RANCHI UNIVERSITY, RANCHI

UNIVERSITY DEPARTMENT
OFMATHEMATICS

STUDY OF SOME FIXED POINT

THEOREMS IN S-METRIC SPACE

A DISSERTATION
SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF M.Sc. [MATHEMATICS]

SUBMITTED BY:-
NAME: SHAMA PERWEEN
UNIVERSITY ROLL NO.: 21MS5402956
REGISTRATION NO.: RU2021036926
SESSION.: 2021-2023

SUPERVISIOR:-
Prof. Himanshu Kumar Pandey,
Assistant Professor,
UNIVERSITY DEPARTMENT OF MATHEMATICS
R.U., Ranchi
2023
 UNIVERSITY DEPARTMENT OF MATHEMATICS
Ranchi University, Ranchi- 834008 (Jharkhand)

CERTIFICATE

This is to certify that the project report entitled with "SOME FIXED POINT
THEOREMS IN S-METRIC SPACE” submitted by Shama Perween
Registration No. - RU2021036926, University Roll No.- 21MS5402956 and
Class Roll No:-53 in the session 2021-2023 to the University Department of
Mathematics, Ranchi University, Ranchi for the partial fulfillment of
requirement for the degree of Master of Science in Mathematics is a bonafide
record of review work carried out by her under my supervision and guidance.
The contents of this project, in full or in parts, have not been submitted to
any other institute or university for the award of any degree or diploma.

Prof. Himanshu Kumar Pandey

supervisor
University Department of Mathematics,
Ranchi University, Ranchi

Dr. C.S.P. Lugun

Associate Professor
Head of Department

University Department of Mathematics,

Ranchi University, Ranchi

 UNIVERSITY DEPARTMENT OF MATHEMATICS
Ranchi University, Ranchi- 834008 (Jharkhand)

Declaration

I, Shama Perween, Registration No.: RU2021036926, University Roll No.-

21MS5402956 and Class Roll No. -53 in the session: 2021-2023 a bonafide
student of M.Sc. in Mathematics of University Department of Mathematics,
Ranchi University, Ranchi would like to declare that the dissertation entitled
with “S-Metric Space” submitted by me in partial fulfillment of the
requirements for the award of the degree of Master of Science in Mathematics is
an authentic work carried out by me.

I certify that the above declaration made above by the certificate is true.

Name : Shama Perween

University Roll No. : 21MS5402956
Registration No. : RU2021036926
Session : 2021-2023
 Acknowledgment

In the present world of competition there is a race of existence in which those

who hold strong will to come forward, succeed. Project is a bridge between
theoretical andpractical working .With a strong will and curiosity , I would like
to express my special thanks of gratitude towards my worthy Professor of
Mathematics Prof. Himanshu Kumar Pandey (Assistant Professor, Dept. of
Mathematics RU) for sparking the keen interest and curiosity for this important
topic of Mathematics “Some fixed point theorems in S- metric space” and
providing their vital support, motivation, guidance and encouraging me to the
highest peak and providing me the opportunity to prepare the project. And I am
highly obeliged in taking the opportunity to express my gratitude towards Dr.
C.S.P. Lugun (Head of Department, Dept. of Mathematics, RU) for providing
healthy and researchfriendly environment of students.

Secondly, I would like to express my sincere gratitude and grateful thanks to my

parents who are always the foremost pillars in any student’s upbringing, with
love and encouragement in every stage. Also I am deeply grateful thanks to all
the teaching staffs, Dept. of Mathematics, Ranchi University for their
suggestions and co-operation in carrying out this research. And heartfelt thanks
to my fellow friends from whom the data for this project were collected.

Name : Shama Perween

University Roll No. : 21MS5402956
Registration No. : RU2021036926
Session : 2021-2023
 CONTENTS

Chapter 1. Some fixed-point theorem in S metric space …1-11

1.1 Introduction
1.2 Preliminaries
 S-metric space
 examples
 Some basic definitions
 Fixed point theory
 Examples

Chapter 2. Some fixed-point results in S-metric space …12-18

2.1 Introduction
2.2 Preliminaries
 Common fixed-point theorem
 Main result

Chapter 3. Common fixed points of single valued and multi valued

mapping in s metric space …19-29
3.1 Introduction
3.2 Preliminaries
 Common fixed points of single and multi-valued
mapping
 Main result

Chapter 4. Common fixed-point theorems for three self-maps of a complete

S metric space …30-38
4.1Introduction
4.2 Preliminaries
 Common fixed-point theorem for three self-maps
 Main result

CONCLUSION …39

REFERENCE …40
 CHAPTER 1
SOME FIXED-POINT THEOREM IN S-METRIC METRIC
SPACE

1.1 Introduction
In the present paper, we introduce the concept of S-metric spaces and give some
properties of them. Then a common fixed point theorem for two multi-valued
mappings on complete S-metric spaces is given. In addition, we give an illustrative
example for the single valued case.
We begin with the following definition.
Definition 1.1.1. Let 𝑋 be a non-empty set. An S-metric on 𝑋 is a function 𝑆 ∶
𝑋3 → [0, ∞) that satisfies the following conditions, for each 𝑥, 𝑦, 𝑧, 𝑎 ∈ 𝑋,
1. 𝑆(𝑥, 𝑦, 𝑧) ≥ 0,
2. 𝑆(𝑥, 𝑦, 𝑧) = 0 if and only if 𝑥 = 𝑦 = 𝑧
3. S(x, y, z) ≤ S(x, x, a) + S(y, y, a) + S(z, z, a).
The pair (X, S) is called an S-metric space.
Immediate examples of such S-metric spaces are:
1.𝐿𝑒𝑡 𝑋 = 𝑅 𝑛 𝑎𝑛𝑑 || · || a norm on 𝑋, then 𝑆(𝑥, 𝑦, 𝑧) = ||𝑦 + 𝑧 − 2𝑥|| +
||𝑦 − 𝑧|| is an S-metric on 𝑋.
2. Let 𝑋 = 𝑅 𝑛 and || · || a norm on 𝑋, then 𝑆(𝑥, 𝑦, 𝑧) = ||𝑥 − 𝑧|| + ||𝑦 −
𝑧|| is an S-metric on 𝑋.
3. Let 𝑋 be a nonempty set, d is ordinary metric on 𝑋, then 𝑆(𝑥, 𝑦, 𝑧) = 𝑑(𝑥, 𝑧) +
𝑑(𝑦, 𝑧) is an S-metric on 𝑋.
Lemma 1.1.2. In an S-metric space, we have S(x, x, y) = S(y, y, x).
Proof. By third condition of S-metric, we have
(1.1) S(x, x, y) ≤ S(x, x, x) + S(x, x, x) + S(y, y, x) = S(y, y, x)
and similarly
(1.2) S(y, y, x) ≤ S(y, y, y) + S(y, y, y) + S(x, x, y) = S(x, x, y).
Hence by (1.1) and (1.2), we get S(x, x, y) = S(y, y, x).
 Definition 1.1.3. Let (X, S) be an S-metric space. For r > 0 and x ∈ X we define
the open
ball 𝐵𝑠 (𝑥, 𝑟) and closed ball 𝐵𝑠 (𝑥, 𝑟) with center 𝑥 and radius 𝑟 as follows
respectively:
𝐵𝑠 (𝑥, 𝑟) = {𝑦 ∈ 𝑋 ∶ 𝑆(𝑦, 𝑦, 𝑥 ) < 𝑟},
𝐵𝑠 (𝑥, 𝑟) = {𝑦 ∈ 𝑋 ∶ 𝑆(𝑦, 𝑦, 𝑥) ≤ 𝑟}.
Example 1.1.4. Let 𝑋 = 𝑅. Denote 𝑆(𝑥, 𝑦, 𝑧) = |𝑦 + 𝑧 − 2𝑥| + |𝑦 − 𝑧| for
all 𝑥, 𝑦, 𝑧 ∈ 𝑅.
Thus 𝐵𝑠 (1, 2) = {𝑦 ∈ 𝑅 ∶ 𝑆(𝑦, 𝑦, 1) < 2} = {𝑦 ∈ 𝑅 ∶ |𝑦 − 1| < 1} =
(0, 2).
Definition 1.1.5. Let (𝑋, 𝑆) be an S-metric space and 𝐴 ⊂ 𝑋.
1. If for every x ∈ A there exists r > 0 such that BS(x, r) ⊂ A, then the subset A is
called open subset of X.
2. Subset 𝐴 of X is said to be S-bounded if there exists 𝑟 > 0 such that
𝑆(𝑥, 𝑥, 𝑦) < 𝑟 for all 𝑥, 𝑦 ∈ 𝐴.
3. A sequence {𝑥𝑛 } in X converges to x if and only if 𝑆(𝑥𝑛 , 𝑥𝑛 , 𝑥 )→ 0 as n → ∞.
That is for each 𝜀 > 0
4. there exists n0 ∈ 𝑁 such that ∀ 𝑛 ≥ 𝑛0 ⇒ 𝑆(𝑥𝑛 , 𝑥𝑛 , 𝑥 ) < 𝜀 and we denote by
lim 𝑥𝑛 = 𝑥.
𝑛→∞
5. Sequence {𝑥𝑛 } in 𝑋 is called a Cauchy sequence if for each 𝜀 > 0 , there exists
𝑛0 ∈ 𝑁 such that
S(𝑥𝑛 , 𝑥𝑛 ,𝑥𝑚 ) < ε for each 𝑛, 𝑚 ≥ 𝑛0 .
5. The S-metric space (𝑋, 𝑆) is said to be complete if every Cauchy sequence is
convergent.
6. Let τ be the set of all 𝐴 ⊂ 𝑋 with 𝑥 ∈ 𝐴 if and only if there exists 𝑟 > 0 such
that 𝐵𝑠 (𝑥, 𝑟) ⊂ 𝐴. Then τ is a topology on 𝑋 (induced by the S-metric S).
Lemma 1.1.6. Let (𝑋, 𝑆) be an S-metric 𝑟 > 0 space. If 𝑟 > 0 and 𝑥 ∈ 𝑋, then
the ball 𝐵𝑠 (𝑥, 𝑟) is open subset of 𝑋.
Proof. Let 𝑦 ∈ 𝐵𝑠 (𝑥, 𝑟), hence 𝑆(𝑦, 𝑦, 𝑥) < 𝑟. If set 𝛿 = 𝑆(𝑥, 𝑥, 𝑦) and 𝑟 ′ =
𝑟−𝛿
then we prove that 𝐵𝑠 (𝑦, 𝑟′ ) ⊆ 𝐵𝑠 (𝑥, 𝑟). Let 𝑧 ∈ 𝐵𝑠 (𝑦, 𝑟′ ), then 𝑆(𝑧, 𝑧, 𝑦) <
2
𝑟′ .
By third condition of S-metric we have
 𝑆(𝑧, 𝑧, 𝑥 ) ≤ 𝑆(𝑧, 𝑧, 𝑦) + 𝑆(𝑧, 𝑧, 𝑦) + 𝑆(𝑥, 𝑥, 𝑦) < 2𝑟 ′ + 𝛿 = 𝑟
Hence 𝐵𝑠 (𝑦, 𝑟′ ) ⊆ 𝐵𝑠 (𝑥, 𝑟).
That is the ball 𝐵𝑠 (𝑥, 𝑟) is a open subset of 𝑋.
Lemma 1.1.7. Let (X, S) be an S-metric space. If sequence {𝑥𝑛 } in 𝑋 converges to
𝑥, then 𝑥 is unique.
Proof. Let {𝑥𝑛 } converges to 𝑥 and 𝑦, then for each 𝜀 > 0 there exist 𝑛1, 𝑛2 ∈ 𝑁
such that
∀ 𝑛 ≥ 𝑛1 ⇒ 𝑆(𝑥𝑛 , 𝑥𝑛 , 𝑥) < 𝜀/4
And,
∀ 𝑛 ≥ 𝑛2 ⇒ 𝑆(𝑥𝑛 , 𝑥𝑛 , 𝑦) < 𝜀/2 . If set 𝑛0 = 𝑚𝑎𝑥 {𝑛1 , 𝑛2 }, then for every 𝑛 ≥
𝑛0 by third condition S-metric we have:
𝑆(𝑥, 𝑥, 𝑦) ≤ 2𝑆(𝑥, 𝑥, 𝑥𝑛 ) + 𝑆(𝑦, 𝑦, 𝑥𝑛 ) < 𝜀/ 2 + 𝜀/ 2 = 𝜀.
Hence 𝑆(𝑥, 𝑥, 𝑦) = 0 𝑠𝑜 𝑥 = 𝑦.
Lemma 1.1.8. Let (𝑋, 𝑆) be an S-metric space. If sequence 𝑥𝑛 in 𝑋 is converges to
𝑥, then {𝑥𝑛 } is a Cauchy sequence.
Proof. Since lim 𝑥𝑛 = 𝑥 then for each 𝜀 > 0 there exists 𝑛1 , 𝑛2 ∈ 𝑁 such
𝑛→∞
that 𝑛 ≥ 𝑛1 ⇒ 𝑆(𝑥𝑛 𝑥𝑛, 𝑥) < 𝜀 /4
and
𝑚 ≥ 𝑛2 ⇒ 𝑆(𝑥𝑚 𝑥𝑚 , 𝑥) < 𝜀/2.
If set 𝑛0 = 𝑚𝑎𝑥{𝑛1 , 𝑛2 }, then for every 𝑛, 𝑚 ≥ 𝑛0 by third condition of S-metric
we have:
𝜀 𝜀
𝑆(𝑥𝑛 , 𝑥𝑛 , 𝑥𝑚 ) ≤ 2𝑆 (𝑥𝑛 , 𝑥𝑛 , 𝑥 ) + 𝑆(𝑥𝑚 , 𝑥𝑚 , 𝑥 ) < + = 𝜀.
2 2

Hence {𝑥𝑛 } is a Cauchy sequence.

Lemma 1.1.9. Let (𝑋, 𝑆) be an S- metric space. If there exist sequences {𝑥𝑛 } and
{𝑦𝑛 } 𝑠uch that lim 𝑥𝑛 = 𝑥 and lim 𝑦𝑛 = 𝑦, then
𝑛→∞ 𝑛→∞

lim 𝑆(𝑥𝑛 , 𝑥𝑛 , 𝑦𝑛 ) = 𝑆(𝑥, 𝑥, 𝑦).

𝑛→∞
 Proof. Since lim 𝑥𝑛 = 𝑥 and lim 𝑦𝑛 = 𝑦, then for each ε > 0 there exist
𝑛→∞ 𝑛→∞
𝑛1 , 𝑛2 ∈ N such that
∀ 𝑛 ≥ 𝑛1 ⇒ 𝑆(𝑥𝑛 , ) < 𝜀/ 4
and
∀ 𝑛 ≥ 𝑛2 ⇒ 𝑆(𝑦𝑛 , 𝑦𝑛 , 𝑦) < 𝜀 /4
. If set 𝑛0 = 𝑚𝑎𝑥{𝑛1, 𝑛2 }, then for every 𝑛 ≥ 𝑛0 by third condition of S-metric
we have:
𝑆(𝑥𝑛 , 𝑥𝑛 , 𝑦𝑛 ) ≤ 2𝑆(𝑥𝑛 , 𝑥𝑛 , 𝑥) + 𝑆(𝑦𝑛 , 𝑦𝑛 , 𝑥)
≤ 2𝑆(𝑥𝑛 , 𝑥𝑛 , 𝑥) + 2𝑆(𝑦𝑛 , 𝑦𝑛 , 𝑦) + 𝑆(𝑥, 𝑥, 𝑦)
< 𝜀/ 2 + 𝜀/ 2 + 𝑆(𝑥, 𝑥, 𝑦) = 𝜀 + 𝑆(𝑥, 𝑥, 𝑦).
Hence we have:
(1.3) 𝑆(𝑥𝑛 , 𝑥𝑛 , 𝑥) − 𝑆(𝑥, 𝑥, 𝑦) < 𝜀.
On the other hand, we have
𝑆(𝑥, 𝑥, 𝑦) ≤ 2𝑆(𝑥, 𝑥, 𝑥𝑛 ) + 𝑆(𝑦, 𝑦, 𝑥𝑛 )
≤ 2𝑆(𝑥, 𝑥, 𝑥𝑛 ) + 2𝑆(𝑦, 𝑦, 𝑦𝑛 ) + 𝑆(𝑥𝑛 , 𝑥𝑛 , 𝑦𝑛 )
< 𝜀 /2 + 𝜀/ 2 + 𝑆(𝑥𝑛 , 𝑥𝑛 , 𝑦𝑛 ) = 𝜀 + 𝑆(𝑥𝑛 , 𝑥𝑛 , 𝑦𝑛 ),
that is
(1.4) 𝑆(𝑥, 𝑥, 𝑦) − 𝑆(𝑥𝑛 , 𝑥𝑛 , 𝑦𝑛 ) < 𝜀.
Therefore by relations (1.3) and (1.4) we have |𝑆(𝑥𝑛 , 𝑥𝑛 , 𝑦𝑛 ) − 𝑆(𝑥, 𝑥, 𝑦)| < 𝜀,
that is
lim 𝑆(𝑥𝑛 , 𝑥𝑛 , 𝑦𝑛 ) = 𝑆(𝑥, 𝑥, 𝑦).
𝑛→∞

Let (𝑋, 𝑆) be an S-metric space, 𝐶(𝑋) denotes the family of all nonempty closed
subsets of 𝑋. For 𝐴 and 𝐵 two nonempty subsets of 𝑋 we define; 𝑑𝑖𝑠𝑡(𝑥, 𝐴) =
𝑖𝑛𝑓 𝑎 ∈ 𝐴 {𝑆(𝑥, 𝑥, 𝑎)}
And
𝑆(𝐴, 𝐴, 𝐵) = 𝑠𝑢𝑝 𝑎 ∈ 𝐴, 𝑏 ∈ 𝐵 {𝑆(𝑎, 𝑎, 𝑏)}. By the definition of 𝑑𝑖𝑠𝑡(𝑥, 𝐴), it is
clear that 𝑑𝑖𝑠𝑡(𝑥, 𝐴) = 0 ⇔ 𝑥 ∈ 𝐴.
1.2. Implicit Relations
Implicit relations on metric spaces have been used in many articles. For examples,
[1], [2], [3], [4], [5], [6], [7], [8]. Let R+ be the set of nonnegative real numbers and
let T be the set of all functions 𝑇 ∶ 𝑹 𝟔+ → 𝑹 satisfying the following condition.
𝑇0 ∶ 𝑇( lim 𝑖𝑛𝑓 𝑝𝑛 ) ≤ lim inf 𝑇(𝑝𝑛 ) for any 𝑝𝑛 ∈ 𝑅+6 , where 𝑙𝑖𝑚 𝑖𝑛𝑓 𝑛 →
𝑛→∞ 𝑛→∞
∞ 𝑝𝑛 means component-wise lim inf.
𝑇1 ∶ 𝑇(𝑡1, . . . , 𝑡6) is non increasing in 𝑡2, . . . , 𝑡6
𝑇2: there exists a continuous strictly increasing function 𝜙 ∶ 𝑅+ → 𝑅 + with
𝜙(𝑡) < 𝑡 𝑓𝑜𝑟 𝑡 > 0 𝑎𝑛𝑑 𝜀 > 0 such that the inequalities
𝑢 ≤ 𝑤 + 𝜀
and
𝑇(𝑤, 𝑣, 𝑣, 𝑢, 2𝑢 + 𝑣, 0) ≤ 0 𝑜𝑟 𝑇(𝑤, 𝑣, 𝑢, 𝑣, 0, 2𝑢 + 𝑣) ≤ 0
implies 𝑤 ≤ 𝜙(𝑣).
𝑇3 ∶ 𝑇(𝑤, 0, 𝑣, 0, 0, 𝑣) ≤ 0 𝑎𝑛𝑑 𝑇(𝑤, 0, 0, 𝑣, 𝑣, 0) ≤ 0 implies 𝑤 ≤ 𝜙(𝑣), where
𝜙 is the function in 𝑇2.
1
Example1.2.1. 𝑇(𝑡1, . . . , 𝑡6 ) = 𝑡1 − 𝑓(𝑚𝑎𝑥{𝑡2, 𝑡3 , 𝑡4 , (𝑡 5 + 𝑡6 )}), where f :
3
𝑅+ → 𝑅 + continuous strictly increasing function with 𝑓(𝑡) < 𝑡 𝑓𝑜𝑟 𝑡 > 0.
𝑇0 and 𝑇1 : Obviously.
𝑇2 : Let u > 0, then choose ε > 0 so that f(u)+ε < u (this is possible since f(u) < u).
Now let u ≤ w + ε and T(w, v, v, u, 2u + v, 0) = w − f(max{u, v}) ≤ 0. If u ≥ v, then
u ≤ w +ε ≤ f(u) +ε < u, a contradiction. Thus u < v and w ≤ f(v). Similarly, u ≤ w +ε
and T(w, v, u, v, 0, 2u + v) ≤ 0 imply w ≤ f(v). If u = 0, then w ≤ f(v). Thus 𝑇2 is
satisfied with ϕ = f.
𝑇3 : T(w, 0, v, 0, 0, v) = T(w, 0, 0, v, v, 0) = w − f(v) ≤ 0 ⇒ w ≤ f(v) = ϕ(v).
1.3.Fixed Point Theory
Our main result for fixed point theory of this work as follows.
Theorem1.3.1 Let (𝑋, 𝑆) be a complete S-metric space, 𝑥0 ∈ 𝑋, 𝑟 > 0 with F, G :
𝐵𝑆 (𝑥0 , 𝑟) → 𝐶(𝑋). Suppose, for all 𝑥, 𝑦 ∈ 𝐵𝑠 (𝑥0 , 𝑟) sets 𝐹 𝑥 , 𝐺𝑦 are bounded and
(3.1) T(S(𝐹 𝑥 , 𝐹𝑦 , 𝐺𝑦 ), S(x, x, y), dist(x, 𝐹𝑥 ), dist(y, 𝐺𝑦 ), dist(x,𝐺𝑦 ), dist(y,𝐹𝑥 )) ≤
0 where T ∈ 𝜏. Also assume the following conditions are satisfied:
𝑟 − 𝜙(𝑟)
(3.2) 𝑑𝑖𝑠𝑡(𝑥0 , 𝐹𝑥0 ) < and
2
( 𝑟 − 𝜙(𝑟) 𝜙(𝑟)
(3.3) ∑∞
𝑖=1 ∅
𝑖
) ≤ where ϕ is the function in 𝑇2 . Then there
2 2
exists 𝑥 ∈ 𝐵𝑆 (𝑥0 , 𝑟) 𝑥 ∈ 𝐹𝑥 with and 𝑥 ∈ 𝐺𝑥 .
Proof. From (3.2) we can choose 𝑥1 ∈ 𝐹 𝑥0 with
𝑟 − 𝜙(𝑟)
(3.4) 𝑆(𝑥0 , 𝑥0 , 𝑥1 ) <
2

Hence 𝑆𝑥1 , 𝑥2 , 𝑥0 ) < 𝑟 𝑠𝑜 𝑥1 ∈ 𝐵𝑠 [𝑥0 , 𝑟]. Since ϕ is strictly increasing by (3.4)

we can choose ε > 0 such that
( 𝑟 − 𝜙(𝑟))
(3.5) 𝜙(𝑆(𝑥0 , 𝑥0 , 𝑥1 )) + 𝜀 < 𝜙 .
2

On the other hand, for this ε there is 𝑥2 ∈ 𝐺𝑥1 so that

(3.6) 𝑆(𝑥1, 𝑥1 , 𝑥2 ) ≤ 𝑑𝑖𝑠𝑡(𝑥1 , 𝐺𝑥1 ) + 𝜀 ≤ 𝑆(𝐹𝑥0 𝐹𝑥0 , 𝐺𝑥1 ) + 𝜀.

Now since 𝑥0 , 𝑥1 ∈ 𝐵𝑠 [𝑥0 , 𝑟] we can use the inequality (3.1) to obtain

T(S(𝐹𝑥0 , 𝐹𝑥0 , 𝐺𝑋1 ), S(𝑥0 , 𝑥0 , 𝑥1 ), dist(𝑥0 , 𝐹𝑥0 ), dist(𝑥1 , 𝐺𝑥1 ), dist(𝑥0 , 𝐺𝑥1 ), dist(𝑥1 , F
𝑥0 )) ≤ 0.
From 𝑇1 we have
𝑇(𝑆(𝐹𝑥0 , 𝐹𝑥0 , 𝐺𝑥1 ), 𝑆(𝑥0 , 𝑥0 , 𝑥1 ), 𝑆(𝑥0 , 𝑥0 , 𝑥1), 𝑆(𝑥1 , 𝑥1 , 𝑥2 ), 𝑆(𝑥0 , 𝑥0 , 𝑥1 ), 0) ≤ 0,

that is
𝑇(𝑤, 𝑣, 𝑣, 𝑢, 2𝑢 + 𝑣, 0) ≤ 0,
where 𝑤 = 𝑆(𝐹𝑥0 , 𝐹𝑥0 , 𝐺𝑥1 ), 𝑣 = 𝑆(𝑥0 , 𝑥0 , 𝑥1 ) and 𝑢 = 𝑆(𝑥1 , 𝑥1 , 𝑥2 ).

Therefore, from 𝑇2,

𝑆(𝐹𝑥0 , 𝐹𝑥0 , 𝐺𝑥1 ) ≤ 𝜙(𝑆(𝑥0 , 𝑥0 , 𝑥1 ))

and (3.6) yields

𝑆(𝑥1 , 𝑥1, 2) ≤ 𝜙(𝑆(𝑥0 , 𝑥0 , 𝑥1 )) + 𝜀.
Thus from (3.5) we have:
 ( 𝑟 − 𝜙(𝑟)
(3.7) 𝑆(𝑥1 , 𝑥1, 𝑥2 ) < 𝜙
2

Now by (3.3), (3.4), (3.7) and third condition of S-metric have:

𝑆(𝑥2 , 𝑥2 , 𝑥0 ) = 𝑆(𝑥0 , 𝑥,0 𝑥2 )
≤ 2𝑆(𝑥0 , 𝑥0 , 𝑥1 ) + 𝑆(𝑥1 , 𝑥1, 𝑥2 )
𝑟 − 𝜙(𝑟)
< 𝑟 − 𝜙(𝑟) + 𝜙 ( )
2
𝑟 − 𝜙(𝑟)
< 𝑟 − 𝜙(𝑟) + 2 ∑∞ 𝑖
𝑖=1 ∅ ( ) ≤ 𝑟
2

so 𝑥2 ∈ 𝐵𝑠 [𝑥0 , 𝑟]. Again by (3.7) and strictly increasing 𝜙 there is 𝛿 > 0 so that
𝑟 – 𝜙(𝑟)
(3.8) 𝜙(𝑆(𝑥1, 𝑥1 , 𝑥2 )) + 𝛿 < 𝜙 2 ( ),
2

also for this 𝛿 > 0 there is 𝑥3 ∈ 𝐹𝑥2 so that

(3.9) 𝑆(𝑥2 , 𝑥2 , 𝑥3 ) ≤ 𝑑𝑖𝑠𝑡(𝑥2 , 𝐹 𝑥2 ) + 𝛿 ≤ 𝑆(𝐺𝑥1 , 𝐺𝑥1, 𝐹 𝑥2 ) + 𝛿.

As above, since 𝑥1 , 𝑥2 ∈ 𝐵𝑠 [𝑥0 , r] we can use the inequality (3.1) to obtain

𝑇(𝑆(𝐹𝑥2 𝐹 𝑥2 , 𝐺𝑥1 ), 𝑆(𝑥2 , 𝑥2 , 𝑥1 ), 𝑑𝑖𝑠𝑡(𝑥2 , 𝐹 𝑥2 ),

𝑑𝑖𝑠𝑡(𝑥1 , 𝐺𝑥1 ), 𝑑𝑖𝑠𝑡(𝑥2 , 𝐺𝑥1 ), 𝑑𝑖𝑠𝑡(𝑥1 , 𝐹𝑥2 )) ≤ 0

and so from 𝑇1 we have

𝑇(𝑆(𝐹𝑥2 , 𝐹 𝑥2 , 𝐺𝑥1 ), 𝑆(𝑥2 , 𝑥2 , 𝑥1), 𝑆(𝑥2 , 𝑥2 , 𝑥3 ), 𝑆(𝑥1 , 𝑥1 , 𝑥2 ), 0, 𝑆(𝑥1, 𝑥1 , 𝑥3 )) ≤ 0
that is
𝑇(𝑤, 𝑣, 𝑢, 𝑣, 0, 2𝑢 + 𝑣) ≤ 0,
where 𝑤 = 𝑆(𝐹 𝑥2 , 𝐹 𝑥2 , 𝐺𝑥1 ), 𝑣 = 𝑆(𝑥1 , 𝑥1 , 𝑥2 ) and 𝑢 = 𝑆(𝑥2 , 𝑥2 , 𝑥3 ).
Therefore from 𝑇2
w ≤ ϕ(v)
that is
𝑆(𝐹𝑥2 , 𝐹𝑥2 , 𝐺𝑥1 ) ≤ 𝜙(𝑆(𝑥1, 𝑥1 , 𝑥2 ))

and so (3.9) gives

𝑆(𝑥2 , 𝑥2 , 𝑥3 ) ≤ 𝜙(𝑆(𝑥1 , 𝑥1, 𝑥2 )) + 𝛿.
Thus from (3.8) we have
𝑟 − 𝜙(𝑟)
(3.10) 𝑆(𝑥2 , 𝑥,2 𝑥3 ) < ∅2 ( ).
2

Now (3.3), (3.4), (3.7), (3.10) and third condition of S-metric implies:
𝑆(𝑥3, 𝑥3, 𝑥0) = 𝑆(𝑥0, 𝑥0, 𝑥3) ≤ 2𝑆(𝑥0, 𝑥0, 𝑥1) + 2𝑆(𝑥1, 𝑥1, 𝑥2) +
𝑆(𝑥2, 𝑥2, 𝑥3)
< 𝑟 − 𝜙(𝑟) + 2𝜙 ( 𝑟 − 𝜙(𝑟) 2 ) + 𝜙 2 ( 𝑟 − 𝜙(𝑟) 2 )
≤ 𝑟 − 𝜙(𝑟) + 2∑∞ 𝑖 = 1 𝜙 𝑖 ( 𝑟 − 𝜙(𝑟) 2 ) ≤ 𝑟
Thus 𝑥3 ∈ 𝐵𝑠 [𝑥0 , 𝑟].
Continuing this way we can obtain sequence 𝑥𝑛 ⊆ 𝐵𝑠 [𝑥0 , 𝑟] such that 𝑥2𝑛+1 ∈
𝐺𝑥2𝑛+1 and𝑥2𝑛+1 ∈ 𝐹𝑥2𝑛 for 𝑛 ≥ 0 and
𝑟 − 𝜙(𝑟)
𝑆(𝑥𝑛 , 𝑥𝑛 , 𝑥𝑛+1 ) < 𝜙 𝑛 ( ).
2
Next we show that {𝑥𝑛 } is a Cauchy sequence. Notice by (3.3) and above inequality
for each 𝑛, 𝑚 ∈ 𝑁 with 𝑚 > 𝑛 we have:
𝑚−2

𝑆(𝑥𝑛 , 𝑥𝑛 , 𝑥𝑚 ) ≤ 2 ∑ 𝑆(𝑥𝑖 , 𝑥𝑖 , 𝑥𝑖+1 + 𝑆(𝑥𝑚−1 , 𝑥𝑚−1, 𝑥𝑚 )

𝑖=𝑛
𝑚−1

≤ 2 ∑ 𝑆(𝑥𝑖 , 𝑥𝑖 , 𝑥𝑖+1)
𝑖=𝑛
𝑟 − 𝜙(𝑟)
< 2 ∑𝑚−1 𝑖
𝑖=𝑛 ∅ ( )
2
𝑟 − 𝜙(𝑟)
≤ 2 ∑∞ 𝑖
𝑖=𝑛 ∅ ( )
2

so (3.3) guarantees that {𝑥𝑛 } is a Cauchy sequence. Thus there exists 𝑥 ∈ 𝐵𝑠 [𝑥0 , 𝑟]
with 𝑥 𝑛 → 𝑥. It remains to show 𝑥 ∈ 𝐹𝑥 and 𝑥 ∈ 𝐺𝑥 . For 𝑛 even (since 𝑥𝑛 , 𝑥 ∈
𝐵𝑠 [𝑥0 , 𝑟]) we can use the inequality (3.1), we have
𝑇(𝑆(𝐹 𝑥 , 𝐹𝑥 , 𝐺𝑥𝑛 ), 𝑆(𝑥, 𝑥, 𝑥𝑛−1), 𝑑𝑖𝑠𝑡(𝑥, 𝐹𝑥 ), 𝑑𝑖𝑠𝑡(𝑥𝑛−1, 𝐺𝑥𝑛 ), 𝑑𝑖𝑠𝑡(𝑥, 𝐺𝑥𝑛−1 ),

𝑑𝑖𝑠𝑡(𝑥𝑛−1, 𝐹𝑥 ) ≤ 0.
Now taking limit inferior 𝑎𝑠 𝑛 → ∞ (using 𝑇0 we have (notice 𝑑𝑖𝑠𝑡(𝑥, 𝐺𝑥𝑛 − 1) ≤
𝑆(𝑥, 𝑥, 𝑥𝑛 ) → 0, 𝑎𝑛𝑑 𝑎𝑙𝑠𝑜 𝑑𝑖𝑠𝑡(𝑥𝑛−1, 𝐺𝑥𝑛−1) ≤ 𝑆(𝑥𝑛−1, 𝑥𝑛−1, 𝑥𝑛 ) → 0)
 𝑇( lim 𝑖𝑛𝑓 𝑆(𝐹𝑥 , 𝐹𝑥 , 𝐺𝑥𝑛−1 ), 0, 𝑑𝑖𝑠𝑡(𝑥, 𝐹𝑥 ), 0, 0, 𝑑𝑖𝑠𝑡(𝑥, 𝐹𝑥 )) ≤ 0.
𝑛→∞

From 𝑇3 we have
𝑖𝑛𝑓 𝑆(𝐹𝑥 , 𝐹𝑥 , 𝐺𝑥𝑛 − 1) ≤ 𝜙(𝑑𝑖𝑠𝑡(𝑥, 𝐹𝑥 )).

Now 𝑑𝑖𝑠𝑡 (𝑥, 𝐹𝑥 ) ≤ 2𝑆 (𝑥, 𝑥, 𝑥𝑛 ) + dist(𝑥𝑛 , Fx ) ≤ 2𝑆(𝑥, 𝑥, 𝑥𝑛 ) +

𝑆(𝐺𝑥𝑛−1, 𝐺𝑥𝑛−1 𝐹𝑥 )
and so
𝑑𝑖𝑠𝑡(𝑥, 𝐹 𝑥) ≤ 0 + lim 𝑖𝑛𝑓 𝑆(𝐹𝑥 , 𝐹𝑥 , 𝐺𝑥𝑛−1 ) ≤ 𝜙(𝑑𝑖𝑠𝑡(𝑥, 𝐹𝑥 )).
𝑛→∞

Thus 𝑑𝑖𝑠𝑡(𝑥, 𝐹𝑥 ) = 0 𝑠𝑖𝑛𝑐𝑒 𝜙(𝑡) < 𝑡 𝑓𝑜𝑟 𝑡 > 0, 𝑠𝑜 𝑥 ∈ 𝐹 𝑥 = 𝐹𝑥 .

For 𝑛 odd ,
𝑑𝑖𝑠𝑡(𝑥, 𝐺𝑥 ) ≤ 𝑆(𝑥, 𝑥, 𝑥𝑛 ) + 𝑑𝑖𝑠𝑡(𝑥𝑛 , 𝐺𝑥 ) ≤ 𝑆(𝑥, 𝑥, 𝑥𝑛 ) +
𝑆( 𝐹𝑥𝑛−1, 𝐹𝑥𝑛−1 , 𝐺𝑥 ),

and as above we obtain 𝑑𝑖𝑠𝑡(𝑥, 𝐺𝑥 ) = 0, 𝑠𝑜 𝑥 ∈ 𝐺𝑥 .

Now we give some corollaries.
Corollary 1.3.2. Let (𝑋, 𝑆) be a complete S-metric space, 𝑥0 ∈ 𝑋, r > 0 with 𝐹, 𝐺 ∶
𝐵𝑠 [𝑥0, 𝑟] → 𝐶(𝑋). Suppose, for all x, y ∈𝐵𝑠 [𝑥0 , r] sets 𝐹𝑥 , 𝐺𝑦 are bounded and
𝑆(𝐹𝑥 , 𝐹𝑥 , 𝐺𝑦 ) ≤
𝑑𝑖𝑠𝑡(𝑥,𝐺𝑦 ) 𝑑𝑖𝑠𝑡(𝑦,𝐹 𝑥 )
𝑘 𝑚𝑎𝑥{𝑆(𝑥, 𝑥, 𝑦), 𝑑𝑖𝑠𝑡(𝑥, 𝐹𝑥 ), 𝑑𝑖𝑠𝑡(𝑦, 𝐺𝑦 ), , } 𝑤ℎ𝑒𝑟𝑒 0 < 𝑘 <
3 3
1−𝑘
1. Also assume the following condition is satisfied: 𝑑𝑖𝑠𝑡(𝑥0 , 𝐹𝑥0 ) < 𝑟
2

Then there exists 𝑥 ∈ 𝐵𝑠 [𝑥0 , 𝑟] with 𝑥 ∈ 𝐹 𝑥 and 𝑥 ∈ 𝐺𝑥

Proof: By Theorem 3.1 , it is enough to set 𝑇(𝑡1, 𝑡2 ,..., 𝑡6) = 𝑡1 −
𝑡5 𝑡6
𝑘 max{𝑡2, 𝑡3 , 𝑡4 , , }.
3 3

In this case, 𝜙(𝑡) = 𝑘𝑡 and

∞
𝑟 − 𝜙(𝑟) 𝑘𝑟 𝜙(𝑟)
∑ ∅𝑖 ( ) = = .
2 2 2
𝑖=1

Corollary 1.3.3. Let (𝑋, 𝑆) be a complete S-metric space, 𝑥0 ∈ 𝑋, 𝑟 > 0 with

𝐹, 𝐺 ∶ 𝐵𝑠 [𝑥0 , 𝑟] → 𝑋. Suppose for all 𝑥, 𝑦 ∈ 𝐵𝑠 [𝑥0 , 𝑟],
. Corollary 3.3. Let (𝑋, 𝑆) be a complete S-metric space, 𝑥 0 ∈ 𝑋, 𝑟 > 0 with 𝐹, 𝐺 ∶
𝐵𝑠 [𝑥0 , 𝑟] → 𝑋. Suppose for all 𝑥, 𝑦 ∈ 𝐵𝑠 [𝑥0 , 𝑟],
𝑆(𝐹 𝑥, 𝐹𝑥 , 𝐺𝑦 )
≤ 𝑘 𝑚𝑎𝑥{𝑆(𝑥, 𝑥, 𝑦), 𝑆(𝑥, 𝑥, 𝐹 𝑥), 𝑆(𝑦, 𝑦, 𝐺𝑦), 𝑆(𝑥, 𝑥, 𝐺𝑦) 3 , 𝑆(𝑦, 𝑦, 𝐹 𝑥) 3 }
where 0 < 𝑘 < 1. Also assume the following condition is satisfied:
1 − 𝑘
𝑆(𝑥0 , 𝑥0 , 𝐹𝑥0 ) < 𝑟.
2
Then there exists a unique 𝑥 ∈ 𝐵𝑠 [𝑥0 , 𝑟] 𝑤𝑖𝑡ℎ 𝐹𝑥 = 𝐺𝑥 = 𝑥.
Proof. By Corollary 3.2 , there exists an 𝑥 ∈ 𝑋 such that 𝐹𝑥 = 𝐺𝑥 = 𝑥. It is
enough prove that 𝑥 is unique.
Let y be another common fixed point of F and G, that is 𝑦 = 𝐹𝑦 = 𝐺𝑦 , then we
have 𝑆(𝑥, 𝑥, 𝑦) = 𝑆(𝐹 𝑥, 𝐹 𝑥, 𝐺𝑦) ≤ 𝑘 𝑚𝑎𝑥{𝑆(𝑥, 𝑥, 𝑦), 𝑆(𝑥, 𝑥, 𝑥), 𝑆(𝑦, 𝑦, 𝑦)}
= 𝑘𝑆(𝑥, 𝑥, 𝑦)
, which is a contradiction. Therefore 𝐹 and 𝐺 have a unique common fixed point in
𝐵𝑆[𝑥0, 𝑟].
Corollary 1.3.4. Let (𝑋, 𝑆) be a complete S-metric space,𝑥0 ∈ 𝑋, 𝑟 > 0 with 𝐹 ∶
𝐵𝑠 [𝑥0 , 𝑟] → 𝑋. Suppose for all 𝑥, 𝑦 ∈ [𝑥0 , 𝑟],
𝑆(𝑥, 𝑥, 𝐹𝑦 ) 𝑆(𝑦, 𝑦, 𝐹𝑥 )
𝑆(𝐹𝑥 , 𝐹𝑥 , 𝐹𝑦 ) ≤ 𝑘 𝑚𝑎𝑥{𝑆(𝑥, 𝑥, 𝑦), 𝑆(𝑥, 𝑥, 𝐹𝑥 ), 𝑆(𝑦, 𝑦, 𝐹𝑦 ), , }
3 3
where 0 < 𝑘 < 1. Also assume the following condition is satisfied:
1−𝑘
𝑆(𝑥0 , 𝑥0 , 𝐹𝑥0 ) < 𝑟.
2

Then there exists a unique 𝑥 ∈ 𝐵𝑠 [𝑥0, 𝑟] with 𝐹𝑥 = 𝑥.

Now we give an example.
Example 1.3.5. Let 𝑋 = 𝑅 and 𝑆(𝑥, 𝑦, 𝑧) = |𝑥 − 𝑧| + |𝑦 − 𝑧| .Then (𝑋, 𝑆) is
a complete S-metric space. Let 𝑥0 = 1 and 𝑟 = 6, then
𝐵𝑠 [𝑥0 , 𝑟] = 𝐵𝑠 [1, 6]
= {𝑦 ∈ 𝑋 ∶ 𝑆(𝑦, 𝑦, 𝑥) ≤ 6}
= [−2, 4].
Now let 𝐹 ∶ 𝐵𝑆[𝑥0, 𝑟] → 𝑋, 𝐹 𝑥 = 𝑥 2 𝑎𝑛𝑑 𝑙𝑒𝑡 𝑘 = 1 2 , then
 3 (1 – 𝑘)
𝑆(𝑥0 , 𝑥0 , 𝐹𝑥0 ) = 𝑆(1, 1, 1 /2 ) = 1 < = 𝑟.
2 2
Also, for all 𝑥, 𝑦 ∈ 𝐵𝑠 [𝑥0 , 𝑟], we have
𝑆(𝐹𝑥 , 𝐹𝑦 , 𝐹𝑦 ) = 2 |𝐹𝑥 − 𝐹𝑦 |

= |𝑥 − 𝑦|
1
= (2 |𝑥 − 𝑦|)
2
1
= 𝑆(𝑥, 𝑥, 𝑦)
2
1 𝑆(𝑥,𝑥,𝐹𝑦 ) 𝑆(𝑦,𝑦,𝐹𝑥 )
≤ 𝑚𝑎𝑥{𝑆(𝑥, 𝑥, 𝑦), 𝑆(𝑥, 𝑥, 𝐹𝑥 ), 𝑆(𝑦, 𝑦, 𝐹𝑦 ), , }.
2 3 3

Therefore all conditions of Corollary 3.4 are satisfied, thus F has a unique fixed
point in 𝐵𝑠 [𝑥0 , 𝑟] = [−2, 4].
 CHAPTER 2
SOME FIXED-POINT RESULTS ON S-METRICS SPACES

2.1 Introduction and preliminaries: - In 1922, Banach proposed a theorem, which

is well-known as Banach’s Fixed Point Theorem to establish the existence of
solutions for nonlinear operator equations and integral equations. Since then,
because of simplicity and usefulness, it has become a very popular tool in solving a
variety of problems such as control theory, economic theory, nonlinear analysis and
global analysis. Later, a huge amount of literature is witnessed on applications,
generalizations and extensions of this theorem. They are carried out by several
authors in different directions, e.g., by weakening the hypothesis, using different
setups.
Many mathematics problems require one to find a distance between two or more
objects which is not easy to measure precisely in general. There exist different
approaches to obtaining the appropriate concept of a metric structure. Due to the
need to construct a suitable framework to model several distinguished problems of
practical nature, the study of metric spaces has attracted and continues to attract the
interest of many authors. Over last few decades, a number of generalizations of
metric spaces have thus appeared in several papers, such as 2-metric spaces, G-
metric spaces, D -metric spaces, partial metric spaces and cone metric spaces. These
generalizations were then used to extend the scope of the study of fixed-point
theory. For more discussions of such generalizations. Sadeghi et al have introduced
the notion of an S-metric space and proved that this notion is a generalization of a
G-metric space and a D -metric space. Also, they have proved properties of S-metric
spaces and some fixed-point theorems for a self map on an S-metric space.
The Banach contraction principle is the most powerful tool in the history of fixed
point theory. Boyd and Wong extended the Banach contraction principle to the
nonlinear contraction mappings. We begin by briefly recalling some basic
definitions and results for S-metric spaces that will be needed in the sequel.
Definition 2.1.1: - Let X be a non-empty set, an S-metric on 𝑋 is a function
𝑆: 𝑋 3 → [0, +∞) that satisfies the following conditions, for each x, y, z, a ∈ X,
(1). 𝑆 (𝑥, 𝑦, 𝑧) ≥ 0,
(2). 𝑆 (𝑥, 𝑦, 𝑧) = 0 if and only if 𝑥 = 𝑦 = 𝑧,
(3). 𝑆 (𝑥, 𝑦, 𝑧) ≤ 𝑆 (𝑥, 𝑥, 𝑎) + 𝑆 (𝑦, 𝑦, 𝑎) + 𝑆 (𝑧, 𝑧, 𝑎),
for all 𝑥, 𝑦, 𝑧, 𝑎 ∈ 𝑋. The pair (𝑋, 𝑆) is called an S-metric space.
Immediate examples of such S-metric spaces are:
Example 2.1.2: - Let 𝑋 = 𝑅 𝑛 and ‖. ‖a norm on 𝑋, then
𝑆 (𝑥, 𝑦, 𝑧) = ‖ 𝑦 + 𝑧 − 2𝑥 ‖ + ‖ y − z ‖ is an S-metric on 𝑋.
Let 𝑋 be a nonempty set, 𝑑 is ordinary metric on 𝑋, then
𝑆 (𝑥, 𝑦, 𝑧) = 𝑑 (𝑥, 𝑧) + 𝑑 (𝑦, 𝑧)
is an S-metric on 𝑋. This S-metric is called the usual S-metric on 𝑋.
Definition 2.1.3: - Let (𝑋, 𝑆) be an S-metric space.
(i) A sequence {𝑥𝑛 } ⊂ 𝑋 converges to 𝑥 ∈ 𝑋 if 𝑆 (𝑥 𝑛 , 𝑥 𝑛 , 𝑥) → 0
as 𝑛 → +∞.That is, for each 𝜀 > 0, there exists 𝑛0 ∈ 𝑁 such
that for all 𝑛 ≥ 𝑛 0 we have 𝑆 (𝑥 𝑛 , 𝑥 𝑛 , 𝑥) < 𝜀. We write 𝑥 𝑛 → 𝑥
for brevity.
(ii) A sequence {𝑥 𝑛 } ⊂ X is a Cauchy sequence if 𝑆 (𝑥 𝑛 , 𝑥 𝑛 , 𝑥𝑚 ) →
0 𝑎𝑠 𝑛, 𝑚 → +∞. That is, for each 𝜀 > 0, there exists 𝑛 0 ∈ 𝑁
such that for all 𝑛, 𝑚 ≥ 𝑛0 we have 𝑆 (𝑥 𝑛 , 𝑥 𝑛 , 𝑥 𝑚 ) < 𝜀.
(iii) The S-metric space (𝑋, 𝑆) is complete if every Cauchy sequence is
a convergent sequence.
Definition 2.1.4: - Let (𝑋, 𝑆) be an S-metric space. For 𝑟 > 0 and 𝑥 ∈ 𝑋 we
define the open ball 𝐵𝑠 (𝑥, 𝑟) and closed ball 𝐵𝑠 [𝑥, 𝑟] with centre 𝑥 and radius 𝑟 as
follows respectively:
𝐵𝑠 (𝑥, 𝑟) = {𝑦 ∈ 𝑋: 𝑆 (𝑦, 𝑦, 𝑥) < 𝑟}, 𝐵𝑠 [𝑥, 𝑟] = {𝑦 ∈ 𝑋: 𝑆 (𝑥, 𝑥, 𝑦) ≤ 𝑟}.
Example 2.1.5: - Let 𝑋 = 𝑅 and 𝑆 (𝑥, 𝑦, 𝑧) = |𝑦 + 𝑧 − 2𝑥| + |𝑦 − 𝑧| for all
𝑥, 𝑦, 𝑧 ∈ 𝑅. Then
𝐵𝑠 (1, 2) = {𝑦 ∈ 𝑅: 𝑆 (𝑦, 𝑦, 1) < 2} = {𝑦 ∈ 𝑅: |𝑦 − 1| < 1}
= {𝑦 ∈ 𝑅: 0 < 𝑦 < 2} = (0, 2).
Lemma 1: - Let (𝑋, 𝑆) be an S-metric space. If 𝑟 > 0 and 𝑥 ∈ 𝑋, then the ball Bs
(𝑥, 𝑟) is open subset of 𝑋.
Lemma 2: - In an S-metric space, we have 𝑆 (𝑥, 𝑥, 𝑦) = 𝑆 (𝑦, 𝑦, 𝑥).
Proof. By third condition of S-metric, we have
𝑆 (𝑥, 𝑥, 𝑦) ≤ 𝑆 (𝑥, 𝑥, 𝑥) + 𝑆 (𝑥, 𝑥, 𝑥) + 𝑆 (𝑦, 𝑦, 𝑥) = 𝑆 (𝑦, 𝑦, 𝑥) … . . . . (1)
𝑆 (𝑦, 𝑦, 𝑥) ≤ 𝑆 (𝑦, 𝑦, 𝑦) + 𝑆 (𝑦, 𝑦, 𝑦) + 𝑆 (𝑥, 𝑥, 𝑦) = 𝑆 (𝑥, 𝑥, 𝑦), … . (2)
hence by (1) and (2), we get 𝑆 (𝑥, 𝑥, 𝑦) = 𝑆 (𝑦, 𝑦, 𝑥).
Lemma 3: - Let (𝑋, 𝑆) be an S-metric space. If sequence {𝑥𝑛 } in converges to 𝑥,
then 𝑥 is unique.
Lemma 4: - Let (𝑋, 𝑆) be an S-metric space. If sequence {𝑥𝑛 } in 𝑋 is converges to
𝑥, then {𝑥𝑛 } is a Cauchy sequence.
Lemma 5: - Let (𝑋, 𝑆) be an S-metric space. If there exist sequences {𝑥𝑛 } and {𝑦𝑛 }
such that lim 𝑥𝑛 = 𝑥 and lim 𝑦𝑛 = 𝑦, then lim 𝑆(𝑥 𝑛 , 𝑥 𝑛 , 𝑥 𝑚 ) = 𝑆 (𝑥, 𝑥, 𝑦).
𝑛→∞ 𝑛→∞ 𝑛→∞

Definition 2.1.6: - Let 𝑋 be a (non-empty) set, a 𝑏-metric on 𝑋 is a function

𝑑: 𝑋 2 → [0, +∞) if there exists a real number b ≥ 1 such that the following
conditions hold for all 𝑥, 𝑦, 𝑧 ∈ 𝑋,
(1) 𝑑 (𝑥, 𝑦) = 0 if and only if 𝑥 = 𝑦,
(2) 𝑑 (𝑥, 𝑦) = 𝑑 (𝑦, 𝑥),
(3) 𝑑 (𝑥, 𝑧) ≤ 𝑏 [𝑑 (𝑥, 𝑦) + 𝑑 (𝑦, 𝑧)].
The pair (𝑋, 𝑑) is called a b-metric space.
Proposition 1: - Let (𝑋, 𝑆) be an S-metric space and let 𝑑 (𝑥, 𝑦) = 𝑆 (𝑥, 𝑥, 𝑦), for
all 𝑥, 𝑦 ∈ 𝑋.
Then we have
(1) 𝑑 is a b-metric on 𝑋;
(2) 𝑥𝑛 → 𝑥 in (𝑋, 𝑆) if and only if 𝑥𝑛 → 𝑥 in (𝑋, 𝑑);
(3) {𝑥𝑛 } is a Cauchy sequence in (𝑋, 𝑆) if and only if {𝑥𝑛 } is a Cauchy sequence
in (𝑋, 𝑑).
Definition 2.1.7: - Let £ be the set of all continuous functions 𝑔: [0, ∞) 4 →
[0, +∞), satisfying the conditions:
(i) 𝑔 (1, 1, 1, 1) < 1,
(ii) 𝑔 is sub-homogeneous., 𝑔 (𝛼𝑥1 , 𝛼𝑥2 , 𝛼𝑥3 , 𝛼𝑥4 ) ≤ 𝛼𝑔 (𝑥1 , 𝑥2 , 𝑥3 , 𝑥4 ), for
all 𝛼 ≥ 0,
(iii) if 𝑥𝑖 , 𝑦𝑖 ∈ [0, +∞),𝑥𝑖 ≤𝑦𝑖 for 𝑖 = 1, . . . , 4 we have 𝑔 (𝑥1 , 𝑥2 , 𝑥3 , 𝑥4 ) ≤
𝑔 (𝑦1 , 𝑦3 , 𝑦4 ).
Example 2.1.8: - The function g (𝑥1 , 𝑥2 , 𝑥3 , 𝑥4 ) = k max {𝑥𝑖 }𝑖 = 0 for 𝑘 ∈ (0, 1)
is in class £.
𝑥3 + 𝑥4
Example 2.1.9: - The function g (𝑥1 , 𝑥2 , 𝑥3 , 𝑥4 ) = k max {𝑥1 , 𝑥2 , } for 𝑘 ∈
2
(0,1) is in class £.
Proposition 2: - If 𝑔 ∈ £ and 𝑢, 𝑣 ∈ [0, +∞] are such that 𝑢 ≤ 𝑔 (𝑣, 𝑣, 𝑣, 𝑢),
then 𝑢 ≤ ℎ𝑣, where ℎ = 𝑔 (1, 1, 1, 1).
Proof: - If 𝑣 < 𝑢, then 𝑢 ≤ 𝑔 (𝑣, 𝑣, 𝑣, 𝑢) ≤ 𝑔 (𝑢, 𝑢, 𝑢, 𝑢) < 𝑢𝑔 (1, 1, 1, 1) =
ℎ𝑢 < 𝑢, which is a contradiction. Thus 𝑢 ≤ 𝑣, which implies 𝑢 ≤
𝑔 (𝑣, 𝑣, 𝑣, 𝑢) ≤ 𝑔 (𝑣, 𝑣, 𝑣, 𝑣) < 𝑣𝑔 (1, 1, 1, 1) = ℎ𝑣.
Corollary 2.1.10: - Let (𝑋, 𝑆) be a complete S-metric space and 𝑇: 𝑋 → 𝑋 a
function such that for, all 𝑥, 𝑦, 𝑧, 𝑎 ∈ 𝑋,
𝑆 (𝑇𝑥 , 𝑇𝑦 , 𝑇𝑧 ) ≤ 𝐿𝑆 (𝑥, 𝑦, 𝑧),
where 𝐿 ∈ (0, 1 2 ). Then there exists a unique point 𝑢 ∈ 𝑋 such that 𝑇𝑢 = 𝑢.
2. Results: -
Now, we give our main result.
Theorem 2.1.11: - Let (𝑋, 𝑆) be a S- metric space and 𝑇: 𝑋 → 𝑋 be a function.
Suppose that there exists 𝑔 ∈ £ and 𝛼 ∈ (0, 1), such that 𝛼 (ℎ + 2) ≤ 1 where
ℎ = 𝑔 (1, 1, 1, 1). Suppose also that 𝛼𝑆 (𝑥, 𝑥, 𝑇𝑥) ≤ 𝑆 (𝑥, 𝑦, 𝑧) implies
𝑆 (𝑇𝑥, 𝑇𝑦, 𝑇𝑧) ≤ 𝑔 (𝑆 (𝑥, 𝑦, 𝑧), 𝑆 (𝑥, 𝑥, 𝑇𝑥), 𝑆 (𝑦, 𝑦, 𝑇𝑦), 𝑆 (𝑧, 𝑧, 𝑇𝑧)),
∀ 𝑥, 𝑦, 𝑧 ∈ 𝑋.
Then F(T) is non-empty set.
Proof: - Fix arbitrary 𝑥0 ∈ 𝑋 and let 𝑇𝑥0 = 𝑥1 . Since 𝛼𝑆 (𝑥0 , 𝑥0 , 𝑇𝑥0 ) <
𝑆 (𝑥0 , 𝑥0 , 𝑥1 ),
then by the hypothesis of the theorem and condition (iii) Definition 5, respectively,
we have
𝑆 (𝑥1 , 𝑥1 , 𝑇𝑥1 ) = 𝑆 ( 𝑇𝑥0 , 𝑇𝑥0 , 𝑇𝑥1 )
≤ 𝑔 (𝑆 (𝑥0 , 𝑥0 , 𝑥1 ), 𝑆 (𝑥0 , 𝑥0 , 𝑇𝑥0 ), 𝑆 (𝑥0 , 𝑥0 , 𝑇𝑥0 ), 𝑆 (𝑥1 , 𝑥1 , 𝑇𝑥1 ))
= 𝑔 (𝑆 (𝑥0 , 𝑥0 , 𝑥1 ), 𝑆 (𝑥0 , 𝑥0 , 𝑥1 ), 𝑆 (𝑥0 , 𝑥0 , 𝑥1 ), 𝑆 (𝑥1 , 𝑥1 , 𝑇𝑥1 ))
Then, by Proposition 2, we have 𝑆 (𝑥1 , 𝑥1 , 𝑇𝑥1 ) ≤ ℎ𝑆 (𝑥0 , 𝑥0 , 𝑥1 ).
Now let 𝑇𝑥1 = 𝑥2 . Since 𝛼𝑆 (𝑥1 , 𝑥1 , 𝑇𝑥1 ) < 𝑆 (𝑥1 , 𝑥1 , 𝑥2 ), by using and the
properties of the function 𝑔 we have
S (𝑥2 , 𝑥2 , 𝑇𝑥2 ) = S (𝑇𝑥1 , 𝑇𝑥1 , 𝑇𝑥2 )
≤ 𝑔 (𝑆 (𝑥1 , 𝑥1 , 𝑥2 ), 𝑆 (𝑥1 , 𝑥1 , 𝑇𝑥1 ), 𝑆 (𝑥1 , 𝑥1 , 𝑇𝑥1 ), 𝑆 (𝑥2 , 𝑥2 , 𝑇𝑥2 ))
= 𝑔 (𝑆 (𝑥1 , 𝑥1 , 𝑥2 ), 𝑆 (𝑥1 , 𝑥1 , 𝑥2 ), 𝑆 (𝑥1 , 𝑥1 , 𝑥2 ), 𝑆 (𝑥2 , 𝑥2 , 𝑇𝑥2 )).
Then, by Proposition 2, we have S (𝑥2 , 𝑥2 , 𝑇𝑥2 ) ≤ h S(𝑥1 , 𝑥1 , 𝑥2 ). In a similar way,
we can let 𝑇𝑥2 = 𝑥3 . So, we have
𝑆 (𝑥2 , 𝑥2 , 𝑥3 ) < ℎ𝑆 (𝑥1 , 𝑥1 , 𝑥2 ) < ℎ2 𝑆 (𝑥0 , 𝑥0 , 𝑥1 ).
By continuing this process, we obtain a sequence {𝑥𝑛 }𝑛 ≥ 1 in 𝑋 𝑠uch that
𝑥𝑛+1 = 𝑇𝑥𝑛 , which satisfies S (𝑥𝑛 , 𝑥𝑛 , 𝑇𝑥𝑛 ) ≤ h S(𝑥𝑛−1 , 𝑥𝑛−1 , 𝑥𝑛 )
and
𝑆 (𝑥𝑛 , 𝑥𝑛 , 𝑥𝑛+1 ) ≤ ℎ𝑛 𝑆 (𝑥0 , 𝑥0 , 𝑥1 ).
If 𝑥𝑚 = 𝑥𝑚+1 for some 𝑚 ≥ 1, then T has a fixed point.
Suppose that 𝑥𝑛 = 𝑥𝑛+1 , for all n ≥ 1. Repeated application of the triangle inequality
implies
S (𝑥𝑛 , 𝑥𝑛 , 𝑥𝑛+𝑚 ) ≤ 2S (𝑥𝑛 , 𝑥𝑛 , 𝑥𝑛+1 ) + S (𝑥𝑛+𝑚 , 𝑥𝑛+𝑚 , 𝑥𝑛+1 )
= 2S (𝑥𝑛 , 𝑥𝑛 , 𝑥𝑛+1 ) + S (𝑥𝑛+1 , 𝑥𝑛+1 , 𝑥𝑛+𝑚 )
≤ 2S (𝑥𝑛 , 𝑥𝑛 , 𝑥𝑛+1 ) + 2S (𝑥𝑛+1 , 𝑥𝑛+1 , 𝑥𝑛+2 ) + S (𝑥𝑛+𝑚 , 𝑥+𝑚𝑛 , 𝑥𝑛+2 )
≤ 2[S (𝑥𝑛 , 𝑥𝑛 , 𝑥𝑛+1 ) + S(𝑥𝑛 , 𝑥𝑛 , 𝑥𝑛+1 ) + · · · + S (𝑥𝑛+𝑚−1 , 𝑥𝑛+𝑚−1 , 𝑥𝑛+𝑚 )]
2ℎ𝑛
≤ 2 ∑𝑘=𝑚−1
𝑘=0 ℎ𝑘+𝑛 𝑆 (𝑥0 , 𝑥0 , 𝑥1 ) ≤ 𝑆 (𝑥0 , 𝑥0 , 𝑥1 ). So, we get
1−ℎ

lim 𝑆 (𝑥𝑛 , 𝑥𝑛 , 𝑥1 ) → ∞
𝑛→∞

and hence {𝑥𝑛 }n ≥1 is a Cauchy sequence in (𝑋, 𝑆). Regarding Definition 2, {𝑥𝑛
}n ≥1 is also a Cauchy sequence in (X, S).
Since (X, S) is a complete S- metric space, by Definition 2, (X, S) is also complete.
Thus {𝑥𝑛 }n ≥1 converges to a limit, say, x ∈ X, that is,
lim 𝑆 (𝑥𝑛 , 𝑥𝑛 , 𝑥) = 0.
𝑛→∞

It is easy to see that lim 𝑆 (𝑥1 , 𝑥1 , 𝑥2 ) = 0. Now, we claim that for each n ≥ 1 one
𝑛→∞
of the relations
𝛼𝑆 (𝑥𝑛 , 𝑥𝑛 , 𝑇𝑥𝑛 ) ≤ 𝑆 (𝑥𝑛 , 𝑥𝑛 , 𝑥)
𝑜𝑟 𝛼𝑆 (𝑥𝑛 , 𝑥𝑛 , 𝑇𝑥𝑛 ) ≤ 𝑆 (𝑥𝑛 , 𝑥𝑛 , 𝑥) holds.
If for some 𝑛 ≥ 1 we have
𝛼S (𝑥𝑛 , 𝑥𝑛 , 𝑇𝑥𝑛 ) > S (𝑥𝑛 , 𝑥𝑛 , 𝑇𝑥𝑛 )
and αS (𝑥𝑛 , 𝑥𝑛 , 𝑇𝑥𝑛 ) > S (𝑥𝑛 , 𝑥𝑛 , 𝑥),
then S (𝑥𝑛 , 𝑥𝑛 , 𝑥𝑛+1 ) ≤ 2S (𝑥𝑛 , 𝑥𝑛 , 𝑥) + S (𝑥𝑛+1 , 𝑥𝑛+1 , 𝑥)
< 2αS (𝑥𝑛 , 𝑥𝑛 , 𝑇𝑥𝑛 ) + αS (𝑥𝑛+1 , 𝑥𝑛+1 , 𝑇𝑥𝑛+1 )
= 2αS (𝑥𝑛 , 𝑥𝑛 , 𝑥𝑛+1 ) + αhS (𝑥𝑛 , 𝑥𝑛 , 𝑥𝑛+1 ).
This results in α (h + 2) > 1, which contradicts the initial assumption. Hence, our
claim is proved. Observe that by the assumption of the theorem, we have either
S (𝑇𝑥𝑛 , 𝑇𝑥𝑛 , 𝑇𝑥 ) ≤ g (S (𝑥𝑛 , 𝑥𝑛 , 𝑥), S (𝑇𝑥𝑛 , 𝑥𝑛 , 𝑥), S (𝑇𝑥𝑛 , 𝑥𝑛 , 𝑥), S (𝑇𝑥 , 𝑥𝑛 , 𝑥𝑛 )), Or
S (𝑇𝑥𝑛+1 , 𝑇𝑥𝑛+1 , 𝑇𝑥 ) ≤ g (S (𝑥𝑛+1 , 𝑥𝑛+1 , 𝑥),), S (𝑇𝑥𝑛 +1 , 𝑥𝑛+1 , 𝑥),
S (𝑇𝑥𝑛 +1 , 𝑥𝑛 , 𝑥), S (𝑇𝑥 , 𝑥𝑛+1 , 𝑥𝑛+1 )).
Therefore, one of the following cases holds.
Case (i). There exists an infinite subset I ⊆ N such that
𝑆 (𝑥𝑛+1 , 𝑥𝑛+1 , 𝑇𝑥)
= 𝑆 (𝑇 𝑥𝑛 , 𝑇 𝑥𝑛 , 𝑇𝑥)
≤ g (𝑆 (𝑥𝑛 , 𝑥𝑛 , 𝑥), 𝑆 (𝑥𝑛 , 𝑥𝑛 , 𝑥), 𝑆 (𝑇 𝑥𝑛 , 𝑥𝑛 , 𝑥), 𝑆 (𝑇𝑥, 𝑥𝑛 , 𝑥𝑛 ))
= g (S (𝑥𝑛 , 𝑥𝑛 , 𝑥), S (𝑥𝑛+1 , 𝑥𝑛+1 , 𝑥), S (𝑇 𝑥𝑛 , 𝑥𝑛 , 𝑥), S (𝑇 𝑥, 𝑥𝑛 , 𝑥𝑛 )).
for all n ∈ I.
Case (ii). There exists an infinite subset J ⊆ N such that
S (𝑥𝑛+2 𝑥𝑛+2 , 𝑇𝑥) = S (𝑇𝑥𝑛+1 , 𝑇𝑥𝑛+1 , 𝑇𝑥 )
≤ g (S (𝑥𝑛+1 , 𝑥𝑛+1 , 𝑥), S (𝑇𝑥𝑛+1 , 𝑥𝑛+1 , 𝑥), S (𝑇𝑥𝑛+1 , 𝑇𝑥𝑛+1 , 𝑇𝑥 ), S (𝑇𝑥, , 𝑥𝑛+1 , 𝑥𝑛+1 ))
= g (S (𝑥𝑛+1 , 𝑥𝑛+1 , 𝑥), S (𝑥𝑛+2 , 𝑥𝑛+1 , 𝑥), S (𝑥𝑛+2 , 𝑥𝑛+1 , 𝑥), S (𝑇𝑥, 𝑥𝑛+1 , 𝑥𝑛+1 )) )).
for all n ∈ I.
In case (i), taking the limit as n → +∞ we obtain
𝑆 (𝑥, 𝑥, 𝑇𝑥 ) ≤ 𝑔 (0, 0, 0, 𝑆 (𝑥, 𝑥, 𝑇𝑥 ))
Now by using Definition 5, Proposition 2, we have 𝑆 (𝑥, 𝑥, 𝑇𝑥 ) = 0, and thus 𝑥 =
𝑇𝑥 .
In case(ii), taking the limit as 𝑛 → ∞ we obtain 𝑆 (𝑥, 𝑥, 𝑇𝑥 ) ≤
𝑔 (0, 0, 0, 𝑆 (𝑥, 𝑥, 𝑇𝑥 ) Now by using definition 5, propositions 2, we have S (x,
x, 𝑇𝑥 ) = 0, and thus 𝑥 = 𝑇𝑥 This completes the proof.
Corollary 2.1.12: - Let (X, S) be a S- metric space and 𝑇: 𝑋 → 𝑋 be a function.
Suppose that there exists 𝑔 ∈ £ and 𝛼 ∈ (0, 1), such that 𝛼 (ℎ + 2) ≤ 1 where
ℎ = 𝑔 (1, 1, 1, 1). Suppose also that αS (y, y, 𝑇𝑦 ) ≤ S (x, y, z) implies
S (𝑇𝑥 , 𝑇𝑦 , 𝑇𝑧 ) ≤ g (S (x, y, z), S (x, x, 𝑇𝑥 ), S (y, y, 𝑇𝑦 ), S (z, z, 𝑇𝑧 ))
for all x, y, z ∈ X. Then F(T) is non-empty.
Corollary 2.1.13: - Let (X, S) be a complete S-metric space and T: X → X a
function such that for all x, y, z ∈ X, S ((𝑇𝑥 , 𝑇𝑦 , 𝑇𝑧 ) ≤ 𝐿𝑆 (𝑥, 𝑦, 𝑧), where 𝐿 ∈
(0, 1). Then there exists a unique point 𝑢 ∈ 𝑋 such that 𝑇𝑢 =u.
Proof: - Let g(, 𝑥2 , 𝑥3 , 𝑥4 ) 𝐿𝑥1 .
Corollary 2.1.14: - Let (X, S) be a complete S-metric space and 𝑇: 𝑋 → 𝑋 a
function such that for all x, y, z ∈ X, ), S (y, y, 𝑇𝑦 ), S (z, z, 𝑇𝑧 )}
where L ∈ (0, 1). Then there exists a unique point u ∈ X such that Tu =u.
Proof: - Let g (𝑥1 , 𝑥2 , 𝑥3 , 𝑥4 )) = L max { 𝑥1 , 𝑥2 , 𝑥3 , 𝑥4 )}.
 CHAPTER-3
COMMON FIXED POINT OF SINGLE-VALUED AND MULTI-
VALUED MAPPING IN S-METRIC SPACE

3.1 INTRODUCTION
Metric spaces are very important in mathematics. Generalized metric spaces can
be pointed out as b-metric, D-metric and fuzzy metric spaces. For more
considerations, see [2, 13, 4, 15]. In 2012, another generalized metric space called
S-metric space was introduced by Sedghi et al. [16]. In the setting of S-metric
space see, for example [5, 9, 12, 14], and the references therein. For application of
fixed points and common fixed points in different fields such as fractional calculus,
existence theory in fractional boundary value problems, see [1, 3, 6, 7, 8, 11]. In
this paper, some common fixed point theorems for single-valued and multi valued
mappings are proved in S-metric spaces by using a generalization of coincidence
point for pairs (f, F), (f, F) and (g, G) in which the mappings f and g are single-
valued and the mappings F and G are multi-valued mappings with values in S-
metric space (CB(X), SH), where SH is the Hausdorff S-metric. In section 2, some
preliminaries are recalled. In section 3, we state our main theorem. Section 4 is the
conclusions.

3.2 PRELIMINARIES
In this section some definitions, lemmas, theorems, and example are recalled.
Definition 3. 2.1. For non empty set 𝑋, 𝑆 ∶ 𝑋 3 → [0, ∞) is called an S-metric on
X if
(1): 𝑆(𝑥, 𝑦, 𝑧) = 0 𝑖𝑓𝑓 𝑥 = 𝑦 = 𝑧;
(2): 𝑆(𝑥, 𝑦, 𝑧) ≤ 𝑆(𝑥, 𝑥, 𝑎) + 𝑆(𝑦, 𝑦, 𝑎) + 𝑆(𝑧, 𝑧, 𝑎), for all 𝑥, 𝑦, 𝑧, 𝑎 ∈
𝑋. (𝑋, 𝑆) is called an S-metric space.
Example 3.2.2. (1): Assume 𝛼 ≥ 0 and 𝑋 = [𝛼, ∞). Define 𝑆 ∶ 𝑋 3 → [0, ∞)
0 𝑖𝑓 𝑥 = 𝑦 = 𝑧.
by 𝑆(𝑥, 𝑦, 𝑧) = { .. The mapping S is an S-metric on
𝑚𝑎𝑥 {𝑥, 𝑦, 𝑧} − 𝛼 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒.
X. We call it the max S-metric.
(2): Let 𝑋 = [0, ∞). Define 𝑆 ∶ 𝑋3 → [0, ∞) by 𝑆(𝑥, 𝑦, 𝑧) = ( 0 𝑖𝑓 𝑥 = 𝑦 =
𝑧; 𝑥 + 𝑦 + 2𝑧 otherwise. Then, 𝑆 is an S-metric on 𝑋.
Definition 3.2.3. In S-metric space (𝑋, 𝑆), assume that 𝑥 is an element of 𝑋, and
𝑟 > 0.
(1): An open ball 𝐵𝑠(𝑥, 𝑟) with center 𝑥 and radius 𝑟 is defined by
𝐵𝑠(𝑥, 𝑟) = {𝑦 ∈ 𝑋 ∶ 𝑆(𝑦, 𝑦, 𝑥) < 𝑟}.
(2): A sequence {𝑦𝑛 } in 𝑋 converges to 𝑦 if 𝑙𝑖𝑚𝑛 → ∞ 𝑆(𝑦𝑛, 𝑦𝑛, 𝑦) = 0. In
this case, we write 𝑦𝑛 → 𝑦 or 𝑙𝑖𝑚 𝑛→∞ 𝑦𝑛 = 𝑦.
(3): A sequence {𝑦𝑛 } in 𝑋 is called a Cauchy sequence if
𝑙𝑖𝑚𝑛,𝑚→∞ 𝑆(𝑦𝑛 , 𝑦𝑛 , 𝑦𝑚 ) = 0.
(4): (𝑋, 𝑆) is called complete if every Cauchy sequence converges.
(5): A subset 𝐴 of 𝑋 is called bounded if there exists ϵ > 0 such that for all a,
𝑏 ∈ 𝐴, 𝑆(𝑎, 𝑎, 𝑏) < 𝜖.
In (𝑋, 𝑆), we set 𝜏 = {𝐴 ⊆ 𝑋 ∶ 𝐴 𝑖𝑠 𝑎 𝑢𝑛𝑖𝑜𝑛 𝑜𝑓 𝑜𝑝𝑒𝑛 𝑏𝑎𝑙𝑙𝑠}. 𝜏 is a topology and
we set 𝐶𝐵(𝑋) = {𝐴 ⊆ 𝑋 ∶ 𝐴 𝑖𝑠 𝑛𝑜𝑛 𝑒𝑚𝑝𝑡𝑦 𝑐𝑙𝑜𝑠𝑒𝑑 𝑎𝑛𝑑 𝑏𝑜𝑢𝑛𝑑𝑒𝑑}.
Example 3.2.4. Consider 𝑋 = [0, ∞) with the max S-metric. Then, for 𝑎 ∈
( [0, 𝑟) 𝑖𝑓 𝑎 < 𝑟;
𝑋 𝑎nd 𝑟 > 0, we have: 𝐵𝑠 (𝑎, 𝑟) = {
{𝑎}𝑖𝑓 𝑎 ≥ 𝑟
Definition 3.2.5. Let (𝑋, 𝑆) be an S-metric space. We say 𝑆 is continuous if
𝑆(𝑥𝑛 , 𝑦𝑛 , 𝑧𝑛 ) → 𝑆(𝑥, 𝑦, 𝑧), whenever 𝑥𝑛 → 𝑥, 𝑦𝑛 → 𝑦, 𝑧𝑛 → 𝑧.
Example 3.2.6.
( 1 𝑖𝑓 (𝑥, 𝑦, 𝑧) = (1, 2, 3);
On 𝑋 = [0, ∞), define 𝑆(𝑥, 𝑦, 𝑧) = {
|𝑥 − 𝑧| + |𝑦 − 𝑧|𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒.
1
𝑆 is a S-metric on 𝑋 and it is not continuous. In fact, we have: 𝑥𝑛 = 1 + →
𝑛
2 3
1, 𝑦𝑛 = 2 + → 2, 𝑧𝑛 = 3 + → 3.
𝑛 𝑛

But 3 = lim 𝑆(𝑥𝑛 , 𝑦𝑛 , 𝑧𝑛 ) ≠ 𝑆(1, 2, 3) = 1.

𝑛→∞

Definition 3.2.7. Let (𝑋, 𝑆) be an S-metric space. We define 𝑆𝐻 ∶ 𝐶𝐵(𝑋)3 →

[0, ∞), 𝑏𝑦
𝑆𝐻 (𝐴, 𝐵, 𝐶) = 𝐻𝑆 (𝐴, 𝐶) + 𝐻𝑆 (𝐵, 𝐶), where 𝐻𝑆 (A, B) = max{ℎ𝑆 (A, B), ℎ𝑆 (B,
A)}, ℎ𝑆 (A, B) = sup{S(a, a, B) : a ∈ A} and S(a, a, B) = inf{S(a, a, b) : b ∈ B}.
For more information see
Theorem 3.2.8.𝑆𝐻 is an S-metric on CB(X). We call 𝑆𝐻 the Hausdorff S-metric on
𝐶𝐵(𝑋) generated by 𝑆.
Remark 3.2.9. In Example 2.2(1) let 𝑢 be a non decreasing continuous function
on 𝑋 = [𝛼, ∞) and let 𝐹(𝑥) = [𝛼, 𝑢(𝑥)]. We have: 𝐻𝑠(𝐹 𝑥, 𝐹 𝑦) = ( 𝑢(𝑦) −
𝛼 𝑖𝑓 𝑦 ≥ 𝑥; 𝑢(𝑥) − 𝛼 𝑖𝑓 𝑥 > 𝑦. Let (X, S) be an S-metric space. The set of all
non empty compact subsets of 𝑋 is denoted 𝑏𝑦 𝐾(𝑋).
Theorem 3.2.10. Let (𝑋, 𝑆) be a complete S-metric spaces.Then, (𝐾(𝑋), 𝑆𝐻 ) is a
complete S-metric space. The converse is also true. In fact, suppose that {𝑥𝑛 } is a
Cauchy sequence in (𝑋, 𝑆). By Theorem, we have 𝑙𝑖𝑚𝑛→∞ 𝑆𝐻 ({𝑥𝑛 }, {𝑥𝑛 }, {𝑥𝑚 }) =
2 𝑙𝑖𝑚𝑛 → ∞ 𝑆({𝑥𝑛 , 𝑥𝑛 , 𝑥𝑚 ) → 0. That is, {{𝑥𝑛 }} is a Cauchy sequence in
(𝐾(𝑋), 𝑆𝐻 ). So, by there exists 𝑥 ∈ 𝑋 such that {𝑥𝑛 } → {𝑥}. That is, 𝑥𝑛 → 𝑥.
Definition 3.2.11. Let (𝑋, 𝑆) be an S-metric space.
(1) The mappings 𝑓 ∶ 𝑋 → 𝑋 and 𝐹 ∶ 𝑋 → 𝐶𝐵(𝑋) are given. We say f and F
have a coincidence point at 𝑎 ∈ 𝑋 𝑖𝑓 𝑓(𝑎) ∈ 𝐹(𝑎), also, we say f and F have a
common fixed point at 𝑎 ∈ 𝑋 𝑖𝑓 𝑓(𝑎) = 𝑎 ∈ 𝐹(𝑎).
(2) The mapping 𝐹 ∶ 𝑋 → 𝐶𝐵(𝑋) is given. We say the mapping 𝑓 ∶ 𝑋 → 𝑋 is F-
weakly commuting at 𝑥 ∈ 𝑋 𝑖𝑓 𝑓(𝑓(𝑥)) ∈ 𝐹(𝑓(𝑥)).
Definition 3.2.12. Let (X, S) be an S-metric space. The mappings 𝑓, 𝑔 ∶ 𝑋 → 𝑋
and 𝐹, 𝐺 ∶ 𝑋 → 𝐶𝐵(𝑋) are given.
(1) We say the pair (𝑓, 𝐹) satisfies the limit property if there exist a sequence {xn}
in 𝑋, some 𝑡 ∈ 𝑋 and 𝐴 ∈ 𝐶𝐵(𝑋) such that lim 𝑓𝑥𝑛 = 𝑡 ∈ 𝐴 = lim 𝐹𝑥𝑛
𝑛→∞ 𝑛→∞

(2) We say The pairs (𝑓, 𝐹) and (𝑔, 𝐺) satisfy the common limit property if there
exist two sequences {𝑥𝑛 } and {{𝑦𝑛 } } in 𝑋, 𝑡 ∈ 𝑋, and 𝐴, 𝐵 ∈ 𝐶𝐵(𝑋) such that
lim 𝐹𝑥𝑛 = 𝐴, lim 𝐺𝑦𝑛 = 𝐵, lim 𝑓𝑥𝑛 = lim 𝑔𝑦𝑛 = 𝑡 ∈ 𝐴 ∩ 𝐵 .
𝑛→∞ 𝑛→∞ 𝑛→∞ 𝑛→∞

3.3 MAIN RESULT

In this section we state our mean theorem. Some examples and theorems follow up.
Theorem 3.3.1. Let 𝑓 be a self-mapping on an S-metric space (𝑋, 𝑆) and let 𝐹 be
a multi-valued mapping from 𝑋 into 𝐶𝐵(𝑋) such that
(1): The pair (𝑓, 𝐹) satisfies the limit property;
(2): For all two distinct elements 𝑥, 𝑦 ∈ 𝑋, 𝑆𝐻 (𝐹𝑥 , 𝐹𝑥 , 𝐹𝑦 ) < 𝑚𝑎𝑥
{𝑆(𝑓𝑥 , 𝑓𝑥 , 𝑓𝑦 ), 𝑆(𝑓𝑥 , 𝑓𝑥 , 𝐹𝑥 ) + 𝑆(𝑓𝑦 , 𝑓𝑦 , 𝐹𝑦 ) 𝑆(𝑓𝑥 , 𝑓𝑥 , 𝐹𝑦 ) +
𝑆(𝑓𝑦 , 𝑓𝑦 , 𝐹𝑥 )}……………………………………………………… (1)
If 𝑓𝑋 is a closed subset of 𝑋, then
(a): 𝑓 and 𝐹 have a coincidence point.
(b): 𝑓 and 𝐹 have a common fixed point provided that for each 𝑣 ∈ 𝐶(𝑓, 𝐹),
the mapping 𝑓 is 𝐹 −weakly commuting at 𝑣 and 𝑓𝑓𝑣 = 𝑓𝑣, where
𝐶(𝑓, 𝐹) = {𝑎 ∈ 𝑋 ∶ 𝑓 𝑎 ∈ 𝐹 𝑎}.
Proof. By assumption, there exist a sequence {𝑥𝑛 }in 𝑋, 𝑡 ∈ 𝑋 and 𝐴 ∈
𝐶𝐵(𝑋) such that lim 𝑓(𝑥𝑛) = 𝑡 ∈ lim 𝐹(𝑥𝑛) = 𝐴. Also there exists 𝑎 ∈ 𝑋 such
𝑛→∞ 𝑛→∞
that 𝑡 = 𝑓(𝑎). We put 𝑥 = 𝑥𝑛 and 𝑦 = 𝑎 in inequality (1) to obtain:
𝑆𝐻 (𝐹 𝑥𝑛 , 𝐹 𝑥𝑛 , 𝐹𝑎 ) < 𝑚𝑎𝑥{𝑆(𝑓 𝑥𝑛 , 𝑓𝑥𝑛 , 𝑓𝑎 ), 𝑆(𝑓 𝑥𝑛 , 𝑓𝑥𝑛 , 𝐹 𝑥𝑛 ) +
𝑆( 𝑓𝑎 , 𝑓𝑎 , 𝐹𝑎 ), 𝑆(𝑓 𝑥𝑛 , 𝑓𝑥𝑛 , 𝐹𝑎 ) + 𝑆(𝑓𝑎 , 𝑓𝑎 , 𝐹𝑥𝑛 )}.
It follows that
lim 𝑆𝐻 ( 𝐹𝑥𝑛 , 𝐹𝑥𝑛 , 𝐹𝑎 ) ) = 𝑆𝐻 (𝐴, 𝐴, 𝐹𝑎 ) ≤ 𝑆(𝑓𝑎 , 𝑓𝑎 , 𝐹𝑎 ).
𝑛→∞

By definition of 𝑆𝐻 we have
2𝑆(𝑓𝑎 , 𝑓𝑎 , 𝐹𝑎 ) ≤ 𝑆𝐻 (𝐴, 𝐴, 𝐹𝑎 ) ≤ 𝑆(𝑓𝑎 , 𝑓𝑎 , 𝐹𝑎 ).
That is, 𝑆(𝑓𝑎 , 𝑓𝑎 , 𝐹𝑎 ) = 0. So, 𝑓(𝑎) ∈ 𝐹(𝑎). This proves (a). To prove (b), by (a),
there exist 𝑡, 𝑎 ∈ 𝑋 𝑠uch that 𝑡 = 𝑓𝑎 ∈ 𝐹𝑎 . Since 𝑎 ∈ 𝐶(𝑓, 𝐹), So 𝑓𝑓𝑎 = 𝑓𝑎 and
ff 𝑎 ∈ 𝐹𝑓𝑎 . Hence, 𝑓 𝑡 = 𝑡 ∈ 𝐹 𝑡.
Example 3.3.2. Consider 𝑋 = [1, ∞) with the max S-metric. Define 𝑓 ∶ 𝑋 →
𝑥 2 +1
𝑋, 𝐹 ∶ 𝑋 → 𝐶𝐵(𝑋) as 𝑓(𝑥) = 𝑥 3 and 𝐹 (𝑥 ) = [1, ]respectively. The pair
2𝑥
(𝑓, 𝐹) satisfies the limit property. In fact, we have
1 1
lim 𝑓(1 + ) = 1 ∈ lim 𝐹(1 + ) …………………..(1)
𝑛→∞ 𝑛 𝑛→∞ 𝑛

For any two distinct elements 𝑥, 𝑦 ∈ 𝑋, the inequality (1) holds. For example, in
the case 𝑥 < 𝑦, by Remark 2.9 we have
𝑦2 + 1
𝑆𝐻 (𝐹𝑥 , 𝐹𝑥 , 𝐹𝑦 ) = 2𝐻𝑠 (𝐹𝑥 , 𝐹𝑦 ) = −2
𝑦
On the other hand, 𝑆(𝑓𝑥 , 𝑓𝑥 , 𝑓𝑦 ) = S(F 𝑥 3 , F 𝑥 3 , F 𝑦 3 ) = 𝑦 3 − 1.
So, 𝑆𝐻 (𝐹𝑥 , 𝐹𝑥 , 𝐹𝑦 ) < 𝑚𝑎𝑥{𝑆(𝑓𝑥 , 𝑓𝑥 , 𝑓𝑦 ), 𝑆(𝑓𝑥 , 𝑓𝑥 , 𝐹𝑥 ) +
𝑆(𝑓𝑦 , 𝑓𝑦 , 𝐹𝑦 ), 𝑆(𝑓𝑥 , 𝑓𝑥 , 𝐹𝑦 ) + 𝑆(𝑓𝑦 , 𝑓𝑦 , 𝐹𝑥 )}.
Hence, by Theorem 3.1, f and F have a coincidence point. That is, 𝑓(1) ∈ 𝐹(1).
Since 𝑓𝑓(1) = 𝑓(1) and f𝑓(1) ∈ 𝐹(1), 𝑓 and 𝐹 have common fixed point 1.
Theorem 3.3.3. Let 𝑓 be a self-mapping on a complete S-metric space (𝑋, 𝑆) and
2
let 𝐹 be a multi-valued mapping from 𝑋 into 𝐾(𝑋) and let 𝜆 ∈ (0, ) be a constant
3
such that for all two distinct members 𝑥, 𝑦 ∈ 𝑋:
𝑆𝐻 (𝐹𝑥 , 𝐹𝑥 , 𝐹𝑦 ) ≤ 𝜆 𝑚𝑎𝑥{𝑆(𝑓𝑥 , 𝑓𝑥 , 𝑓𝑦 ), 𝑆(𝑓𝑥 , 𝑓𝑥 , 𝐹𝑥 ), 𝑆(𝑓𝑦 , 𝑓𝑦 , 𝐹𝑦 ), 𝑆(𝑓𝑥 , 𝑓𝑥 , 𝐹𝑦 ) +
𝑆(𝑓𝑦 , 𝑓𝑦 , 𝐹𝑥 )} … … … … … … … … … … … … … … … … … … … … . . (2)
If fX is a closed subset of 𝑋 𝑎𝑛𝑑 𝐹𝑥 ⊆ 𝐾(𝑓𝑋), then
(a): 𝑓 and 𝐹 have a coincidence point;
(b): 𝑓 and 𝐹 have a common fixed point provided that for each 𝑣 ∈
𝐶(𝑓, 𝐹), 𝑓 is F-weakly commuting at v and 𝑓𝑓𝑣 = 𝑓𝑣, where
𝐶(𝑓, 𝐹) = {𝑎 ∈ 𝑋 ∶ 𝑓𝑎 ∈ 𝐹𝑎 }.
Proof. Since for each 𝑥0 ∈ 𝑋, ∅ ≠ 𝐹 𝑥0 ⊆ 𝑓𝑋, there exists 𝑥1 ∈ 𝑋 such that 𝑦1
= f𝑥1 ∈𝐹0 . So, by Lemma 3.11 there exists 𝑦2 = 𝑓𝑥2 ∈ 𝐹𝑥0 such that
1
𝑆(𝑦1 , 𝑦2 , 𝑦3 ) < 𝑆𝐻 (𝐹𝑥0 , 𝐹𝑥0 , 𝐹𝑥1 ) + 𝜆.
2

We obtain a sequence {𝑦𝑛 } such that 𝑦𝑛 = 𝑓𝑥𝑛 ∈ 𝐹𝑥𝑛−1 and

1
S(𝑦𝑛 , 𝑦𝑛 , 𝑦𝑛−1) < 𝑆𝐻 (𝐹𝑥𝑛−1 , 𝐹𝑥𝑛−1 , 𝐹𝑥𝑛 ) + 𝜆𝑛
2
𝜆
≤ max{S(𝑓𝑥𝑛−1 , 𝑓𝑥𝑛−1 , 𝑓𝑥𝑛 ) S(𝑓𝑥𝑛−1 , 𝑓𝑥𝑛−1 , 𝐹𝑥𝑛−1 )
2

S(𝑓𝑥𝑛 , 𝑓𝑥𝑛 , 𝐹𝑥𝑛 ), 𝑆(𝑓𝑥𝑛−1 , 𝑓𝑥𝑛−1 , 𝐹𝑥𝑛 ) + 𝑆(𝑓𝑥𝑛 , 𝑓𝑥𝑛 , 𝐹𝑥𝑛−1 )} + 𝜆𝑛 .

Set 𝑎𝑛 = 𝑆(𝑦𝑛 , 𝑦𝑛 , 𝑦𝑛+1). Since f𝑥𝑛 ∈ F 𝑥𝑛−1, S(𝑓𝑥𝑛 , 𝑓𝑥𝑛 , 𝐹𝑥𝑛−1) = 0. So,
𝜆
𝑎𝑛 < max{𝑎𝑛−1, S(𝑓𝑥𝑛−1 , 𝑓𝑥𝑛−1 , 𝐹𝑥𝑛−1 ), S(𝑓𝑥𝑛 , 𝑓𝑥𝑛 , 𝐹𝑥𝑛 ),
2

S(𝑓𝑥𝑛−1 , 𝑓𝑥𝑛−1 , 𝐹𝑥𝑛 )} +𝜆𝑛 .

We know
S(𝑓𝑥𝑛−1 , 𝑓𝑥𝑛−1 , 𝐹𝑥𝑛−1 ) ≤S(𝑓𝑥𝑛−1 , 𝑓𝑥𝑛−1 , 𝑓𝑥𝑛 ) =,𝑎𝑛−1 S(𝑓𝑥𝑛 , 𝑓𝑥𝑛 , 𝐹𝑥𝑛 ),)≤ 𝑎𝑛 ,
𝑆(𝑓𝑥𝑛−1 , 𝑓𝑥𝑛−1 , 𝐹𝑥𝑛 )}) ≤ 𝑆(𝑦𝑛−1, 𝑦𝑛−1, 𝑦𝑛+1) ≤ 2S(𝑦𝑛−1, 𝑦𝑛−1 , 𝑦𝑛 ) +
𝑆(𝑦𝑛+1, 𝑦𝑛+1, 𝑦𝑛 ) = 2𝑎𝑛−1 + 𝑎𝑛 .
𝜆 λ λn
So, 𝑎𝑛 < (2𝑎𝑛−1 + 𝑎𝑛 ) +𝜆𝑛 . That is, 𝑎𝑛 < λ 𝑎𝑛−1 + λ . By induction, we have
2 1− 1−
2 2

𝑛
λ λ λ λ
𝑎𝑛 < ( λ ) [𝑎0 + 1 + (1 − ) + (1 − )2 + ⋯ + (1 − )𝑛−1]
1− 2 2 2
2

𝑛
λ λ
≤ ( λ ) [𝑎0 + 1 + (𝑛 − 1) (1 − )].
1− 2
2

𝑛
λ λ
Set 𝑏𝑛 = ( λ ) [𝑎0 + 1 + (𝑛 − 1) (1 − )]
1− 2
2

𝑏𝑛+1 λ
Since lim = λ < 1, 𝑠𝑜, lim 𝑎𝑛 = 0
𝑛→∞ 𝑏𝑛 1−2 𝑛→∞

Now, we show that {𝑦𝑛 } is a Cauchy sequence.

For all 𝑚, 𝑛 ∈ 𝑁, 𝑚 > 𝑛, by Lemma 3.1
∞

𝑆(𝑦𝑛 , 𝑦𝑛 , 𝑦𝑚 ) ≤ 2 ∑ 𝑎𝑖 + 𝑎𝑚−1
𝑖=𝑛
2𝜆 𝜆 2𝜆 𝑖 λ 2𝜆
≤ 2 ∑∞
𝑖=𝑛( ) [𝑎0 + 1 + (i − 1)(1 − )] + ( ) [𝑎0 +1+(i-1) 1 − ]+(( )m−1[𝑎0 +
2−𝜆 2 2−𝜆 2 2−𝜆
λ
1 +(m-2)( 1 − )].
2

Therefore, lim 𝑆(𝑦𝑛 , 𝑦𝑛 , 𝑦𝑚 ) = 0. So, there exists 𝑢 ∈ 𝑋 such that lim 𝑓𝑥𝑛 =
𝑛→∞ 𝑛→∞
lim 𝑦𝑛 = 𝑢. Since 𝑓𝑋 is closed, there exists 𝑎 ∈ 𝑋 such that 𝑓 𝑎 = 𝑢. By
𝑛→∞
putting 𝑥 = 𝑥𝑛 , 𝑦 = 𝑥𝑚 in (2) :
𝑆𝐻 (𝐹𝑥𝑛 , 𝐹𝑥𝑛 , 𝐹𝑥𝑚 ) ≤ 𝜆 𝑚𝑎𝑥{𝑆(𝑓𝑥𝑛 , 𝑓𝑥𝑛 , 𝑓𝑥𝑚 ), 𝑆(𝑓𝑥𝑛 , 𝑓𝑥𝑛 , 𝐹𝑥𝑛 ),
S(𝑓𝑥𝑚 , 𝑓𝑥𝑚 , 𝐹𝑥𝑚 ), S(𝑓𝑥𝑚 , 𝑓𝑥𝑚 , 𝐹𝑥𝑚 ) + S(𝑓𝑥𝑚 , 𝑓𝑥𝑚 , 𝐹𝑥𝑛 )}……………… (3)
Also:
S(𝑓𝑥𝑛 , 𝑓𝑥𝑛 , 𝑓𝑥𝑛 ) ≤ S(𝑦𝑛 , 𝑦𝑛 , 𝑦𝑚 ); ……………………….(4)
𝑆(𝑓𝑥𝑚 , 𝑓𝑥𝑚 , 𝐹𝑥𝑚 ) ≤ 𝑆(𝑦𝑚 , 𝑦𝑚 , 𝑦𝑚+1); …………………...(5)
S𝑓𝑥𝑛 , 𝑓𝑥𝑛 , 𝐹𝑥𝑚 ) ≤S(𝑦𝑛 , 𝑦𝑛 , 𝑦𝑚+1); ………………...........(6)
S(𝑓𝑥𝑚 , 𝑓𝑥𝑚 , 𝐹𝑥𝑛 ) ≤S(𝑦𝑚 , 𝑦𝑚 , 𝑦𝑛+1). ………………….(7)
Relations (4 − 7) imply lim 𝑆𝐻 (𝐹𝑥𝑛 , 𝐹𝑥𝑛 , 𝐹𝑥𝑚 ) = 0.
𝑛→∞

So, {𝐹𝑥𝑛 } is a Cauchy sequence. Hence, by Theorem 2.10 there exists 𝐴 ∈ 𝐾(𝑋)
1
such that lim 𝐹𝑥𝑛 = A. Since 𝑆(𝑦𝑛 , 𝑦𝑛 , 𝐴) ≤ 𝑆𝐻 (𝐹𝑥𝑛−1 , 𝐹𝑥𝑛−1 , 𝐴). So, lim
𝑛→∞ 2 𝑛→∞
𝑆(𝑦𝑛 , 𝑦𝑛 , 𝐴) = 0. By Lemma 3.4 , for every 𝑛, there exists 𝛼𝑛 ∈ 𝐴 such that
𝑆(𝑦𝑛 , 𝑦𝑛 , 𝐴) = 𝑆(𝑦𝑛 , 𝑦𝑛 , 𝛼𝑛 ). Hence, lim 𝑆(𝑦𝑛 , 𝑦𝑛 , 𝛼𝑛 ) = 0. Lemma 2.1 ,
𝑛→∞
implies lim 𝛼𝑛 = u ∈ A. So (𝑓, 𝐹) satisfies the limit property. The rest of the proof
𝑛→∞
is similar to Theorem 3.1.
Example 3.3.4. Consider 𝑋 = [0, 1] with the max S-metric. For 𝑓𝑥 = 𝑥 3 and
𝑥3
𝐹 𝑥 = [0, ], the inequality (2) holds for all two distinct members 𝑥, 𝑦 ∈ 𝑋. For
8
𝑦3
example, in case 𝑥 < 𝑦, by Remark 2.9, (𝐹 𝑥 , 𝐹𝑦 ) = . Hence
4
1
𝑆𝐻 (𝐹𝑥 , 𝐹𝑥 , 𝐹𝑦 ) = 2𝐻𝑆 (𝐹𝑥 , 𝐹𝑦 ) ≤ 𝑆(𝑓𝑥 , 𝑓𝑥 , 𝑓𝑦 ) ≤
4
1
𝑚𝑎𝑥{𝑆(𝑓𝑥 , 𝑓𝑥 , 𝑓𝑦 ), 𝑆(𝑓𝑥 , 𝑓𝑥 , 𝐹𝑦 ), 𝑆(𝑓𝑥 , 𝑓𝑥 , 𝐹𝑦 ), 𝑆(𝑓𝑥 , 𝑓𝑦 , 𝐹𝑦 ) + 𝑆(𝑓𝑥 , 𝑓𝑥 , 𝐹𝑥 )}.
4

We have 𝑓𝑋 = 𝑋 𝑎𝑛𝑑 𝐹 𝑋 ⊆ 𝐾(𝑓𝑋). So all conditions of Theorem 3.3 are

satisfied. Hence, 𝑓 and 𝐹 have commn fixed point 0.
Theorem 3.3.5. Let 𝑓, 𝑔 be two self-mappings on an S-metric (𝑋, 𝑆) and let 𝐹, 𝐺
be two multi-valued mappings from 𝑋 into 𝐶𝐵(𝑋) such that
(1): The pairs (𝑓, 𝐹) and (𝑔, 𝐺) satisfy the common limit property;
(2): For all two distinct members 𝑥, 𝑦 ∈ 𝑋: 𝑆𝐻 (𝐹𝑥 , 𝐹𝑥 , 𝐺𝑦 ) <
𝑚𝑎𝑥{𝑆(𝑓𝑥 , 𝑓𝑥 , 𝑔𝑦), 𝑆(𝑓𝑥 , 𝑓𝑥 , 𝐹 𝑥) + 𝑆(𝑔𝑦, 𝑔𝑦, 𝐺𝑦 ), 𝑆(𝑓𝑥, 𝑓𝑥, 𝐺𝑦 ) +
𝑆(𝑔𝑦, 𝑔𝑦, 𝐹𝑥 )}………………... (8) If 𝑓𝑋, 𝑔𝑋 are closed subsets of 𝑋, then
(a): 𝑓 and 𝐹 have coincidence point;
(b): 𝑔 and 𝐺 have coincidence point;
(c): 𝑓 and 𝐹 have common fixed point provided that for each 𝑣 ∈ 𝐶(𝑓, 𝐹), 𝑓 be an
F-weakly commuting at 𝑣 and 𝑓𝑓𝑣 = 𝑓𝑣;
(d): 𝑔 and 𝐺 have common fixed point provided that for each 𝑣 ∈ 𝐶(𝑔, 𝐺), 𝑔 be a
G-weakly commuting at v and ggv = gv;
(e): If (c) and (d) hold, then 𝑓, 𝑔, 𝐹 and 𝐺 have common fixed point.
Proof. By assumption, there exist sequences {𝑥𝑛 }, {𝑦𝑛 } in X and 𝑢 ∈ 𝑋, 𝐴, 𝐵 ∈
𝐶𝐵(𝑋) such that lim 𝐹𝑥𝑛 = 𝐴, lim 𝐺𝑦𝑛 = 𝐵 𝑎𝑛𝑑 lim 𝑓𝑥𝑛 = 𝑙𝑖𝑚𝑛 → ∞ 𝑔𝑦𝑛 =
𝑛→∞ 𝑛→∞ 𝑛→∞
 𝑢 ∈ 𝐴 ∩ 𝐵. Assume that 𝑣, 𝑤 ∈ 𝑋 such that lim 𝑓𝑥𝑛 = 𝑓𝑣and 𝑙𝑖𝑚𝑛 →
𝑛→∞
∞ 𝑔𝑦𝑛 = 𝑔𝑤. We have 𝑓𝑣 = 𝑔𝑤 = 𝑢 ∈ 𝐴 ∩ 𝐵. To prove (a), we show, 𝑢 =
𝑓𝑣 ∈ 𝐹 𝑣. 𝑃𝑢𝑡 𝑥 = 𝑣 𝑎𝑛𝑑 𝑦 = 𝑦𝑛 in (8) and approach n to ∞, then
𝑆𝐻(𝐹 𝑣, 𝐹 𝑣, 𝐵) ≤ 𝑆(𝑓𝑣, 𝑓𝑣, 𝐹 𝑣). Since
𝑢 = 𝑓𝑣 ∈ 𝐵, 2𝑆(𝑓𝑣, 𝑓𝑣, 𝐹 𝑣) 6 𝑆𝐻(𝐹 𝑣, 𝐹 𝑣, 𝐵),
so, 𝑆(𝑓𝑣, 𝑓𝑣, 𝐹 𝑣) = 0. Therefore, 𝑢 = 𝑓𝑣 ∈ 𝐹 𝑣. Similarly, put 𝑥 = 𝑥𝑛, 𝑦 =
𝑤 in (8) and we have 𝑢 = 𝑔𝑤 ∈ 𝐺𝑤. Properties (c), (d), (e) are similar to
Theorem 3.1(b).
Example 3.3.6. Consider 𝑋 = [0, ∞) with the max S-metric. For 𝑓𝑥 =
𝑥 3 , 𝐹 𝑥 = [0, 𝑥 3 8 ] 𝑎𝑛𝑑 𝑔𝑥 = 𝑥 4 , 𝐺𝑥 = [0, 𝑥 4 8 ], the pairs (𝑓, 𝐹) and
(𝑔, 𝐺) satisfy the common limit property, in fact
1 1 1 1
lim 𝑓( ) = lim 𝑔( ) = 0, lim 𝐹( ) = lim 𝐺( ) = {0}.
𝑛→∞ 𝑛 𝑛→∞ 𝑛 𝑛→∞ 𝑛 𝑛→∞ 𝑛
For all distinct members 𝑥, 𝑦 ∈ 𝑋, the inquality (8) holds. For example, in case
𝑥 < 𝑦, first assume 𝑥 3 < 𝑦 4 . Since, for 𝑡 ∈ 𝐹𝑥 , 𝑆(𝑡, 𝑡, 𝐺𝑦 ) =
0, 𝑠𝑜, ℎ𝑆 (𝐹𝑥 , 𝐺𝑦 ) = 0. 𝐴𝑙𝑠𝑜 𝑠𝑖𝑛𝑐𝑒, 𝑓𝑜𝑟 𝑡 ∈ 𝐺𝑦 , we have
0 𝑖𝑓 𝑡 ∈ 𝐹𝑥 ,
S(t, t, 𝐹𝑥 ) = {
𝑡 𝑖𝑓 𝑡 ∈ 𝐺𝑦 − 𝐹𝑥
𝑦4
so, ℎ𝑠 (𝐺𝑦 , 𝐹𝑥 ) = sup{S(t, t, 𝐹𝑥 ) : t ∈ 𝐺𝑦 } = .
8
𝑦4 𝑦4
Hence,𝐻𝑆 (𝐹𝑥 , 𝐺𝑦 ) = and 𝑆𝐻 (𝐹𝑥 , 𝐹𝑥 , 𝐺𝑦 ) =.
4 4

On the other hand, we have 𝑆(𝑓𝑥, 𝑓𝑥, 𝑔𝑦) = 𝑥 3 , therefore the inequality (8)
holds. We have 𝑓𝑓0 = 𝑓0 = 0 ∈ 𝐹 𝑓0, 𝑎𝑛𝑑 𝑔𝑔0 = 𝑔0 = 0 ∈ 𝐺𝑔 0. So, all
conditions of Theorem 3.5 are satisfied. Therefore, f, g, F and G have common
fixed point. That is, f0 = g0 = 0 ∈ F0 ∩ G0 = {0}.
Corollary 3.3.7. If in Theorem 3.5 we set 𝐹 = 𝐺, 𝑎𝑛𝑑 𝑓 = 𝑔, Theorem 3.1
follows.
Theorem 3.3.8. Let f, g be two self-mappings on a complete S-metric space
(𝑋, 𝑆) and let 𝐹, 𝐺 be two multi-valued mappings from 𝑋 into 𝐾(𝑋) and let 𝜆 ∈
2
(0, ) be a constant such that for all two distinct members 𝑥, 𝑦 ∈ 𝑋 :
3

𝑆𝐻 (𝐹𝑥 , 𝐹𝑥 , 𝐺𝑦 ) 6 λ max{S(𝑓𝑥 , 𝑓𝑥 , 𝑔𝑦 ), S𝑔𝑦 , 𝑔𝑦 , 𝐺𝑦 ), S(,𝑓𝑥 , 𝑓𝑥 , 𝐺𝑦 ), S(,𝑓𝑥 , 𝑓𝑥 , 𝐺𝑦 +

S( 𝑔𝑦 , 𝑔𝑦 , 𝐹𝑥 )}……………….. (9)
If 𝑓𝑋, 𝑔𝑋 are closed subsets of 𝑋 and F 𝑋 ⊆ 𝐾(𝑔𝑋), 𝐺𝑋 ⊆ 𝐾(𝑓𝑋), then
(a): 𝑓 and 𝐹 have coincidence point;
(b): 𝑔 and 𝐺 have coincidence point;
(c): 𝑓 and 𝐹 have common fixed point provided that for each 𝑣 ∈ 𝐶(𝑓, 𝐹), 𝑓 be an
F-weakly commuting mapping at v and 𝑓𝑓𝑣 = 𝑓𝑣;
(d): g and G have common fixed point provided that for each v ∈ C(g, G), g is an
G-weakly commuting mapping at 𝑣 𝑎𝑛𝑑 𝑔𝑔𝑣 = 𝑔𝑣;
(e): If (c) and (d) hold, then 𝑓, 𝑔, 𝐹 𝑎𝑛𝑑 𝐺 have common fixed point.
Proof. For 𝑥0 ∈ 𝑋, there exists 𝑥1 ∈ X such that 𝑦1 = 𝑔𝑥1 ∈ 𝐹 𝑥0 . So, by Lemma
3.11, there exists 𝑦2 ∈ 𝐺𝑥1 such that
1
S(𝑦1 , 𝑦1 , 𝑦2 ) < 𝑆𝐻 (𝐹𝑥0 , 𝐹𝑥0 , 𝐺𝑥1 ) + λ.
2

There exists 𝑥2 ∈ X such that 𝑦2 = 𝑓𝑥2 ∈ 𝐺𝑥1 . So, there exists 𝑦3 ∈ 𝐹𝑥2 such that
1
S(𝑦2 , 𝑦2 , 𝑦3 , ,) < 𝑆𝐻 (𝐺𝑥1 , 𝐺𝑥1 , 𝐹𝑥2 ) + 𝜆2 .
2

We obtain a sequence {𝑦𝑛 } such that for every 𝑛 ≥ 1,

𝑦2𝑛 = 𝑓𝑥2𝑛 ∈ 𝐺𝑥2𝑛−1 , 𝑦2𝑛+1 = 𝑔𝑥2𝑛+1 ∈ 𝐹𝑥2𝑛 .
We have
1
𝑆(𝑦2𝑛 . 𝑦2𝑛 . 𝑦2𝑛+1) < 𝑆𝐻 (𝐺𝑥2𝑛−1 , 𝐺𝑥2𝑛−1 , 𝐹𝑥2𝑛 ) + 𝜆2 ;
2
1
S(y2n−1, y2n−1, y2n) < 𝑆𝐻 (F𝑥2𝑛−2,F) + 𝜆2𝑛 .
2

Set an = S(𝑦𝑛 , 𝑦𝑛 , 𝑦𝑛−1, ,). Similar to Theorem 3.3, it can be shown that
𝜆 𝜆
𝑎2𝑛 < (2𝑎𝑛−1 + 𝑎2𝑛 ) + 𝜆2𝑛 , 𝑎2𝑛−1 < (2𝑎2𝑛−2 + 𝑎2𝑛−1𝜆 ) + 𝜆2𝑛−1 .
2 2

So, for every 𝑛 ∈ 𝑁, we have

𝜆
𝑎𝑛 < (2𝑎𝑛−1 + 𝑎𝑛 ) + 𝜆 𝑛 . Similar to Theorem 3.3, we have lim 𝑎𝑛 =
2 𝑛→∞
0 𝑎𝑛𝑑 {𝑦𝑛 } is a Cauchy sequence. So, there exists 𝑢 ∈ 𝑋 such that lim 𝑦𝑛 = 𝑢.
𝑛→∞
Hence, lim 𝑓𝑥2𝑛 = lim 𝑔𝑥2𝑛+1 = 𝑢, and there exist 𝑎, 𝑏 ∈ 𝑋 such that 𝑓 𝑎 =
𝑛→∞ 𝑛→∞
𝑔𝑏 = 𝑢. To show {𝐹𝑥2𝑛 } is a Cauchy sequence, we have
𝑆𝐻 (𝐹𝑥2𝑛 , 𝐹𝑥2𝑛 , 𝐹𝑥2𝑚 ) ≤ 2𝑆𝐻 (𝐹𝑥2𝑛 , 𝐹𝑥2𝑛 , 𝐺𝑥2𝑛+1 ) +
𝑆𝐻(𝐹𝑥2𝑛 , 𝐹𝑥2𝑛 , 𝐹𝑥2𝑛+1 )……………………. (10)
By (9) we have:
𝑆𝐻 (𝐹𝑥2𝑛 , 𝐹𝑥2𝑛 , 𝐹𝑥2𝑚 )
≤ 𝜆 𝑚𝑎𝑥{𝑆(𝑓𝑥2𝑛 , 𝑓𝑥2𝑛 , 𝑔𝑥2𝑛+1 ), 𝑆(𝑓𝑥2𝑛 , 𝑓𝑥2𝑛 , 𝑓𝑥2𝑛 ), 𝑆(𝑔𝑥2𝑛 , 𝑔𝑥2𝑛 , 𝐺𝑥2𝑛+1 ), 𝑆(𝐹𝑥2𝑛 , 𝐹𝑥2𝑛 , 𝐹𝑥2𝑛+1 )
+ 𝑆(𝐹𝑥2𝑛 , 𝐹𝑥2𝑛 , 𝐹𝑥2𝑛+1 )}. 𝑆𝑜, lim 𝑆𝐻((𝐹𝑥2𝑛 , 𝐹𝑥2𝑛 , 𝐹𝑥2𝑛+1 ) = 0.
𝑛→∞

Similarly, we have lim 𝑆𝐻(𝐹𝑥2𝑚 , 𝐹𝑥2𝑚 , 𝐺𝑥2𝑛+1 ) = 0. It follows from (10)

𝑛,𝑚→∞
that 𝑙𝑖𝑚𝑛, 𝑚 → ∞ 𝑆𝐻(𝐹𝑥2𝑚 , 𝐹𝑥2𝑚 , 𝐺𝑥2𝑛+1 ) = 0. So, by Theorem 2.10, there exists
𝐴 ∈ 𝐾(𝑋) 𝑠uch that lim 𝐹𝑥2𝑛 = 𝐴. Now, assume that the left side of inequality
𝑛→∞
(9) is 𝑆(𝑓𝑥 , 𝑓𝑥 , 𝑔𝑦 ). Then, we have
𝑆𝐻 (𝐹𝑥 , 𝐹𝑥 , 𝐺𝑦 ) 6 𝜆 𝑆(𝑓𝑥, 𝑓𝑥, 𝑔𝑦)……………………………. (11)
Put 𝑥 = 𝑥2𝑛 , 𝑦 = 𝑥2𝑛+1 in (11) to obtain
𝑆𝐻 (𝐹𝑥2𝑛 , 𝐹𝑥2𝑛 , 𝐺𝑥2𝑛+1) ≤ 𝜆 𝑆(𝑓𝑥2𝑛 , 𝑓𝑥2𝑛 , 𝑔𝑥2𝑛+1).
So,
𝑙𝑖𝑚𝑛 → ∞ 𝑆𝐻(𝑓𝑥2𝑛 , 𝑓𝑥2𝑛 , 𝑔𝑥2𝑛+1) = 0.
Therefore, by Lemma 2.1, lim 𝐺𝑥2𝑛+1 = 𝐴. Similarly, if the left side of
𝑛→∞
inequality (9) is 𝑆(𝑓𝑥, 𝑓𝑥, 𝐹 𝑥) 𝑜𝑟 𝑆(𝑔𝑦, 𝑔𝑦, 𝐺𝑦), we have , lim 𝐺𝑥2𝑛+1 = 𝐴. On
𝑛→∞
the other hand, we have:
1
𝑆(𝑦2𝑛+1, 𝑦2𝑛+1 , 𝐴) ≤ 𝑆𝐻(𝐹𝑥2𝑛 , 𝐹𝑥2𝑛 , 𝐴).
2

So, lim 𝑆(𝑦2𝑛+1, 𝑦2𝑛+1, 𝐴) = 0. By Lemma 3.4 , for every 𝑛, there exists
𝑛→∞
𝛼2𝑛+1 ∈ 𝐴 such that,
𝑆(𝑦2𝑛+1, 𝑦2𝑛+1 , 𝐴) =
𝑆(𝑦2𝑛+1, 𝑦2𝑛+1 , 𝛼2𝑛+1). 𝑆𝑜, lim 𝑆(𝑦2𝑛+1, 𝑦2𝑛+1 , 𝛼2𝑛+1) = 0. Hence, by
𝑛→∞
Lemma 2.1 lim 𝛼2𝑛+1= u. So 𝑢 ∈ 𝐴. That is, (𝑓, 𝐹), (𝑔, 𝐺) satisfy the common
𝑛→∞
limit property. The rest of the proof is similar to Theorem 3.5.
Example 3. 3.9. In Example 3.6, for all distinct members 𝑥, 𝑦 ∈ 𝑋:
1
𝑆𝐻 (𝑓𝑥 , 𝑓𝑥 , 𝐺𝑦 ) = 𝑆(𝑓𝑥 , 𝑓𝑥 , 𝑔𝑦 )
4
1
≤ 𝑚𝑎𝑥{𝑆(𝑓𝑥 , 𝑓𝑥 , 𝑔𝑦 ), 𝑆(𝑓𝑥 , 𝑓𝑥 , 𝐺𝑦 ), 𝑆(𝑔𝑦 , 𝑔𝑦 , 𝐹𝑥 ),
4
𝑆(𝑓𝑥 , 𝑓𝑥 , 𝐺𝑦 ) + 𝑆(𝑔𝑦 , 𝑔𝑦 , 𝐹𝑥 )}.
So, all conditions of Theorem 3.8 are satisfied. That is, 𝑓, 𝑔, 𝐹 𝑎𝑛𝑑 𝐺 ℎave
common fixed point.
Corollary 3.3.10. If in Theorem 3.8 we set 𝐹 = 𝐺, 𝑎𝑛𝑑 𝑓 = 𝑔, Theorem 3.3
follows.
 CHAPTER-4
COMMON FIXED POINT THEROMS FOR THREE SELF MAPS
OF A COMPLETE S-METRIC SPACE

4.1. Introduction
In an attempt to generalize metric space Gahler introduced the notion of 2-metric
spaces while B.C.Dhage initiated the notion of D- metric spaces.Subsequently
several researchers have proved that most of their claims are not valid.As probable
modification to D- metric spaces,very recently Shaban Sedghi, Nabi Shobe and
Haiyun Zhou introduced D ∗ - metric spaces. In 2006 Zead Mustafa and Brailey
Sims have initiated G- metric spaces, while Shaban Sedghi, Nabi Shobe and
Abdelkrm Aliouche considered S-metric spaces in 2012.Of these three
generalizations, the S-metric space seen evinced interest in many researchers. The
purpose of this paper is to prove a common fixed point theorem for three self maps
of a S-metric space.Also as a consequence, we prove a common fixed point
theorem for three self maps of a complete S-metric space. Further we show that a
common fixed point theorem for three self maps of a metric space proved by
S.L.Singh and S.P.Singh pp 1584-1586) follows as a particular case of our
theorem. Now we recall some basic definitions and lemmas required in the sequel
in section 4.2 and establish main results in section 4.3.

4 .2. Preliminaries
Defination 4.2.1 Let X be a non empty set .By S-metric we mean a function 𝑆 ∶
𝑋3 → [0,∞) which satisfies the following conditions for each 𝑥, 𝑦, 𝑧, 𝑤 ∈ 𝑋
(a) 𝑆(𝑥, 𝑦, 𝑧) ≥ 0
(b) 𝑆(𝑥, 𝑥, 𝑦) = 0 𝑖𝑓 𝑎𝑛𝑑 𝑜𝑛𝑙𝑦 𝑖𝑓 𝑥 = 𝑦 = 𝑧.
(c) 𝑆(𝑥, 𝑦, 𝑧) ≤ 𝑆(𝑥, 𝑥, 𝑤) + 𝑆(𝑦, 𝑦, 𝑤) + 𝑆(𝑧, 𝑧, 𝑤)
In this case (𝑋, 𝑆) is called a S-metric space
Example 4.2.2. Let 𝑋 = 𝑅 and 𝑆 ∶ 𝑅 3 → [0, ∞) be defined by 𝑆(𝑥, 𝑦 , 𝑧) =
|𝑦 + 𝑧 − 2𝑥| + |𝑦 − 𝑧| 𝑓𝑜𝑟 𝑥, 𝑦, 𝑧 ∈ 𝑅, then (𝑋, 𝑆) is a S-metric space.
Example 4.2.3. Let (𝑋, 𝑑) be a metric space. Define 𝑆𝑑 ∶ 𝑋 3 → [0, ∞) by
𝑆𝑑(𝑥, 𝑦, 𝑧) = 𝑑(𝑥, 𝑦) + 𝑑(𝑦, 𝑧) + 𝑑(𝑧, 𝑥) then 𝑆𝑑 is a S-metric on 𝑋 and we call
this as the S-metric induced by 𝑑.
Remark 4. 2.4. It is shown in a S-metric space that
𝑆(𝑥, 𝑥, 𝑦) = 𝑆(𝑦, 𝑦, 𝑥) for all x, y ∈ X. Also we need the following notions given
in [6].
Definition 4.2.5. Let (𝑋, 𝑆) be an S-metric space. Let 𝑥 ∈ 𝑋 and 𝑟 > 0, then the
open ball with centre at 𝑥 and radius r is given by 𝐵(𝑥, 𝑟) = {𝑦 ∈ 𝑋 ∶
𝑆(𝑦, 𝑦, 𝑥) < 𝑟}
Remark 4. 2.6. Let (𝑋, 𝑆) be an S-metric space and 𝐴 ⊂ 𝑋.
(1) It has been proved in that 𝐵(𝑥, 𝑟) is an open set in 𝑋 and that the topology
generated by the open balls as a basis is a topology called the topology induced by
the S-metric on 𝑋.
(2) If for every 𝑥 ∈ 𝐴, there exists a 𝑟 > 0 such that 𝐵𝑠(𝑥, 𝑟) ⊂ 𝐴, then the
subset 𝐴 is called an open subset of 𝑋.
(3) A sequence {𝑥𝑛 } 𝑖𝑛 𝑋 said to converge to 𝑥 if 𝑆(𝑥𝑛 , 𝑥𝑛 , 𝑥) → 0; that is for
each 𝜀 > 0, there exists an 𝑛0 ∈ 𝑁 such that for all 𝑛 ≥ 𝑛0, we have S(𝑥𝑛 , 𝑥𝑛 , 𝑥)
< ε and we write this by lim 𝑥𝑛 = 𝑥 in this case.
𝑛→∞

(4)A sequence {𝑥𝑛 } in X is called a Cauchy sequence if to each ε > 0, there exists
n0 ∈ N such that S(𝑥𝑛 , 𝑥𝑛 , 𝑥) < ε for each 𝑛, 𝑚 ≥ 𝑛 0
(5) In it has been proved that if {𝑥𝑛 } is a sequence in S-metric space (X,S) that
converges to x is unique and that {𝑥𝑛 }is a Cauchy sequence.
(6) An S-metric space (𝑋, 𝑆) is said to be complete if every Cauchy sequence in it
converges.
Definition 4.2.7. Let (𝑋, 𝑆) be an S-metric space, If there exists sequences {𝑥𝑛 }
and {𝑦𝑛 } such that lim 𝑥𝑛 = x and lim 𝑦= y then lim n→∞ S(𝑥𝑛 , 𝑥𝑛 , 𝑦𝑛 )
𝑛→∞ 𝑛→∞
= 𝑆(𝑥, 𝑥, 𝑦), then we say that 𝑆(𝑥, 𝑦, 𝑧) is continuous in 𝑥 and 𝑦.
It is well known now that the commutativity of maps is generalized as follows
Defination.4.2.8. If 𝑔 and 𝑓 are self maps of a S-metric space (𝑋, 𝑆) such that for
every sequence {𝑥𝑛 } in 𝑋 with lim 𝑔𝑥𝑛 = lim 𝑓𝑥𝑛 = 𝑡 for some 𝑡 ∈ 𝑋 we
𝑛→∞ 𝑛→∞
have
lim 𝑆(𝑔𝑓𝑥𝑛 , 𝑔𝑓𝑥𝑛 , 𝑓𝑔𝑥𝑛 ) = 0 the𝑛 𝑔 and 𝑓 are said to be compatible
𝑛→∞

Trivially commuting self maps of a S-metric space are compatible but not
conversely. For example
Example 4.2.9. Let 𝑋 = [0,1] and 𝑆(𝑥, 𝑦, 𝑧) = |𝑥 − 𝑦| + |𝑦 − 𝑧| + |𝑧 − 𝑥| for
𝑥2 𝑥2
𝑥, 𝑦, 𝑧 ∈ 𝑋. Defining 𝑓 ∶ 𝑋 → 𝑋, 𝑔 ∶ 𝑋 → 𝑋 by gx = and f x = for 𝑥 ∈
2 3
𝑋 then it is easy to see that 𝑔, 𝑓 are compatible but not commutative.
Lemma 4.2.10. Let (𝑋, 𝑑) be any metric space and Sd be the S-metric induced by
d. For any sequence {𝑥𝑛 } in (𝑋, 𝑆𝑑 ), is a Cauchy sequence if and only if {𝑥𝑛 } is a
Cauchy sequence in (X,d).
Proof. First observe that d(𝑥, 𝑦) ≤ 𝑆𝑑 (𝑥, 𝑥, 𝑦) ≤ 2𝑑(𝑥, 𝑦) for all 𝑥, 𝑦 ∈ 𝑋. Now
the lemma follows immediately in view of the above inequality
Corollary 4.2.11. Let (𝑋, 𝑑) be any metric space and 𝑆 𝑑 be the S-metric on 𝑋.
Then (𝑋, 𝑆 𝑑 ) is a complete if and only if (𝑋, 𝑑) is complete
Proof. Follows from Lemma 2.1
Definition 4.2.12. If 𝑓, 𝑔 and h be self maps of a non empty set 𝑋 such that 𝑓(𝑋) ∪
𝑔(𝑋) ⊆ ℎ(𝑋), then for any 𝑥0 ∈ 𝑋,there is a sequence {𝑥𝑛 } in X such that
𝑓 𝑥2𝑛 = ℎ𝑥2𝑛+1 , 𝑔𝑥2𝑛+1 = ℎ𝑥2𝑛+2 𝑓𝑜𝑟 𝑛 ≥ 0 then{𝑥𝑛 } is called an associated
sequence of 𝑥0 relative to three self maps 𝑓, 𝑔 𝑎𝑛𝑑 ℎ.
The existence of an associated sequence of 𝑥0 relative to 𝑓, 𝑔 𝑎𝑛𝑑 ℎ is ensured.
In fact, if 𝑥 0∈ X then 𝑓𝑥0 ∈ f(X) and 𝑓(𝑋) ⊆ ℎ(𝑋) imply that there is 𝑎 𝑥1 ∈
𝑋 𝑠𝑢𝑐ℎ 𝑡ℎ𝑎𝑡𝑓𝑥0 = ℎ𝑥1 . Now ℎ𝑥1 ∈ g(X) and 𝑔(𝑋) ⊆ ℎ(𝑋) imply that there is a
𝑥2 ∈ X with 𝑔𝑥1 .. =ℎ𝑥2 .. Again 𝑓𝑥2 . ∈ 𝑓(𝑋) and 𝑓(𝑋) ⊆ ℎ(𝑋) then we get 𝑥3 ∈ X
with𝑓𝑥2 . = ℎ3 . and 𝑔𝑥3 . ∈ 𝑔(𝑋), 𝑔(𝑋) ⊆ ℎ(𝑋) gives 𝑔𝑥3 . = ℎ𝑥4 . for some 𝑥4 ∈ X.
Repeating this process, alternatively using the fact 𝑓(𝑋) ⊆ ℎ(𝑋) 𝑎𝑛𝑑 𝑔(𝑋) ⊆
ℎ(𝑋) we can find a sequence {𝑥𝑛 }with 𝑓𝑥2𝑛 = ℎ𝑥2𝑛+1and 𝑔𝑥2𝑛+1 = ℎ𝑥2𝑛+2for n ≥ 0.

It may be noted that for a given point𝑥0 ∈ X there may be more than one sequence
{𝑥𝑛 }} with the above condition. For example
Example 4.2.13. Suppose 𝑋 = 𝑅 with 𝑆(𝑥, 𝑦, 𝑧) = |𝑥 − 𝑦| + |𝑦 − 𝑧| + |𝑧 −
𝑥| 𝑓𝑜𝑟 𝑥, 𝑦, 𝑧 ∈ 𝑋. Define self maps f∶ 𝑋 → 𝑋, 𝑔 ∶ 𝑋 → 𝑋 and ℎ ∶ 𝑋 → 𝑋
𝑥2
by 𝑓 𝑥 = 𝑔𝑥 = 𝑎𝑛𝑑 ℎ(𝑥) = 𝑥 2 . Then as explained above we get a sequence
3
𝑥0
{xn} with 𝑓𝑥2𝑛 = ℎ𝑥2𝑛 +1 and 𝑔𝑥 for n ≥ o where each xn has two choices viz
√3
–𝑥0
or for 𝑛 ≥ 0. Hence to each x0 ∈ X, there are infinitely many associated
√3
sequences {𝑥𝑛 }.
Definition 4.2.14. A mapping 𝜑 ∶ [0, ∞) → [0, ∞) is said to be a contractive
modulus if φ(0) = 0 and 𝜑(𝑡) < 𝑡 𝑓𝑜𝑟 𝑡 > 0 Example 2.15. The mapping φ :
[0,∞) → [0,∞) defined by φ(t) = ct where 0 ≤ c < 1 is a contractive modulus.
4.3. Main Results
Theorem 4.3.1. Suppose 𝑓, 𝑔 and ℎ be three self maps of Smetric space (𝑋, 𝑆)
satisfying the conditions
(i) 𝑓(𝑋) ∪ 𝑔(𝑋) ⊆ ℎ(𝑋)
𝑆(𝑓 𝑥, 𝑓 𝑥, 𝑔𝑦) ≤ 𝜑 𝜆(𝑥, 𝑦) for all x, y ∈ X where φ is an upper semi continuous
contractive modulus and 𝜆(𝑥, 𝑦) =
𝑚𝑎𝑥{𝑆(ℎ𝑥, ℎ𝑥, ℎ𝑦), 𝑆(𝑓 𝑥, 𝑓 𝑥, ℎ𝑥), 𝑆(𝑔𝑦, 𝑔𝑦, ℎ𝑥), [𝑆(𝑓 𝑥, 𝑓 𝑥, ℎ𝑦) +
𝑆(𝑔𝑦, 𝑔𝑦, ℎ𝑥)]}
(iii) Either (𝑓, ℎ) 𝑜𝑟 (𝑔, ℎ) is compatible pair and
(iv) ℎ is continuous Further if

(v) There is a point 𝑥𝑜 ∈ X and an associated sequence {𝑥𝑛 } of 𝑥0 relative to the

three self maps such that the sequence 𝑓𝑥0 ,𝑔𝑥1 𝑓𝑥2 ,𝑔𝑥3 ,··· 𝑓 𝑥2𝑛 ,𝑔𝑥2𝑛+1,···
converge to some point 𝑧 ∈ 𝑋.
Then z is the unique common fixed point for 𝑓, 𝑔 𝑎𝑛𝑑 ℎ. Before proving the main
theorem, we establish a lemma which is noteworthy.
Lemma 4.3.2. Suppose 𝑓, 𝑔 and h be three self maps of S-metric space (𝑋, 𝑆)
satisfying the conditions (i),(ii),(iv) and (v) of the Theorem 3.1. Then for the
associated sequence {𝑥𝑛 } of 𝑥0 relative to f,g and h we have (a)
lim λ(hx2n , x2n+1)) = 𝑆(𝑧, 𝑧, ℎ𝑧) if (𝑓, ℎ) is compatible (b) 𝑙𝑖𝑚 𝑛 →
𝑛→∞
∞ λ(hx2n , x2n+1) = 𝑆(𝑧, 𝑧, ℎ𝑧) 𝑖𝑓 (𝑔, ℎ) is compatible
Proof. Since by (v), each of the sequence f x2n and gx2n+1 converges to z ∈ X and
𝑠𝑖𝑛𝑐𝑒 𝑓𝑥2𝑛 = ℎ𝑥2𝑛+1 𝑎𝑛𝑑 𝑔𝑥2𝑛+1 = ℎ𝑥2𝑛+2𝑓𝑜𝑟 𝑛 ≥ 0, we have
𝑓𝑥2𝑛 , 𝑔𝑥2𝑛+1, ℎ𝑥2𝑛+1, ℎ𝑥2𝑛+2 → 𝑧 𝑎𝑠 𝑛 → ∞ …………………….(3.1)
Now since h is continuous, we have
ℎ 𝑓𝑥2𝑛 → ℎ𝑧, ℎ 2 𝑥2𝑛 → ℎ𝑧 𝑎𝑠 𝑛 → ∞…………………………….. (3.2)
(a) If the pair (𝑓, ℎ) is compatible, we have
𝑙𝑖𝑚𝑛 → ∞ 𝑆(ℎ 𝑓𝑥2𝑛 , ℎ 𝑓𝑥2𝑛 , 𝑓 ℎ𝑥2𝑛 ) = 0………………………… …..(3.3)
since 𝑓𝑥2𝑛 , ℎ𝑥2𝑛 → 𝑧 𝑎𝑠 𝑛 → ∞ by 3.1 Now, in view of 3.2 and 3.3, we get
 𝑓ℎ𝑥2𝑛 → ℎ𝑧 𝑎𝑠 𝑛 → ∞ ……………………………………… (3.4)
Also from (ii) we have
λ(ℎ𝑥2𝑛 , 𝑥2𝑛+1) =
max{S(
1
ℎ 2 𝑥2𝑛 , ℎ 2 𝑥2𝑛 , ℎ𝑥2𝑛+1 ), 𝑆(𝑓 ℎ𝑥2𝑛 , 𝑓 ℎ𝑥2𝑛 , ℎ 2 𝑥2𝑛 ), 𝑆(𝑔𝑥2𝑛+1, 𝑔𝑥2𝑛+1, ℎ𝑥2𝑛+1 ), [𝑆(𝑓 ℎ
2
1) + 𝑆(𝑔𝑥2𝑛+1, 𝑔𝑥2𝑛+1, ℎ2 𝑥2𝑛 )]}
So that, in view of Remark 2.4 , we have
lim λ( ℎ𝑥2𝑛 , 𝑥2𝑛+1) =
𝑛→∞
𝑚𝑎𝑥{𝑆(ℎ𝑧, ℎ𝑧, 𝑧), 𝑆(ℎ𝑧, ℎ𝑧, ℎ𝑧), 𝑆(𝑧, 𝑧, 𝑧)), 1 2 [𝑆(ℎ𝑧, ℎ𝑧, 𝑧) + 𝑆(𝑧, 𝑧, ℎ𝑧)]} =
𝑆(𝑧, 𝑧, ℎ𝑧)
Proving part (a) of the lemma
(b) If the pair (g,h) is compatible, we have by 3.1
𝑙𝑖𝑚 𝑛 → ∞ 𝑆(ℎ𝑔𝑥2𝑛 + 1, ℎ𝑔𝑥2𝑛+1, 𝑔ℎ𝑥2𝑛 ) = 0 ………………………….(3.5)
Also since ℎ is continuous, we have again by 3.1, that
ℎ2 𝑥2𝑛+1 → ℎ𝑧 𝑎𝑛𝑑 ℎ𝑔𝑥2𝑛+1 → ℎ𝑧 𝑎𝑠 𝑛 → ∞ …………………………….(3.6)
Now, in view of 3.5 and 3.6, we get
𝑔ℎ𝑥2𝑛+1 → ℎ𝑧 as 𝑛 → ∞ ………………………………..(3.7)
Now, from (ii) we have
𝜆(𝑥2𝑛, ℎ𝑥2𝑛 + 1) = 𝑚𝑎𝑥{𝑆(ℎ𝑥2𝑛 , ℎ𝑥2𝑛 , ℎ2 𝑥2𝑛+1),………………. (3.8)
S(f 𝑥2𝑛 , f𝑥2𝑛 ,h𝑥2𝑛 ),S(gh𝑥2𝑛+11,gh𝑥2𝑛+1,………………………. (3.9)
1
ℎ 2 𝑥2𝑛+1), [S(f 𝑥2𝑛 , f 𝑥2𝑛 ,ℎ 2 𝑥2𝑛+1 )+ …….(3.10)
2

S(𝑔ℎ𝑥2𝑛 + 1, 𝑔ℎ𝑥2𝑛+1, ℎ𝑥2𝑛 )]} …………….(3.11)

Now, letting n → ∞ in 3.8 and using the continuity of 𝑆(𝑥, 𝑦, 𝑧) in x and y
3.1, 3.6, 3.7 we get
𝑙𝑖𝑚 𝑛 → ∞ 𝜆(𝑥2𝑛 , ℎ𝑥2𝑛+1 )
= 𝑚𝑎𝑥{𝑆(𝑧, 𝑧, ℎ𝑧), 𝑆(𝑧, 𝑧, 𝑧), 𝑆(ℎ𝑧, ℎ𝑧, ℎ𝑧)), 1 2 [𝑆(𝑧, 𝑧, ℎ𝑧)
+ 𝑆(ℎ𝑧, ℎ𝑧, 𝑧)]} = 𝑆(𝑧, 𝑧, ℎ𝑧)
Proving part (b) of the lemma
Proof of the Theorem 4.3.1 In this section we first prove the existence of a
common fixed point in one of the two cases of the condition (iii) and the other case
follows similarly with appropriate changes. Here we prove in case the pair (𝑓, ℎ) is
compatible. Now from (ii), we have
S(f h𝑥2𝑛 , f h𝑥2𝑛 ,g𝑥2𝑛+11) ≤ φ λ(h𝑥2𝑛 , 𝑥2𝑛+1)………….(3.12)
in which on letting 𝑛 → ∞ and using Lemma 3.2 and the continuity of 𝑆(𝑥, 𝑦, 𝑧)
in x and y we get
𝑆(ℎ𝑧, ℎ𝑧, 𝑧) ≤ 𝜑 𝑆(ℎ𝑧, ℎ𝑧, 𝑧) ………………… (3.13)
And this leads to a contradiction if ℎ𝑧 ≠ 𝑧. Therefore ℎ𝑧 = 𝑧. Again from the
condition (ii), we have
𝑆(𝑓 𝑧, 𝑓 𝑧, 𝑔𝑥2𝑛+1) ≤ 𝜑 𝜆(𝑧, 𝑥2𝑛+1) …………………………..(3.14)
But
λ(z, 𝑥2𝑛+1)=𝑚𝑎𝑥{𝑆(ℎ𝑧, ℎ𝑧, ℎ𝑥2𝑛+1)𝑆(𝑓 𝑧, 𝑓 𝑧, ℎ𝑧), 𝑆(𝑔𝑥2𝑛+1, 𝑔𝑥2𝑛+1, ℎ𝑥2𝑛+1)),
1
(𝑓 𝑧, 𝑓 𝑧, ℎ𝑥2𝑛+1) + 𝑆(𝑔𝑥2𝑛+1, 𝑔𝑥2𝑛+1, ℎ𝑧)]}
2
In which on letting 𝑛 → ∞, we find lim 𝑛 → ∞ 𝜆(𝑧, 𝑥2𝑛+1)
= 𝑚𝑎𝑥{𝑆(ℎ𝑧, ℎ𝑧, 𝑧), 𝑆(𝑓 𝑧, 𝑓 𝑧, 𝑧), 𝑆(𝑧, 𝑧, 𝑧), 1 2 [𝑆(𝑓 𝑧, 𝑓 𝑧, 𝑧) + 𝑆(𝑧, 𝑧, 𝑧)
= 𝑚𝑎𝑥{0, 𝑆(𝑓 𝑧, 𝑓 𝑧, 𝑧),0, 1 2 [𝑆(𝑓 𝑧, 𝑓 𝑧, 𝑧) + 0]} = 𝑆(𝑓 𝑧, 𝑓 𝑧, 𝑧)
Now, letting 𝑛 → ∞ in 3.14, we get by the upper semicontinuity of φ, that
𝑆(𝑓 𝑧, 𝑓 𝑧, 𝑧) ≤ 𝜑 𝑆(𝑓 𝑧, 𝑓 𝑧, 𝑧) ………………………. (3.15)
which leads to a contradiction if f 𝑧 ≠ 𝑧. Therefore 𝑓 𝑧 = 𝑧. Now, again from
the condition (ii), we have
𝑆(𝑓𝑥2𝑛 , 𝑓 𝑥2𝑛 𝑔𝑧) ≤ 𝜑 𝜆(𝑥2𝑛 , 𝑧) …………… (3.16)
But 𝜆(, 𝑧) =
𝑚𝑎𝑥{𝑆(ℎ𝑥2𝑛 , ℎ𝑥2𝑛 , ℎ𝑧), 𝑆(𝑓 𝑥2𝑛 𝑓𝑥2𝑛 , ℎ𝑥2𝑛 ), 𝑆(𝑔𝑧, 𝑔𝑧, ℎ𝑧)), 1 2 [𝑆(𝑓𝑥2𝑛 , 𝑓𝑥2𝑛 , ℎ𝑧) +
𝑆(𝑔𝑧, 𝑔𝑧, ℎ𝑥2𝑛 )]}
so that
𝑙𝑖𝑚 𝑛 → ∞ 𝜆(𝑥2𝑛 , 𝑧)
1
= 𝑚𝑎𝑥{𝑆(𝑧, 𝑧, ℎ𝑧), 𝑆(𝑧, 𝑧, ℎ𝑧), 𝑆(𝑔𝑧, 𝑔𝑧, 𝑧), [𝑆(𝑧, 𝑧, 𝑧)
2
+ 𝑆(𝑔𝑧, 𝑔𝑧, 𝑧)]}
= 𝑆(𝑔𝑧, 𝑔𝑧, 𝑧) = 𝑆(𝑧, 𝑧, 𝑔𝑧)
Now, letting 𝑛 → ∞ in 3.16, we get by the upper semicontinuity of φ, that
𝑆(𝑧, 𝑧, 𝑔𝑧) ≤ 𝜑 𝑆(𝑧, 𝑧, 𝑔𝑧) ………………. (3.17) and this leads to a contradiction
if 𝑔𝑧 ≠ 𝑧. Therefore 𝑔𝑧 = 𝑧. Hence 𝑧 = 𝑓 𝑧 = 𝑔𝑧 = ℎ𝑧
Showing that z is a common fixed point of f,g and h. We now prove the uniqueness
of the common fixed point.If possible let z 0 be another common fixed point of f,g
and h. Then from condition (ii), we have
𝑆(𝑧, 𝑧, 𝑧 0 ) = 𝑆(𝑓 𝑧, 𝑓 𝑧, 𝑔𝑧0 ) ≤ 𝜑 𝜆(𝑧, 𝑧 0 ) ……….. (3.18)
where
𝜆(𝑧, 𝑧 ′ )
1
= 𝑚𝑎𝑥{𝑆(ℎ𝑧, ℎ𝑧, ℎ𝑧 ′ ), 𝑆(𝑓 𝑧, 𝑓 𝑧, ℎ𝑧), 𝑆(𝑔𝑧 ′ , 𝑔𝑧 ′ , ℎ𝑧 ′ )),
[𝑆(𝑓 𝑧, 𝑓 𝑧, ℎ𝑧 ′ )
2
+ 𝑆(𝑔𝑧 , 𝑔𝑧 , ℎ𝑧)]} = 𝑚𝑎𝑥{𝑆(𝑧, 𝑧, 𝑧 ),0,0, 1 2 [𝑆(𝑧, 𝑧, 𝑧 ) + 𝑆(𝑧 ′ , 𝑧 ′ , 𝑧)]}
′ ′ ′ ′

= 𝑆(𝑧, 𝑧, 𝑧 ′ )
Therefore 3.18 gives
𝑆(𝑧, 𝑧, 𝑧 ′ ) ≤ 𝜑 𝑆(𝑧, 𝑧, 𝑧 ′ ) …………… (3.19)
which leads to a contradiction if 𝑧 ≠ 𝑧 ′ . Hence 𝑧 is the unique common fixed
point of 𝑓, 𝑔 𝑎𝑛𝑑 ℎ.
Hence the Theorem 3.1 is completely proved.
Theorem 4.3.3. Suppose (𝑋, 𝑆) is a S-metric space satisfying the conditions (i) to
(iv)of Theorem 3.1. Further if (v)’ (𝑋, 𝑆) is complete
Then f,g and h have a unique common fixed point 𝑧 ∈ 𝑋.
Before proving the main theorem, we establish an essential lemma.
Lemma 4.3.4. Suppose (X,S) is a S-metric space (X,S) and f,g and h be three self
maps of X such that
(i) 𝑓(𝑋) ∪ 𝑔(𝑋) ⊆ ℎ(𝑋)
1
(ii) 𝑆(𝑓 𝑥, 𝑓 𝑥, 𝑔𝑦) ≤ 𝑐 𝜆(𝑥, 𝑦) 𝑓𝑜𝑟 𝑎𝑙𝑙 𝑥, 𝑦 ∈ 𝑋 𝑤ℎ𝑒𝑟𝑒 0 ≤ 𝑐 <
2
(iii) 𝑎𝑛𝑑 𝜆(𝑥, 𝑦) =
1
𝑚𝑎𝑥{2𝑆(ℎ𝑥, ℎ𝑥, ℎ𝑦), 𝑆(𝑓 𝑥, 𝑓 𝑦, ℎ𝑥), 𝑆(𝑔𝑦, 𝑔𝑦, ℎ𝑥), [𝑆(𝑓 𝑥, 𝑓 𝑥, ℎ𝑦) +
2
𝑆(𝑔𝑦, 𝑔𝑦, ℎ𝑥)]} and
(iv) (𝑋, 𝑆) is complete Then for any x0 ∈ X and for any of its associated
sequence {𝑥𝑛} relative to the three self maps, the sequence
𝑓 𝑥0 , 𝑔𝑥1 , 𝑓 𝑥2 , 𝑔𝑥3 ,··· 𝑓 𝑥4 , 𝑔𝑥5 ,··· converge to some point 𝑧 ∈ 𝑋.
Proof. Suppose f,g and h be self maps of a S-metric space (𝑋, 𝑆) for which
condition (i) and (ii) hold.
Let 𝑥0 ∈ X and {𝑥𝑛 } be an associated sequence of x0 relative to the three self
maps.Then since 𝑓 𝑥2𝑛 = ℎ𝑥2𝑛+1 and g = h𝑥2+2𝑛 for 𝑛 ≥ 0.
Note that 𝜆(𝑥2𝑛 , 𝑥2𝑛+11) = 𝑚𝑎𝑥{2𝑆(ℎ𝑥2𝑛 , ℎ𝑥2𝑛 , ℎ𝑥2𝑛+2),
1
S(f𝑥2𝑛 , f𝑥2𝑛 ,h𝑥2𝑛 ),S(g𝑥2𝑛+1,g𝑥2𝑛+1,h𝑥2𝑛+1), [S(f𝑥2𝑛 , f𝑥2𝑛 ,h𝑥2+1𝑛 )
2
+S(g𝑥2𝑛+1,g𝑥2𝑛+1,h𝑥2𝑛 )]} =
max{2S(h𝑥2𝑛 ,h𝑥2𝑛 ,h𝑥2𝑛+1 ),S(h𝑥2𝑛+1 1,h𝑥2𝑛+1,h𝑥2𝑛 ), S(h𝑥2𝑛 +2,h𝑥2𝑛+2,h𝑥2𝑛+1),
1
[S(h𝑥2𝑛+1,h𝑥2𝑛+1,h𝑥2𝑛+1) + S(h𝑥2𝑛+2,h𝑥2𝑛+2,h𝑥2𝑛 )]} =
2
max{2S(h𝑥2𝑛 ,h𝑥2𝑛 ,h𝑥2𝑛+1 ),S(h𝑥2𝑛+2 ,h𝑥2𝑛+2,h𝑥2𝑛+1),
2S(h𝑥2𝑛+1+2,h, 𝑥2𝑛+2h𝑥2𝑛 )}
andsince
1 1
S(h𝑥2𝑛+2,h𝑥2𝑛+2,h𝑥2𝑛 ) ≤ S(h𝑥2𝑛+2,h𝑥2𝑛+2,h𝑥2𝑛+1) + S(hx2n,hx2n,hx2n+1)
2 2
and α +β ≤ 2max(α,β) for any α ≥ 0,β ≥ 0,
we get
1 1
S(h𝑥2𝑛+1,h𝑥2𝑛+1,h𝑥2𝑛 ) ≤ 2max{S(h𝑥2𝑛+2,h𝑥2𝑛+2,h𝑥2𝑛+1),
2 2
S(h𝑥2𝑛+2,h𝑥2𝑛+2,h𝑥2𝑛 )}
= max{2S(h𝑥2𝑛+2+2,h𝑥2𝑛+2,h𝑥2𝑛 ),S(h𝑥2𝑛 ,h𝑥2𝑛 ,h𝑥2𝑛+1)}
It follows that λ(𝑥2𝑛 , 𝑥2𝑛+1) ≤ max{2S(h𝑥2𝑛 ,h𝑥2𝑛 ,h𝑥2𝑛+1),
2S(h𝑥2𝑛+2,h𝑥2𝑛+2,h𝑥2𝑛+2 )}
Now by (ii) and ?? we have S(h𝑥2𝑛+1,h𝑥2𝑛+1,h𝑥2𝑛+2) = S(f𝑥2𝑛 , f𝑥2𝑛 ,g𝑥2𝑛+1) ≤
cλ(𝑥2𝑛 , 𝑥2𝑛+1) ≤ 2c max{S(h𝑥2𝑛 ,h𝑥2𝑛 ,h𝑥2𝑛+1), S(h𝑥2𝑛+2,h𝑥2𝑛+2,h𝑥2𝑛+1)}
therefore, in view of Remark 2.4 and the fact 0 < 2c < 1, we get
S(h𝑥2𝑛+1,h𝑥2𝑛+1,h𝑥2𝑛+2) ≤ 2cS(h𝑥2𝑛 ,h𝑥2𝑛 ,h𝑥2𝑛+1)……………….. (3.20)
Similarly we can prove that
S(h𝑥2𝑛 ,h𝑥2𝑛 h𝑥2𝑛+1) ≤ 2cS(h𝑥2𝑛−1,h𝑥2𝑛−1,h𝑥2𝑛 )…………….. (3.21)
From 3.20 and 3.21 we have for any m ≥ 1 that
S(h𝑥𝑚 ,h𝑥𝑚 ,h𝑥𝑚+1) ≤ 2cS(h𝑥𝑚−1,h𝑥𝑚−1,h𝑥𝑚 )…………… (3.22)
which on repeated application yields
S(h𝑥𝑚 ,h𝑥𝑚 ,h𝑥𝑚+1) ≤ 2cS(h𝑥𝑚−1,h𝑥𝑚−1,h𝑥𝑚 ) ≤ 4c 2 S(h𝑥𝑚−2,h𝑥𝑚−1,h𝑥𝑚−11) ···
··· ≤ (2c) m S(h𝑥1 ,h𝑥1 ,h𝑥0 )
which imply that the sequence {h𝑥𝑛 } and hence
𝑓𝑥0 , 𝑔𝑥1 , 𝑓 𝑥2 , 𝑔𝑥3 ,··· 𝑓𝑥2𝑛 𝑔𝑥2𝑛+1,··· is a Cauchy sequence in the complete
metric space (𝑋, 𝑆) and therefore converges to a point say 𝑧 ∈ 𝑋, proving the
lemma.
Remark 4.3.5. The converse of the above lemma is not true. That is, suppose f,g
and h are self maps of a S-metric space (𝑋, 𝑆) satisfying conditions (i) and (ii) of
Lemma 3.4. Even if for each x0 ∈ X and for each associated sequence {𝑥𝑛} of x0
relative to 𝑓, 𝑔 𝑎𝑛𝑑 ℎ.
The sequence 𝑓𝑥0 , 𝑔𝑥1 , 𝑓 𝑥2 , 𝑔𝑥3 ,··· 𝑓𝑥2𝑛 𝑔𝑥2𝑛+1,··· converges in X
Then (𝑋, 𝑆) need not be complete as shown in the following example.

Corollary 4.3.6. (pp1584-1586) Let f,g and h be self maps of a metric

space (𝑋, 𝑑) such that
(i) 𝑓(𝑋) ∪ 𝑔(𝑋) ⊆ ℎ(𝑋)
(ii) 𝑑(𝑓 𝑥, 𝑔𝑦) ≤ 𝑐λ0 (𝑥, 𝑦) for all 𝑥, 𝑦 ∈ 𝑋 where λ0 (𝑥, 𝑦) =
1
𝑚𝑎𝑥{𝑑(ℎ𝑥, ℎ𝑦), 𝑑(𝑓 𝑥, ℎ𝑥), 𝑑(𝑔𝑦, ℎ𝑥), [𝑑(𝑓 𝑥, ℎ𝑦) +
2
𝑑(𝑔𝑦, ℎ𝑥)]} 𝑎𝑛𝑑 0 ≤ 𝑐 < 1
(iii) ℎ is continuous and
(iv) 𝑓 ℎ = ℎ 𝑓 and 𝑔ℎ = ℎ𝑔 Further if
(v) 𝑋 is complete.
(vi) Then 𝑓, 𝑔 and h have a unique common fixed point 𝑥 ∈ 𝑋.
Proof. Given that (𝑋, 𝑑) is a metric space satisfying condition (i) to (v) of the
corollary. If
𝑆𝑑 (𝑥, 𝑦, 𝑧) = 𝑑(𝑥, 𝑦) + 𝑑(𝑦, 𝑧) + 𝑑(𝑧, 𝑥)𝑡ℎ𝑒𝑛 (𝑋, 𝑆𝑑 )is a S-metric space.
Also (ii) can be written as 𝑆𝑑 (f x, f y, f y)) ≤ cλ(x, y) for all 𝑥, 𝑦 ∈ 𝑋 where
𝜆(𝑥, 𝑦) =
𝑚𝑎𝑥{𝑆𝑑 (ℎ𝑥, ℎ𝑦, ℎ𝑦), 𝑆𝑑 (𝑓 𝑥, 𝑓 𝑥, ℎ𝑥), 𝑆𝑑(𝑔𝑦, 𝑔𝑦, 𝑔𝑦), 1 2 [𝑆𝑑 (𝑓 𝑥, 𝑓 𝑥, ℎ𝑦) +
𝑆𝑑 (𝑔𝑦, 𝑔𝑦, ℎ𝑥)]}
which is same as the condition (ii) of Theorem 3.3. Also since (𝑋, 𝑑) is
complete, we have (𝑋, 𝑆) is complete, by Theorem.
Now, 𝑓, 𝑔 𝑎𝑛𝑑 ℎ are self maps on (𝑋, 𝑆) satisfying conditions of Theorem 3.3
and hence the Corollary follows.
 CONCLUSIONS
We generalized some theorems in fixed point theorem work. Theorem 3.1 is a
generalization of Theorem 3.4 of Tayyab Kamran, 2004 [10]. Theorem 3.5 and
Theorem 3.8 are generalizations of Theorem 2.3 and Theorem 2.8 of Yicheng Liu,
Jun Wu, Zhixiang Li, 2005 [19], for single-valued and multi-valued mappings on
S-metric and SH-metric spaces respectively. We showed that not every S-metric is
necessarily continuous. The notion of compatible for single-valued and multi-
valued mappings can be defined to investigate the existence of fixed points in S-
metric spaces. Also, the existence of solution for certain nonlinear integral
equations can be investigated in a future work. Acknowledgement The authors
would like to express their appreciation to the referee for his/ her valuable
comments and suggestions that improved the original version.
 References:
[1]I. Altun, Fixed points and homotropy results for multivalued mappings
satisfying an implicit relation, J. Fixed Point Theory and Appl., 9(2011), 125-134.
[2]I. Altun and D. Turkoglu, some fixed-point theorems for weakly compatible
mappings satisfying an implicit relation, Taiwanese J. Math., 13(4) (2009), 1291-
1304.
[3]I. Beg, A. R. Butt, Fixed point for set-valued mappings satisfying an implicit
relation in partially ordered metric spaces, Nonlinear Anal., 71(9) (2009), 3699-
3704.
[4]M. Imdad, S. Kumar and M. S. Khan, Remarks on some fixed-point theorems
satisfying implicit relations, Rad. Math., 11(1) (2002), 135-143.
[5]V. Popa, A general coincidence theorem for compatible multivalued mappings
satisfying an implicit relation, Demonstration Math., 33(1) (2000), 159-164.
[6]V. Popa, some fixed-point theorems for compatible mappings satisfying an
implicit relation, demonstration Math., 32(1) (1999), 157-163.
[7]S. Sedghi, I. Altun and N. Shobe, A fixed point theorem for multi-maps
satisfying an implicit relation on metric spaces, Appl. Anal. Discrete Math.,
2(2)(2008), 189-196.
[8] B.C. Dhage, Generalized metric spaces and mappings with fixed point, Bull.
Cal. Math. Soc., 84(1992), 329–336.
[9] S. Gahler, 2- metrische Raume and ihre topologische struktur, Math. Nachr,
26(1963), 115–148.
[10] S. Gahler, Zur geometric 2-metriche raume, Revue Roumaine de
Mathmatiques pures et Appliquees, 11(1966), 665–667.
[11] S. Sedghi, N. shobe and H. Zhou, A Common fixed-point theorem in D∗-
metric spaces, Fixed Point Theory Appl., (2007), 1–3.