of (1) which satisfy (I3).

Corollary 4.4 (Baczyński and Jayaram [6, Corollary 6]). If S1 , S2 are strict t-conorms and I is a fuzzy implication
which is continuous except at the point (0,0), then the following statements are equivalent:
(i) A triple of functions S1 , S2 , I satisfies the functional equation (1) for all x, y, z ∈ [0, 1].
(ii) There exist continuous and strictly increasing functions s1 , s2 : [0, 1] → [0, ∞] with s1 (0) = s2 (0) = 0 and
s1 (1) = s2 (1) = ∞, which are uniquely determined up to positive multiplicative constants, such that S1 , S2 admit
the representation (2) with s1 , s2 , respectively, and a continuous, decreasing function c : [0, 1] → (0, ∞] with
c(x) < ∞ for x ∈ (0, 1], c(0) = ∞, uniquely determined up to a positive multiplicative constant depending on
constants for s1 and s2 , such that I has the form

1 if x = y = 0,
I (x, y) = x, y ∈ [0, 1].
s2−1 (c(x)s1 (y)) otherwise,

Example 4.5 (Baczyński and Jayaram [6, Example 4]). One specific example is the function c(x) = 1/x defined for
all x ∈ [0, 1], with the assumption that 01 = ∞. In this case the solution I is the following:
⎧
⎪
⎨1 if x = y = 0,
I (x, y) =   x, y ∈ [0, 1].
⎪ 1
⎩ s2−1 s1 (y) otherwise,
x
In the special case when s1 = s2 , i.e., S1 = S2 , we obtain Yager’s class of g-generated fuzzy implications (see Yager
[31, p. 202]).

5. On Eq. (1) when S1 , S2 are nilpotent t-conorms

The situation when both t-conorms S1 , S2 are nilpotent has been also considered by Baczyński and Jayaram [6]. In
this case the solutions of the following functional equation:

f (min(x + y, a)) = min( f (x) + f (y), b),

where a, b > 0 are fixed, f : [0, a] → [0, b] and x, y ∈ [0, a], are used in the proofs of main theorems. We present
here only main results connected with this case.

Theorem 5.1 (Baczyński and Jayaram [6, Theorem 13]). Let S1 , S2 be nilpotent t-conorms. For a function I : [0, 1]2
→ [0, 1] the following statements are equivalent:
(i) A triple of functions S1 , S2 , I satisfies the functional equation (1) for all x, y, z ∈ [0, 1].
(ii) There exist continuous and strictly increasing functions s1 , s2 : [0, 1] → [0, ∞] with s1 (0) = s2 (0) = 0, s1 (1) < ∞
and s2 (1) < ∞, which are uniquely determined up to positive multiplicative constants, such that S1 , S2 admit the
representation (2) with s1 , s2 , respectively, and for every fixed x ∈ [0, 1], the vertical section I (x, ·) has one of
the following forms:
I (x, y) = 0, y ∈ [0, 1],
I (x, y) = 1, y ∈ [0, 1],

0 if y = 0,
I (x, y) = y ∈ [0, 1],
1 if y > 0,
I (x, y) = s2−1 (min(cx s1 (y), s2 (1))), y ∈ [0, 1],
with a certain cx ∈ [s2 (1)/s1 (1), ∞), uniquely determined up to a positive multiplicative constant depending on
constants for s1 and s2 , i.e., if s1 (y) = as1 (y), s2 (y) = bs2 (y) for all y ∈ [0, 1] and some a, b ∈ (0, ∞) and
1414 M. Baczyński / Fuzzy Sets and Systems 161 (2010) 1406 – 1419

we assume that

s2−1 (min(cx s1 (y), s2 (1))) = s2−1 (min(cx s1 (y), s2 (1))), y ∈ [0, 1],

then cx = (b/a)cx .

Example 5.2 (Baczyński and Jayaram [6, Example 6]).

(i) If S1 , S2 are both nilpotent t-conorms, then the least solution of (1) which is a fuzzy implication is the following:
⎧
⎨1    if x = 0,
I (x, y) = −1 s2 (1) x, y ∈ [0, 1].
⎩ s2 min s1 (y), s2 (1) if x > 0,
s1 (1)
In the special case, when s1 = s2 , i.e., S1 = S2 , we obtain the following fuzzy implication:

1 if x = 0,
I (x, y) = x, y ∈ [0, 1],
y if x > 0,

which is also the least (S,N)-implication (see [5, Example 1.5]).

(ii) If S1 , S2 are both nilpotent t-conorms, then the greatest solution of (1) which is a fuzzy implication is the greatest
fuzzy implication I1 (see Example 4.2(ii)).

Corollary 5.3 (Baczyński and Jayaram [6, Corollary 7]). If S1 , S2 are nilpotent t-conorms, then there are no contin-
uous solutions I of (1) which satisfy (I3).

Theorem 5.4 (Baczyński and Jayaram [6, Corollary 8]). If S1 , S2 are nilpotent t-conorms and I is a fuzzy implication
which is continuous except at the point (0, 0), then the following statements are equivalent:
(i) A triple of functions S1 , S2 , I satisfies the functional equation (1) for all x, y, z ∈ [0, 1].
(ii) There exist continuous and strictly increasing functions s1 , s2 : [0, 1] → [0, ∞] with s1 (0) = s2 (0) = 0, s1 (1) < ∞
and s2 (1) < ∞, which are uniquely determined up to positive multiplicative constants, such that S1 , S2 admit
the representation (2) with s1 , s2 , respectively, and a continuous decreasing function c : [0, 1] → [ ss21 (1)
(1)
, ∞] with
c(x) < ∞ for x ∈ (0, 1], c(0) = ∞, uniquely determined up to a positive multiplicative constant depending on
constants for s1 and s2 , such that I has the following form:

1 if x = y = 0,
I (x, y) = −1 x, y ∈ [0, 1].
s2 (min(c(x)s1 (y), s2 (1))) otherwise,

Example 5.5 (Baczyński and Jayaram [6, Example 7]). One specific example, when s2 (1) ≤ s1 (1), is again the func-
tion c(x) = 1/x, with the assumption that 01 = ∞. In this case the solution I of (1) can be written by the following
formula:
    
(−1) 1 1
I (x, y) = s2 s1 (y) = s2−1 min s1 (y), s2 (1) , x, y ∈ [0, 1],
x x
(−1)
where s2 is the pseudo-inverse of s2 (see [22]). In the special case when s1 = s2 , i.e., S1 = S2 , we again obtain
Yager’s class of g-generated fuzzy implications (see Yager [31, p. 202]).

6. On Eq. (1) when S1 is a strict t-conorm and S2 is a nilpotent t-conorm

Our main goal in this section is to present the representations of some classes of fuzzy implications that satisfy
Eq. (1) when S1 is a strict t-conorm and S2 is a nilpotent t-conorm. Within this context, we firstly describe the general
solutions of (1) with the above assumption. From this result we will show again that there are no continuous fuzzy
implications I that are solutions for (1) for strict t-conorm S1 and nilpotent t-conorm S2 , and hence we proceed to
investigate non-continuous solutions for I obeying (1).
 M. Baczyński / Fuzzy Sets and Systems 161 (2010) 1406 – 1419 1415

Theorem 6.1. Let S1 be a strict t-conorm and S2 be a nilpotent t-conorm. For a function I : [0, 1]2 → [0, 1] the
following statements are equivalent:
(i) A triple of functions S1 , S2 , I satisfies the functional equation (1) for all x, y, z ∈ [0, 1].
(ii) There exist continuous and strictly increasing functions s1 , s2 : [0, 1] → [0, ∞] with s1 (0) = s2 (0) = 0, s1 (1) = ∞
and s2 (1) < ∞, which are uniquely determined up to positive multiplicative constants, such that S1 , S2 admit the
representation (2) with s1 , s2 , respectively, and for every fixed x ∈ [0, 1], the vertical section I (x, ·) has one of
the following forms:
I (x, y) = 0, y ∈ [0, 1], (14)
I (x, y) = 1, y ∈ [0, 1], (15)

0 if y = 0,
I (x, y) = y ∈ [0, 1], (16)
1 if y > 0,

0 if y < 1,
I (x, y) = y ∈ [0, 1], (17)
1 if y = 1,
I (x, y) = s2−1 (min(cx s1 (y), s2 (1))), y ∈ [0, 1], (18)
with a certain cx ∈ (0, ∞), uniquely determined up to a positive multiplicative constant depending on constants
for s1 and s2 , i.e., if s1 (y) = as1 (y), s2 (y) = bs2 (y) for all y ∈ [0, 1] and some a, b ∈ (0, ∞), and we
assume that

s2−1 (min(cx s1 (y), s2 (1))) = s2−1 (min(cx s1 (y), s2 (1))), y ∈ [0, 1],

then cx = (b/a)cx .

Proof. (ii) ⇒ (i) Let t-conorms S1 , S2 have the representation (2) with some continuous and strictly increasing
functions s1 , s2 : [0, 1] → [0, ∞] such that s1 (0) = s2 (0) = 0 and s1 (1) = ∞ and s2 (1) < ∞. By Theorem 2.4 and
Remark 2.5(ii) the function S1 is a strict t-conorm and S2 is a nilpotent t-conorm. Let us fix arbitrarily x ∈ [0, 1]. We
consider five cases.
If I (x, y) = 0 for all y ∈ [0, 1], then the left side of (1) is I (x, S1 (y, z)) = 0 and the right side of (1) is
S2 (I (x, y), I (x, z)) = S2 (0, 0) = 0, for all y, z ∈ [0, 1].
If I (x, y) = 1 for all y ∈ [0, 1], then the left side of (1) is I (x, S1 (y, z)) = 1 and the right side of (1) is
S2 (I (x, y), I (x, z)) = S2 (1, 1) = 1, for all y, z ∈ [0, 1].
Let I (x, y) have the form (16) for all y ∈ [0, 1]. Fix arbitrarily y, z ∈ [0, 1]. If y = 0, then the left side of (1)
is I (x, S1 (0, z)) = I (x, z) and the right side of (1) is S2 (I (x, 0), I (x, z)) = S2 (0, I (x, z)) = I (x, z). Analogously, if
z = 0 then both sides of (1) are equal to I (x, y). If y > 0 and z > 0, then S1 (y, z) > S1 (0, 0) = 0 since S1 is strict.
Now, the left side of (1) is I (x, S1 (y, z)) = 1 and the right side of (1) is S2 (I (x, y), I (x, z)) = S2 (1, 1) = 1.
Let I have the form (17) for all y ∈ [0, 1]. Fix arbitrarily y, z ∈ [0, 1]. If y = 1, then the left side of (1) is
I (x, S1 (1, z)) = I (x, 1) = 1 and the right side of (1) is S2 (I (x, 1), I (x, z)) = S2 (1, I (x, z)) = 1. Analogously, if z = 1
then both sides of (1) are equal to 1. If y < 1 and z < 1, then S1 (y, z) < S1 (1, 1) = 1 since S1 is strict. Now, the left
side of (1) is I (x, S1 (y, z)) = 0 and the right side of (1) is S2 (I (x, y), I (x, z)) = S2 (0, 0) = 0.
Finally, let I have the form (18) for all y ∈ [0, 1]. For a fixed y, z ∈ [0, 1] we have
I (x, S1 (y, z)) = I (x, s1−1 (s1 (y) + s1 (z)))
= s2−1 (min(cx (s1 (y) + s1 (z)), s2 (1))),

S2 (I (x, y), I (x, z)) = S2 (s2−1 (min(cx s1 (y), s2 (1))), s2−1 (min(cx s1 (z), s2 (1))))
= s2−1 (min(min(cx s1 (y), s2 (1)) + min(cx s1 (z), s2 (1)), s2 (1)))
= s2−1 (min(cx s1 (y) + cx s1 (z), s2 (1)))
= s2−1 (min(cx (s1 (y) + s1 (z)), s2 (1)))
= I (x, S1 (y, z)).
1416 M. Baczyński / Fuzzy Sets and Systems 161 (2010) 1406 – 1419

(i) ⇒ (ii) Let us assume that functions S1 , S2 and I are the solutions of the functional equation (1) satisfying the
required properties. From Theorem 2.4 and Remark 2.5(ii) the t-conorms S1 , S2 admit the representation (2) for some
continuous additive generators s1 , s2 : [0, 1] → [0, ∞] such that s1 (0) = s2 (0) = 0, s1 (1) = ∞ and s2 (1) < ∞.
Moreover, both generators are uniquely determined up to positive multiplicative constants. Now, the functional equation
(1) becomes

I (x, s1−1 (s1 (y) + s1 (z))) = s2−1 (min(s2 (I (x, y)) + s2 (I (x, z)), s2 (1))), (19)

for all x, y, z ∈ [0, 1]. Let x ∈ [0, 1] be arbitrary but fixed. Define the function I x : [0, 1] → [0, 1] by the formula

I x (y) = I (x, y), y ∈ [0, 1].

By substitutions, s2 (1) = b, h x = s2 ◦ I x ◦ s1−1 , u = s1 (y), v = s1 (z), for y, z ∈ [0, 1], from (19) we obtain the
following functional equation:

h x (u + v) = min(h x (u) + h x (v), b), u, v ∈ [0, ∞],

where h x : [0, ∞] → [0, b] is an unknown function. By Proposition 3.4 we get all possible formulas for the function
h x , and, consequently, all the possible formulas for I x . They have the form presented in this theorem.
Finally, let us assume that for some x ∈ [0, 1] the vertical section is given by (18). We know, by Theorem 2.4, that
continuous additive generators s1 and s2 are unique up to positive multiplicative constants. We show that in this case the
constant cx in (18) depends on the above constants. To prove this let s1 (y) = as1 (y), s2 (y) = bs2 (y) for all y ∈ [0, 1]
and some a, b ∈ (0, ∞) and assume that

s2−1 (min(cx s1 (y), s2 (1))) = s2−1 (min(cx s1 (y), s2 (1))), y ∈ [0, 1].

Using the formula for the inverse of the bijection s2 and our assumption we get
 
1
s2−1 (min(cx s1 (y), s2 (1))) = s2−1 min(cx as1 (y), bs2 (1)) , y ∈ [0, 1].
b
Thus
 
cx as1 (y)
min(cx s1 (y), s2 (1)) = min , s2 (1) , y ∈ [0, 1],
b
hence, in particular, cx = (b/a)cx . 

Since we are interested in finding solutions of (1) in the fuzzy logic context, we can easily obtain an infinite number
of solutions which are fuzzy implications. It should be noted that with this assumption the vertical section (14) is not
possible, while for x = 0 the vertical section should be (15). We should also remember that a fuzzy implication is
decreasing in the first variable while it is increasing in the second one.

Example 6.2 (cf. Example 4.2).

(i) If S1 is a strict t-conorm and S2 is a nilpotent t-conorm, then the least solution of (1) which is a fuzzy implication
is the least fuzzy implication I0 .
(ii) If S1 is a strict t-conorm and S2 is a nilpotent t-conorm, then the greatest solution of (1) which is a fuzzy implication
is the greatest fuzzy implication I1 .

Now we are in a position to describe the continuous solutions of (1). The proof of this result is similar to the previous
one, but it is based on Corollary 3.5.

Corollary 6.3. Let S1 be a strict t-conorm and S2 be a nilpotent t-conorm. For a continuous function I : [0, 1]2 →
[0, 1] the following statements are equivalent:
(i) A triple of functions S1 , S2 , I satisfies the functional equation (1) for all x, y, z ∈ [0, 1].
 M. Baczyński / Fuzzy Sets and Systems 161 (2010) 1406 – 1419 1417

(ii) There exist continuous and strictly increasing functions s1 , s2 : [0, 1] → [0, ∞] with s1 (0) = s2 (0) = 0, s1 (1) = ∞
and s2 (1) < ∞, which are uniquely determined up to positive multiplicative constants, such that S1 , S2 admit
the representation (2) with s1 , s2 , respectively, and either I = 0, or I = 1, or there exists a unique continuous
function c : [0, 1] → (0, ∞), such that I has the form

I (x, y) = s2−1 (min(c(x)s1 (y), s2 (1))), x, y ∈ [0, 1]. (20)

Corollary 6.4. If S1 is a strict t-conorm and S2 is a nilpotent t-conorm, then there are no continuous solutions I of
(1) which satisfy (I3).

Proof. Let a continuous function I satisfy (I3) and (1) with some strict t-conorm S1 and some nilpotent t-conorm
S2 with continuous additive generators s1 , s2 , respectively. Then I has the form (20) with a continuous function
c : [0, 1] → (0, ∞), but in this case we get

I (0, 0) = s2−1 (min(c(0)s1 (0), s2 (1))) = s2−1 (min(0, s2 (1))) = 0,

so I does not satisfy the first condition in (I3). 

From Corollary 6.4 it is obvious that we need to look for solutions which are not continuous at the point (0,0). Using
similar methods as earlier we can prove the following fact.

Corollary 6.5. Let S1 be a strict t-conorm and S2 be a nilpotent t-conorm and I : [0, 1]2 → [0, 1] be a fuzzy
implication continuous except at the point (0,0). Then the following statements are equivalent:
(i) A triple of functions S1 , S2 , I satisfies the functional equation (1) for all x, y, z ∈ [0, 1].
(ii) There exist continuous and strictly increasing functions s1 , s2 : [0, 1] → [0, ∞] with s1 (0) = s2 (0) = 0, s1 (1) = ∞
and s2 (1) < ∞, which are uniquely determined up to positive multiplicative constants, such that S1 , S2 admit the
representation (2) with s1 , s2 , respectively, and a unique continuous decreasing function c : [0, 1] → (0, ∞] with
c(x) < ∞ for x ∈ (0, 1], c(0) = ∞, such that I has the form

1 if x = y = 0,
I (x, y) = x, y ∈ [0, 1].
s2−1 (min(c(x)s1 (y), s2 (1))) otherwise,

Remark 6.6. The function I in the above theorem can also be written in the following form

I (x, y) = s2−1 (min(c(x)s1 (y), s2 (1))), x, y ∈ [0, 1],

with the convention that 0 · ∞ = ∞ · 0 = ∞.

7. On Eq. (1) when S1 is a nilpotent t-conorm and S2 is a strict t-conorm

Our main goal in this section is to present the representations of some classes of fuzzy implications that satisfy
Eq. (1) when S1 is a nilpotent t-conorm and S2 is a strict t-conorm. Within this context, we firstly describe the general
solutions of (1) with the above assumption.

Theorem 7.1. Let S1 be a nilpotent t-conorm and S2 be a strict t-conorm. For a function I : [0, 1]2 → [0, 1] the
following statements are equivalent:
(i) A triple of functions S1 , S2 , I satisfies the functional equation (1) for all x, y, z ∈ [0, 1].
(ii) There exist continuous and strictly increasing functions s1 , s2 : [0, 1] → [0, ∞] with s1 (0) = s2 (0) = 0, s1 (1) < ∞
and s2 (1) = ∞, which are uniquely determined up to positive multiplicative constants, such that S1 , S2 admit the
representation (2) with s1 , s2 , respectively, and for every fixed x ∈ [0, 1], the vertical section I (x, ·) has one of
1418 M. Baczyński / Fuzzy Sets and Systems 161 (2010) 1406 – 1419

the following forms:

I (x, y) = 0, y ∈ [0, 1],
I (x, y) = 1, y ∈ [0, 1],

0 if y = 0,
I (x, y) = y ∈ [0, 1].
1 if y > 0,

Proof. (ii) ⇒ (i) The proof in this direction can be easily checked as it is done for Theorem 6.1.
(i) ⇒ (ii) Let us assume that functions S1 , S2 and I are the solutions of the functional equation (1) satisfying the
required properties. From Theorem 2.4 and Remark 2.5(ii) the t-conorms S1 , S2 admit the representation (2) for some
continuous additive generators s1 , s2 : [0, 1] → [0, ∞] such that s1 (0) = s2 (0) = 0, s1 (1) < ∞ and s2 (1) = ∞.
Moreover, both generators are uniquely determined up to positive multiplicative constants. Now, the functional equation
(1) becomes
I (x, s1−1 (min(s1 (y) + s1 (z), s1 (1)))) = s2−1 (s2 (I (x, y)) + s2 (I (x, z))), x, y, z ∈ [0, 1]. (21)
Let x ∈ [0, 1] be arbitrary but fixed. Define the function I x : [0, 1] → [0, 1] by the formula
I x (y) = I (x, y), y ∈ [0, 1].
By substitutions, s1 (1) = a, h x = s2 ◦ I x ◦ s1−1 , u = s1 (y), v = s1 (z), for y, z ∈ [0, 1], from (21) we obtain the
following functional equation:
h x (min(u + v, a)) = h x (u) + h x (v), u, v ∈ [0, a],
where h x : [0, a] → [0, ∞] is an unknown function. By Proposition 3.6 we get all possible formulas for the function
h x , and, consequently, all the possible formulas for I x . They have the form presented in this theorem. 

Example 7.2.
(i) If S1 is a nilpotent t-conorm and S2 is a strict t-conorm, then the least solution of (1) which is a fuzzy implication
is the following:

0 if x > 0 and y = 0,
I (x, y) =
1 otherwise.
(ii) If S1 is a nilpotent t-conorm and S2 is a strict t-conorm, then the greatest solution of (1) which is a fuzzy implication
is the greatest fuzzy implication I1 .

Remark 7.3. In this case we have only trivial continuous solutions I = 0 or I = 1, which are not fuzzy implications.

8. Conclusion

Recently, in [30,9], the authors have studied the distributivity of R- and S-implications over t-norms and t-conorms.
In [6] the authors have studied the distributivity of fuzzy implications over strict or nilpotent t-conorms. In this paper
we have continued these investigations for continuous and Archimedean t-conorms. The key mathematical results are
included in Section 3, where we have shown solutions of two functional equations related with the distributivity of
fuzzy connectives. The main results which can be important from the practical point of view in fuzzy logic systems
are included in Sections 4–7. In our future work we will concentrate on the situation when S1 and S2 are continuous
t-conorms, not necessary Archimedean.

References

[1] J. Aczél, Lectures on Functional Equations and their Applications, Academic Press, New York, 1966.
[2] M. Baczyński, On a class of distributive fuzzy implications, Int. J. Uncertainty Fuzziness Knowledge-Based Syst. 9 (2001) 229–238.
[3] M. Baczyński, Contrapositive symmetry of distributive fuzzy implications, Int. J. Uncertainty Fuzziness and Knowledge-Based Syst. 10 (2002)
135–147.
 M. Baczyński / Fuzzy Sets and Systems 161 (2010) 1406 – 1419 1419

[4] M. Baczyński, J. Drewniak, Monotonic fuzzy implication, in: P.S. Szczepaniak, P.J.G. Lisboa, J. Kacprzyk (Eds.), Fuzzy Systems in Medicine,
Studies in Fuzzines and Soft Computing, Vol. 41, Physica-Verlag, Heidelberg, 2000, pp. 90–111.
[5] M. Baczyński, B. Jayaram, On the characterizations of (S,N)-implications, Fuzzy Sets and Systems 158 (2007) 1713–1727.
[6] M. Baczyński, B. Jayaram, On the distributivity of fuzzy implications over nilpotent or strict triangular conorms, IEEE Trans. Fuzzy Syst. 17
(2009) 590–603.
[7] J. Balasubramaniam, C.J.M. Rao, A lossless rule reduction technique for a class of fuzzy system, in: N.E. Mastorakis (Ed.), Recent Advances
in Simulation, Computational Methods and Soft Computing, Proc. 3rd WSEAS Internet. Conf. on Fuzzy Sets and Fuzzy Systems, Interlaken,
Switzerland, February 12–14, 2002, pp. 228–233.
[8] J. Balasubramaniam, C.J.M. Rao, R-implication operators and rule reduction in Mamdani-type fuzzy systems, in: Proc. 6th Joint Conf. on
Information Sciences, Fuzzy Theory, Technology, Durham, USA, March 8–12, 2002, pp. 82–84.
[9] J. Balasubramaniam, C.J.M. Rao, On the distributivity of implication operators over T- and S-norms, IEEE Trans. Fuzzy Syst. 12 (2004)
194–198.
[10] F. Bernstein, G. Doetsch, Zur Theorie der konvexen Funktionen, Math. Ann. 76 (1915) 514–526.
[11] H. Bustince, P. Burillo, F. Soria, Automorphisms, negation and implication operators, Fuzzy Sets and Systems 134 (2003) 209–229.
[12] H. Bustince, M. Pagola, E. Barrenechea, Construction of fuzzy indices from fuzzy DI-subsethood measures: application to the global comparison
of images, Inf. Sci. 177 (2007) 906–929.
[13] W.E. Combs, J.E. Andrews, Combinatorial rule explosion eliminated by a fuzzy rule configuration, IEEE Trans. Fuzzy Syst. 6 (1998) 1–11.
[14] W.E. Combs, Author’s reply, IEEE Trans. Fuzzy Syst. 7 (1999) 371.
[15] W.E. Combs, Author’s reply, IEEE Trans. Fuzzy Syst. 7 (1999) 478–479.
[16] B. De Baets, E. Kerre, M. Gupta, The fundamentals of fuzzy mathematical morphology, Part 1: basic Concepts, Int. J. General Syst. 23 (1994)
155–171.
[17] B. De Baets, E. Kerre, M. Gupta, The fundamentals of fuzzy mathematical morphology, Part 2: idempotence, convexity and decomposition,
Int. J. General Syst. 23 (1995) 307–322.
[18] S. Dick, A. Kandel, Comments on “Combinatorial rule explosion eliminated by a fuzzy rule configuration”, IEEE Trans. Fuzzy Syst. 7 (1999)
475–477.
[19] J. Fodor, M. Roubens, Fuzzy Preference Modelling and Multicriteria Decision Support, Kluwer Academic Publishers, Dordrecht, 1994.
[20] S. Gottwald, A Treatise on Many-valued Logics, Research Studies Press, Baldock, Hertfordshire, 2001.
[21] L. Kitainik, Fuzzy Decision Procedures with Binary Relations, Kluwer Academic Publishers, Dordrecht, 1993.
[22] E.P. Klement, R. Mesiar, E. Pap, Triangular Norms, Kluwer Academic Publishers, Dordrecht, 2000.
[23] M. Kuczma, An Introduction to the Theory of Functional Equations and Inequalities: Cauchy’s Equation and Jensen’s Inequality. PWN-Polish
Scientific Publishers & Silesian University, Warszawa, Kraków, Katowice, 1985.
[24] C.H. Ling, Representation of associative functions, Publ. Math. Debrecen 12 (1965) 189–212.
[25] J.M. Mendel, Q. Liang, Comments on “Combinatorial rule explosion eliminated by a fuzzy rule configuration”, IEEE Trans. Fuzzy Syst. 7
(1999) 369–371.
[26] D. Ruiz-Aguilera, J. Torrens, Distributivity of strong implications over conjunctive and disjunctive uninorms, Kybernetika 42 (2005) 319–336.
[27] D. Ruiz-Aguilera, J. Torrens, Distributivity of residual implications over conjunctive and disjunctive uninorms, Fuzzy Sets and Systems 158
(2007) 23–37.
[28] B. Schweizer, A. Sklar, Probabilistic Metric Spaces, North-Holland, New York, 1983.
[29] B.A. Sokhansanj, G.H. Rodrigue, J.P. Fitch, Applying URC fuzzy logic to model complex biological systems in the language of biologists, in:
2nd Internat. Conf. on Systems Biology (ICSB 2001), Pasadena, USA, November 4–7, 2001, p. 102.
[30] E. Trillas, C. Alsina, On the law [ p ∧ q → r ] = [( p → r ) ∨ (q → r )] in fuzzy logic, IEEE Trans. Fuzzy Syst. 10 (2002) 84–88.
[31] R.R. Yager, On some new classes of implication operators and their role in approximate reasoning, Inf. Sci. 167 (2004) 193–216.