Hindawi Publishing Corporation

International Journal of Combinatorics

Volume 2016, Article ID 2508156, 4 pages
http://dx.doi.org/10.1155/2016/2508156

Research Article
On Self-Centeredness of Product of Graphs

Priyanka Singh and Pratima Panigrahi

Department of Mathematics, Indian Institute of Technology Kharagpur, Kharagpur 721302, India

Correspondence should be addressed to Priyanka Singh; priyankaiit22@gmail.com

Received 4 April 2016; Revised 28 June 2016; Accepted 12 July 2016

Academic Editor: Laszlo A. Szekely

Copyright © 2016 P. Singh and P. Panigrahi. This is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly
cited.

A graph 𝐺 is said to be a self-centered graph if the eccentricity of every vertex of the graph is the same. In other words, a graph is a self-
centered graph if radius and diameter of the graph are equal. In this paper, self-centeredness of strong product, co-normal product,
and lexicographic product of graphs is studied in detail. The necessary and sufficient conditions for these products of graphs to be a
self-centered graph are also discussed. The distance between any two vertices in the co-normal product of a finite number of graphs
is also computed analytically.

1. Introduction (i) Cartesian product, denoted by 𝐺 = 𝐺1 ◻𝐺2 ◻ ⋅ ⋅ ⋅ ◻𝐺𝑛 ,

where 𝑥 ∼ 𝑦 if and only if 𝑥𝑖 𝑦𝑖 ∈ 𝐸(𝐺𝑖 ) for exactly one
The concept of self-centered graphs is widely used in applica- index 𝑖, 1 ≤ 𝑖 ≤ 𝑛, and 𝑥𝑗 = 𝑦𝑗 for each index 𝑗 ≠ 𝑖,
tions, for example, the facility location problem. The facility (ii) strong product, denoted by 𝐺 = 𝐺1 ⊠ ⋅ ⋅ ⋅ ⊠ 𝐺𝑛 , where
location problem is to locate facilities in a locality (network) 𝑥 ∼ 𝑦 if and only if 𝑥𝑖 𝑦𝑖 ∈ 𝐸(𝐺𝑖 ) or 𝑥𝑖 = 𝑦𝑖 , for every
so that these facilities can be used efficiently. All graphs in this 𝑖, 1 ≤ 𝑖 ≤ 𝑛,
paper are simple and connected graphs. The distance between
two vertices 𝑢 and V in a graph 𝐺, denoted by 𝑑𝐺(𝑢, V) (or (iii) lexicographic product, denoted by 𝐺 = 𝐺1 ∘ ⋅ ⋅ ⋅ ∘ 𝐺𝑛 ,
simply 𝑑(𝑢, V)), is the minimum length of 𝑢-V path in the where 𝑥 ∼ 𝑦 if and only if, for some 𝑗 ∈ {1, 2,
graph. The eccentricity of a vertex V in 𝐺, denoted by ecc𝐺(V), . . . , 𝑛}, 𝑥𝑗 𝑦𝑗 ∈ 𝐸(𝐺𝑗 ) and 𝑥𝑖 = 𝑦𝑖 for each 1 ≤ 𝑖 < 𝑗,
is defined as the distance between V and a vertex farthest from (iv) co-normal product, denoted by 𝐺 = 𝐺1 ∗ 𝐺2 ∗ ⋅ ⋅ ⋅ ∗
V; that is, ecc𝐺(V) = max{𝑑𝐺(V, 𝑢) : 𝑢 ∈ 𝑉(𝐺)}. The radius 𝐺𝑛 , where 𝑥 ∼ 𝑦 if and only if 𝑥𝑖 ∼ 𝑦𝑖 for some 𝑖 ∈
rad(𝐺) and diameter diam(𝐺) of the graph 𝐺 are, respectively, {1, 2, . . . , 𝑛}.
the minimum and maximum eccentricity of the vertices of Self-centered graphs have been broadly studied and surveyed
graph 𝐺; that is, rad(𝐺) = min{ecc(V) : V ∈ 𝑉(𝐺)} and in [1–3]. In [4], the authors described several algorithms to
diam(𝐺) = max{ecc(V) : V ∈ 𝑉(𝐺)}. The center 𝐶(𝐺) of construct self-centered graphs. Stanic [5] proved that the Car-
graph 𝐺 is the induced subgraph of 𝐺 on the set of all vertices tesian product of two self-centered graphs is a self-centered
with minimum eccentricity. A graph 𝐺 is said to be a self- graph. Inductively, one can prove that Cartesian product of
centered graph if the eccentricity of every vertex is the same; 𝑛-self-centered graphs is also a self-centered graph.
that is, 𝐶(𝐺) = 𝐺 or rad(𝐺) = diam(𝐺). If the eccentricity In this paper, we find conditions for self-centeredness of
of every vertex is equal to 𝑑, then 𝐺 is called 𝑑-self-centered strong product, co-normal product, and lexicographic prod-
graph. uct of graphs.
For any kind of graph product 𝐺 of the graphs 𝐺1 , 𝐺2 ,
. . . , 𝐺𝑛 , the vertex set is taken as 𝑉(𝐺) = {(𝑥1 , 𝑥2 , . . . , 𝑥𝑛 ) : 2. Main Results
𝑥𝑖 ∈ 𝑉(𝐺𝑖 )}. Because of their adjacency rules, product names
are different. Let 𝑥 = (𝑥1 , . . . , 𝑥𝑛 ) and 𝑦 = (𝑦1 , . . . , 𝑦𝑛 ) be two In this section, we will discuss the self-centeredness of differ-
vertices in 𝑉(𝐺). Then the product is called ent types of product graphs. As mentioned before, all graphs
2 International Journal of Combinatorics

considered here are simple and connected. The following = max {max {𝑑𝐺𝑖 (𝑥𝑖 , 𝑦𝑖 )} : 𝑦 ∈ 𝑉 (𝐺)} ,
result is given by Stanic [5]. 1≤𝑖≤𝑛
(2)
Theorem 1. If 𝐺1 and 𝐺2 are 𝑚- and 𝑛-self-centered graphs,
respectively, then 𝐺1 ◻𝐺2 is (𝑚 + 𝑛)-self-centered graph. Recip- where 𝑥 = (𝑥1 , . . . , 𝑥𝑛 ) and 𝑦 = (𝑦1 , . . . , 𝑦𝑛 ).
rocally, if 𝐺1 ◻𝐺2 is self-centered, then both graphs 𝐺1 and 𝐺2 First, let 𝐺𝑘 be 𝑑-self-centered graph for some 𝑘 ∈
are self-centered. {1, 2, . . . , 𝑛} and diam(𝐺𝑖 ) ≤ 𝑑 for all 𝑖, 1 ≤ 𝑖 ≤ 𝑛. Since 𝐺𝑘
is 𝑑-self-centered, ecc(𝑥𝑘 ) = 𝑑 and there exists some 𝑦𝑘 in 𝐺𝑘
By method of induction, one can extend the above theo- such that 𝑑(𝑥𝑘 , 𝑦𝑘 ) = 𝑑. As diam(𝐺𝑖 ) ≤ 𝑑 for all 𝑖, 1 ≤ 𝑖 ≤ 𝑛,
rem and get the result given below. the distance between any two vertices in any 𝐺𝑖 cannot exceed
𝑑. Hence, ecc(𝑥) = 𝑑 for all 𝑥 ∈ 𝑉(𝐺) and thus 𝐺 is 𝑑-self-
Theorem 2. Let 𝐺 = 𝐺1 ◻𝐺2 ◻ ⋅ ⋅ ⋅ ◻𝐺𝑛 be the Cartesian prod- centered graph.
uct of graphs 𝐺1 , 𝐺2 , . . . , 𝐺𝑛 . If every 𝐺𝑖 is 𝑑𝑖 -self-centered Conversely, let 𝐺 be a 𝑑-self-centered graph. If, for some
graph, then 𝐺 is 𝑚-self-centered graph, where 𝑚 = ∑𝑛𝑖=1 𝑑𝑖 , 𝑙 ∈ {1, . . . , 𝑛}, diam(𝐺𝑙 ) = 𝑑𝑙 > 𝑑, then there exist vertices
1 ≤ 𝑖 ≤ 𝑛. Conversely, if 𝐺 is a self-centered graph, then every 𝑥𝑙 and 𝑦𝑙 in 𝐺𝑙 such that 𝑑(𝑥𝑙 , 𝑦𝑙 ) = 𝑑𝑙 . Now for 𝑥 = (𝑥1 ,
𝐺𝑖 is a self-centered graph. . . . , 𝑥𝑙 , . . . , 𝑥𝑛 ) and 𝑦 = (𝑦1 , . . . , 𝑦𝑙 , . . . , 𝑦𝑛 ) in 𝑉(𝐺), 𝑑(𝑥, 𝑦) ≥
𝑑(𝑥𝑙 , 𝑦𝑙 ) = 𝑑𝑙 > 𝑑 and so ecc(𝑥) ≥ 𝑑𝑙 > 𝑑. This contradicts
Next we will discuss self-centeredness of strong product the fact that 𝐺 is 𝑑-self-centered graph and thus it is proven
of graphs. that diam(𝐺𝑖 ) ≤ 𝑑 for all 𝑖. Now, our claim is that there
exists 𝑘 ∈ {1, . . . , 𝑛} such that 𝐺𝑘 is 𝑑-self-centered graph.
Theorem 3. Let 𝐺 = 𝐺1 ⊠ ⋅ ⋅ ⋅ ⊠ 𝐺𝑛 be the strong product of
On the contrary, suppose that none of 𝐺𝑖 is 𝑑-self-centered
graphs 𝐺1 , 𝐺2 , . . . , 𝐺𝑛 . Then 𝐺 is 𝑑-self-centered graph if and
graph. Then there exist vertices 𝑥𝑖 ∈ 𝑉(𝐺𝑖 ) for all 𝑖 such
only if, for some 𝑘 ∈ {1, . . . , 𝑛}, 𝐺𝑘 is 𝑑-self-centered graph and
that ecc(𝑥𝑖 ) = 𝑑𝑖 < 𝑑. Let 𝑥 = (𝑥1 , . . . , 𝑥𝑛 ). Then ecc(𝑥) =
diam(𝐺𝑖 ) ≤ 𝑑 for every 𝑖, 1 ≤ 𝑖 ≤ 𝑛.
max1≤𝑖≤𝑛 {𝑑𝑖 } < 𝑑, which contradicts the fact that 𝐺 is 𝑑-self-
Proof. For any two vertices 𝑥 = (𝑥1 , . . . , 𝑥𝑛 ) and 𝑦 = (𝑦1 , centered graph.
. . . , 𝑦𝑛 ), the distance between them is given in [6]: In the following lemma, we determine the formula for the
distance between two vertices in the co-normal product of a
𝑑 (𝑥, 𝑦) = max {𝑑𝐺𝑖 (𝑥𝑖 , 𝑦𝑖 )} . (1) finite number of graphs.
1≤𝑖≤𝑛

Now, the eccentricity of any vertex 𝑥 of 𝐺 is given by Lemma 4. Let 𝐺 = 𝐺1 ∗ 𝐺2 ∗ ⋅ ⋅ ⋅ ∗ 𝐺𝑛 be the co-normal
product of graphs 𝐺1 , 𝐺2 , . . . , 𝐺𝑛 . The distance between 𝑥 =
ecc (𝑥) = max {𝑑 (𝑥, 𝑦) : 𝑦 ∈ 𝑉 (𝐺)} (𝑥1 , . . . , 𝑥𝑛 ) and 𝑦 = (𝑦1 , . . . , 𝑦𝑛 ) in 𝐺 is

{ 1 𝑖𝑓 𝑥𝑖 ∼ 𝑦𝑖 𝑓𝑜𝑟 𝑠𝑜𝑚𝑒 𝑖 ∈ {1, 2, . . . , 𝑛}

{
{
{
{
{
{
{
{𝑑 (𝑥𝑖 , 𝑦𝑖 )
{ 𝑖𝑓 𝐺𝑗 = 𝐾1 , ∀𝑗 ≠ 𝑖
𝑑 (𝑥, 𝑦) = { (3)
{
{
{
{ 2 𝑖𝑓 𝑥 ≁ 𝑦, 𝑥𝑙 ≠ 𝑦𝑙 𝑓𝑜𝑟 𝑒𝑥𝑎𝑐𝑡𝑙𝑦 𝑜𝑛𝑒 𝑖𝑛𝑑𝑒𝑥 𝑙 𝑎𝑛𝑑 𝐺𝑗 ≠ 𝐾1 𝑓𝑜𝑟 𝑠𝑜𝑚𝑒 𝑗 ≠ 𝑙
{
{
{
{
{
{2 𝑖𝑓 𝑥 ≁ 𝑦 𝑎𝑛𝑑 ∃ 𝑎𝑡 𝑙𝑒𝑎𝑠𝑡 𝑡𝑤𝑜 𝑖𝑛𝑑𝑖𝑐𝑒𝑠 𝑘, 𝑙 𝑠.𝑡. 𝑥𝑘 ≠ 𝑦𝑘 𝑎𝑛𝑑 𝑥𝑙 ≠ 𝑦𝑙 .

Proof. Consider two vertices 𝑥 = (𝑥1 , . . . , 𝑥𝑛 ) and 𝑦 = (𝑦1 , Finally, consider the case, where, for at least two indices 𝑘
. . . , 𝑦𝑛 ) of 𝐺. If, for some 𝑖 ∈ {1, 2, . . . , 𝑛}, 𝑥𝑖 ∼ 𝑦𝑖 , then by the and 𝑙, 𝑥𝑘 ≠ 𝑦𝑘 and 𝑥𝑙 ≠ 𝑦𝑙 ; that is, for at least two indices
definition of co-normal product 𝑥 ∼ 𝑦 and thus 𝑑(𝑥, 𝑦) = 1. 𝑘 and 𝑙, 𝐺𝑘 ≠ 𝐾1 and 𝐺𝑙 ≠ 𝐾1 . Since 𝑥 ≁ 𝑦, 𝑥𝑘 ≁ 𝑦𝑘 ,
Next, let 𝐺𝑗 = 𝐾1 for all 𝑗 ≠ 𝑖. In this case, for any and 𝑥𝑙 ≁ 𝑦𝑙 , then from the connectivity of graphs 𝐺𝑘 and
path 𝑃 between 𝑥 and 𝑦, every adjacent pair of vertices in 𝐺𝑙 there exist vertices 𝑧𝑘 ∈ 𝑉(𝐺𝑘 ) and 𝑧𝑙 ∈ 𝑉(𝐺𝑙 ) such that
𝑃 differ only in the 𝑖th coordinate. So 𝑑(𝑥, 𝑦) = 𝑑(𝑥𝑖 , 𝑦𝑖 ). 𝑧𝑘 ∼ 𝑥𝑘 in 𝐺𝑘 and 𝑧𝑙 ∼ 𝑦𝑙 in 𝐺𝑙 . Then we have a vertex
For the third option of the distance formula, we have vertices 𝑧 = (𝑥1 , . . . , 𝑧𝑘 , . . . , 𝑧𝑙 , . . . , 𝑥𝑛 ) ∈ 𝑉(𝐺) such that 𝑥 ∼ 𝑧 and
𝑥 and 𝑦 as 𝑥 = (𝑥1 , 𝑥2 , . . . , 𝑥𝑙 , . . . , 𝑥𝑗 , . . . , 𝑥𝑛 ) and 𝑦 = (𝑥1 , 𝑧 ∼ 𝑦. Thus 𝑥𝑧𝑦 will be an 𝑥-𝑦 path of length two and this
𝑥2 , . . . , 𝑦𝑙 , . . . , 𝑥𝑗 , . . . , 𝑥𝑛 ) such that 𝑥𝑙 ≠ 𝑦𝑙 and 𝐺𝑗 ≠ 𝐾1 proves that 𝑑(𝑥, 𝑦) = 2.
for some 𝑗 ≠ 𝑙. Since 𝐺𝑗 is connected graph, there exists a
vertex 𝑧𝑗 ∈ 𝑉(𝐺𝑗 ) such that 𝑥𝑗 ∼ 𝑧𝑗 and thus we get a vertex
𝑧 = (𝑥1 , 𝑥2 , . . . , 𝑥𝑙 , . . . , 𝑧𝑗 , . . . , 𝑥𝑛 ) ∈ 𝐺 such that 𝑥 ∼ 𝑧 and The following theorem gives necessary and sufficient
𝑧 ∼ 𝑦 (because 𝑥𝑗 = 𝑦𝑗 ) and 𝑥𝑧𝑦 is a path of length two and conditions for a co-normal product of graphs to be a self-
hence 𝑑(𝑥, 𝑦) = 2. centered graph.
International Journal of Combinatorics 3

Theorem 5. Let 𝐺 = 𝐺1 ∗ 𝐺2 ∗ ⋅ ⋅ ⋅ ∗ 𝐺𝑛 be the co-normal where 𝑖 is the smallest index for which 𝑥𝑖 ≠ 𝑦𝑖 . For 𝑖 ≥ 2, let
product of graphs 𝐺1 , 𝐺2 , . . . , 𝐺𝑛 with |𝑉(𝐺𝑖 )| = 𝑛𝑖 . Then the 𝑟 = min{𝑑𝐺𝑖 (𝑥𝑖 , 𝑦𝑖 ), 2}. Then 𝑟 ≤ 2. Since 𝑑 ≥ 2, we get 𝑟 ≤ 𝑑.
following hold: Now, for 𝑥 ∈ 𝑉(𝐺),

(i) Let 𝐺𝑖 ≠ 𝐾1 and 𝐺𝑗 = 𝐾1 for all 𝑗 ≠ 𝑖. Then 𝐺 is ecc (𝑥) = max {𝑑 (𝑥, 𝑦) : 𝑦 ∈ 𝑉 (𝐺)}
𝑑-self-centered graph if and only if 𝐺𝑖 is 𝑑-self-centered
graph. = max {𝑑𝐺1 (𝑥1 , 𝑦1 ) , 𝑟 : 𝑦1 ∈ 𝑉 (𝐺1 )} (7)
(ii) Let there be at least two values of 𝑖 such that 𝐺𝑖 ≠ 𝐾1 . = 𝑑,
Then 𝐺 is 2-self-centered graph if and only if there exists
an index 𝑙 such that Δ(𝐺𝑙 ) ≠ 𝑛𝑙 − 1, where Δ(𝐺) is the because ecc(𝑥1 ) = 𝑑 and there exists 𝑦1 ∈ 𝐺1 such that
maximum degree of a vertex in 𝐺. 𝑑(𝑥1 , 𝑦1 ) = 𝑑. This proves that ecc(𝑥) = 𝑑 for all 𝑥 ∈ 𝑉(𝐺)
and hence 𝐺 is a 𝑑-self-centered graph.
Proof. (i) The result is true because 𝐺 is isomorphic to 𝐺𝑖 in Next, let 𝑘 > 1. Since |𝑉(𝐺1 )| = 1, there is no 𝑦1 ∈ 𝐺1 such
this case through the isomorphism that 𝑥1 ≠ 𝑦1 . So, first option in the distance formula will not
arise. Since the degree of the vertex in 𝐺𝑗 for 𝑗 = 1, 2, . . . , 𝑘−1
𝑓 : 𝑉 (𝐺) 󳨀→ 𝑉 (𝐺𝑖 ) (4) is zero, if 𝑖 = 𝑘 in the above distance formula then 𝑑(𝑥, 𝑦) =
𝑑𝐺𝑘 (𝑥𝑘 , 𝑦𝑘 ). Since 𝐺𝑘 ≠ 𝐾1 and is connected deg(𝑥𝑘 ) ≠ 0. So
with 𝑓(𝑥1 , . . . , 𝑥𝑖 , . . . , 𝑥𝑛 ) = 𝑥𝑖 . if 𝑖 ≥ 𝑘 + 1 in the above formula, 𝑑(𝑥, 𝑦) = min{𝑑𝐺𝑖 (𝑥𝑖 , 𝑦𝑖 ), 2}
(ii) Let 𝐺 be a 2-self-centered graph. If, for all the indices and thus the above formula to calculate the distance reduces
𝑖, Δ(𝐺𝑖 ) = 𝑛𝑖 − 1, then there are vertices 𝑥𝑖 ∈ 𝑉(𝐺𝑖 ), 1 ≤ 𝑖 ≤ 𝑛, to
such that deg(𝑥𝑖 ) = 𝑛𝑖 −1. Now, the vertex 𝑥 = (𝑥1 , 𝑥2 , . . . , 𝑥𝑛 ),
ecc(𝑥) = 1, which contradicts the fact that 𝐺 is 2-self-centered {𝑑𝐺𝑘 (𝑥𝑘 , 𝑦𝑘 ) if 𝑖 = 𝑘
graph. Hence there exists an index 𝑙 such that Δ(𝐺𝑙 ) ≠ 𝑛𝑙 − 1. 𝑑 (𝑥, 𝑦) = { (8)
Conversely, let there be an index 𝑙 such that Δ(𝐺𝑙 ) ≠ 𝑛𝑙 −1. min {𝑑𝐺𝑖 (𝑥𝑖 , 𝑦𝑖 ) , 2} if 𝑖 ≥ 𝑘 + 1,
{
Then for any vertex 𝑥 = (𝑥1 , 𝑥2 , . . . , 𝑥𝑙 , 𝑥𝑙+1 , . . . , 𝑥𝑛 ) in 𝐺 there
exists another vertex 𝑦 = (𝑥1 , 𝑥2 , . . . , 𝑦𝑙 , 𝑥𝑙+1 , . . . , 𝑥𝑛 ), where where 𝑖 is the smallest index for which 𝑥𝑖 ≠ 𝑦𝑖 . For 𝑖 ≥ 𝑘 + 1
𝑦𝑙 ∈ 𝑉(𝐺𝑙 ) and 𝑥𝑙 ≁ 𝑦𝑙 . Since 𝑥 ≁ 𝑦, from the third option of let 𝑟1 = min{𝑑𝐺𝑖 (𝑥𝑖 , 𝑦𝑖 ), 2}. Then 𝑟1 ≤ 2. Since 𝑑 ≥ 2, we get
the distance formula given in Lemma 4, ecc(𝑥) = 2. Since 𝑥 𝑟1 ≤ 𝑑. Thus, for any vertex 𝑥 ∈ 𝑉(𝐺), we have
is an arbitrary vertex, 𝐺 is 2-self-centered graph.
ecc (𝑥) = max {𝑑 (𝑥, 𝑦) : 𝑦 ∈ 𝑉 (𝐺)}
In the following two theorems, we discuss self-centered-
= max {𝑑𝐺𝑘 (𝑥𝑘 , 𝑦𝑘 ) , 𝑟1 : 𝑦𝑘 ∈ 𝑉 (𝐺𝑘 )} (9)
ness of lexicographic product of graphs.
= 𝑑.
Theorem 6. Let 𝐺 = 𝐺1 ∘ 𝐺2 ∘ ⋅ ⋅ ⋅ ∘ 𝐺𝑛 be the lexicographic
product of graphs 𝐺1 , 𝐺2 , . . . , 𝐺𝑛 and let 𝑘 ≥ 1 be the smallest This proves that ecc(𝑥) = 𝑑 for all 𝑥 ∈ 𝑉(𝐺) and hence 𝐺 is a
index for which 𝐺𝑘 ≠ 𝐾1 . If 𝐺𝑘 is 𝑑-self-centered graph, where 𝑑-self-centered graph.
𝑑 ≥ 2, then 𝐺 is 𝑑-self-centered graph. The converse is true for Conversely, let 𝐺 be a 𝑑-self-centered graph, where 𝑑 ≥ 3.
𝑑 ≥ 3. Then ecc(𝑥) = 𝑑 for all 𝑥 ∈ 𝑉(𝐺). Notice that, for any vertex
𝑥 = (𝑥1 , 𝑥2 , . . . , 𝑥𝑛 ) in 𝐺,
Proof. For vertices 𝑥 = (𝑥1 , . . . , 𝑥𝑛 ) and 𝑦 = (𝑦1 , . . . , 𝑦𝑛 ) of 𝐺,
the following distance formula is due to Hammack et al. [6]:
ecc (𝑥) = max {𝑑 (𝑥, 𝑦) : 𝑦 ∈ 𝑉 (𝐺)}
𝑑 (𝑥, 𝑦)
{max {ecc (𝑥1 ) , 𝑟 : 𝑥1 ∈ 𝑉 (𝐺1 )} if 𝑘 = 1 (10)
𝑑𝐺1 (𝑥1 , 𝑦1 ) if 𝑥1 ≠ 𝑦1
={
{
{ max {ecc (𝑥𝑘 ) , 𝑟1 : 𝑥𝑘 ∈ 𝑉 (𝐺𝑘 )} if 𝑘 > 1,
{
{ (5) {
= {𝑑𝐺𝑖 (𝑥𝑖 , 𝑦𝑖 ) if 𝑑𝐺𝑙 (𝑥𝑙 ) = 0 ∀1 ≤ 𝑙 < 𝑖
{
{
{ where 𝑟 and 𝑟1 are as defined above. Since ecc(𝑥) (which is
{min {𝑑𝐺𝑖 (𝑥𝑖 , 𝑦𝑖 ) , 2} if 𝑑𝐺𝑙 (𝑥𝑙 ) ≠ 0 for some 1 ≤ 𝑙 < 𝑖, the maximum of ecc(𝑥𝑘 ) and 𝑟 or 𝑟1 ) is equal to 𝑑, 𝑑 ≥ 3 and
𝑟, 𝑟1 ≤ 2, we get ecc(𝑥𝑘 ) = 𝑑 for all 𝑥𝑘 ∈ 𝑉(𝐺𝑘 ). So 𝐺𝑘 is
where 𝑖 is the smallest index for which 𝑥𝑖 ≠ 𝑦𝑖 .
𝑑-self-centered graph.
Let |𝑉(𝐺𝑖 )| = 1 for 𝑖 = 1, 2, . . . , 𝑘 − 1 and let 𝐺𝑘 be 𝑑-self-
centered graph, where 𝑑 ≥ 2. First let 𝑘 = 1. Since |𝑉(𝐺1 )| > If we take 𝑑 = 2, then ecc(𝑥) = 2 may not imply that
1, 𝐺1 is connected and degree of no vertex in 𝐺1 is zero; then ecc(𝑥𝑘 ) = 2 (there may be ecc(𝑥𝑘 ) < 2 and 𝑟 or 𝑟1 is equal to
the second option in the distance formula will not arise. Then 2; see example below).
the above formula to calculate the distance reduces to
Example 7. Here we consider the lexicographic product of
{𝑑𝐺1 (𝑥1 , 𝑦1 ) if 𝑖 = 1 three graphs, 𝐺1 , 𝐺2 , and 𝐺3 , where 𝐺1 = 𝐾2 , 𝐺2 = 𝑃4 ,
𝑑 (𝑥, 𝑦) = { (6) and 𝐺3 = 𝐾2 . Let 𝑉(𝐺1 ) = {𝑥, 𝑦}, 𝑉(𝐺2 ) = {𝑎, 𝑏, 𝑐, 𝑑}, and
min {𝑑𝐺𝑖 (𝑥𝑖 , 𝑦𝑖 ) , 2} if 𝑖 ≥ 2,
{ 𝑉(𝐺3 ) = {1, 2}. The lexicographic product 𝐺 = 𝐾2 ∘ 𝑃4 ∘ 𝐾2
4 International Journal of Combinatorics

(x, a, 1) (x, a, 2) (x, b, 1) (x, b, 2) (x, c, 1) (x, c, 2) (x, d, 1) (x, d, 2)

(y, a, 1) (y, a, 2) (y, b, 1) (y, b, 2) (y, c, 1) (y, c, 2) (y, d, 1) (y, d, 2)

Figure 1

of graphs 𝐾2 , 𝑃4 , and 𝐾2 is shown in Figure 1. One can check Competing Interests

that the eccentricity of every vertex of 𝐺 is two and hence 𝐺 is
a 2-self-centered graph. However, 𝐺1 is not a 2-self-centered The authors declare that there are no competing interests
graph. regarding the publication of this paper.

In the theorem below, we present the general version of References

the 2-self-centered product graphs included in the previous
example. [1] F. Buckley, “Self-centered graphs,” Annals of the New York
Academy of Sciences, vol. 576, pp. 71–78, 1989.
Theorem 8. Let 𝐺 = 𝐺1 ∘ 𝐺2 ∘ ⋅ ⋅ ⋅ ∘ 𝐺𝑛 be the lexicographic [2] F. Buckley, Z. Miller, and P. J. Slater, “On graphs containing a
product of graphs 𝐺1 , 𝐺2 , . . . , 𝐺𝑛 with |𝑉(𝐺𝑖 )| = 𝑛𝑖 , let 𝐺𝑘 be given graph as center,” Journal of Graph Theory, vol. 5, no. 4, pp.
1-self-centered graph for some 𝑘 ∈ {1, . . . , 𝑛 − 1}, and let 𝐺𝑖 (if 427–434, 1981.
it exists) be 𝐾1 for all 𝑖 < 𝑘. Then 𝐺 is a 2-self-centered graph if [3] T. N. Janakiraman, “On special classes of self-centred graphs,”
and only if Δ(𝐺𝑗 ) ≠ 𝑛𝑗 − 1 for some 𝑗 ≥ 𝑘 + 1. Discrete Mathematics, vol. 126, no. 1–3, pp. 411–414, 1994.
[4] T. N. Janakiraman, M. Bhanumathi, and S. Muthammai, “Self-
Proof. First let 𝐺 be a 2-self-centered graph. It is given that, centered super graph of a graph and center number of a graph,”
for some 𝑘 ∈ {1, . . . , 𝑛 − 1}, 𝐺𝑘 is 1-self-centered graph and let Ars Combinatoria, vol. 87, pp. 271–290, 2008.
𝐺𝑖 be 𝐾1 for all 𝑖 < 𝑘. Our claim is that Δ(𝐺𝑗 ) ≠ 𝑛𝑗 −1 for some [5] Z. Stanic, “Some notes on minimal self-centered graphs,” AKCE
𝑗 ≥ 𝑘 + 1. On the contrary, let Δ(𝐺𝑗 ) = 𝑛𝑗 − 1 for all 𝑗 ≥ 𝑘 + 1. International Journal of Graphs and Combinatorics, vol. 7, pp.
Then there are vertices 𝑔𝑖 ∈ 𝐺𝑖 such that ecc(𝑔𝑖 ) = 1 for every 97–102, 2010.
𝑖, 𝑘 ≤ 𝑖 ≤ 𝑛. Now, by using above distance formula, for every [6] R. Hammack, W. Imrich, and S. Klavžar, Handbook of Product
𝑥 = (𝑥1 , . . . , 𝑥𝑘−1 , 𝑔𝑘 , . . . , 𝑔𝑛 ) in 𝐺, one gets ecc(𝑥) = 1. This Graphs, CRC Press, 2nd edition, 2011.
contradicts the fact that 𝐺 is a 2-self-centered graph.
Conversely, let Δ(𝐺𝑙 ) ≠ 𝑛𝑙 − 1 for some 𝑙 ≥ 𝑘 + 1. Then
for any vertex 𝑥𝑙 ∈ 𝐺𝑙 there exists 𝑦𝑙 ∈ 𝐺𝑙 such that 𝑥𝑙 ≁ 𝑦𝑙 .
For any vertex 𝑥 = (𝑥1 , . . . , 𝑥𝑘 , . . . , 𝑥𝑙 , . . . , 𝑥𝑛 ) there exists a
vertex 𝑦 = (𝑥1 , . . . , 𝑥𝑘 , . . . , 𝑦𝑙 , . . . , 𝑥𝑛 ) such that 𝑥 ≁ 𝑦. So,
ecc(𝑥) ≥ 2. Since 𝐺𝑖 = 𝐾1 for all 𝑖 < 𝑘 (if any), the distance
formula will be

{𝑑𝐺𝑖 (𝑥𝑖 , 𝑦𝑖 ) if 𝑖 = 𝑘
𝑑 (𝑥, 𝑦) = { (11)
min {𝑑𝐺𝑖 (𝑥𝑖 , 𝑦𝑖 ) , 2} if 𝑖 ≥ 𝑘 + 1,
{

where 𝑖 is the smallest index for which 𝑥𝑖 ≠ 𝑦𝑖 . Since 𝐺𝑘 is a 1-

self-centered graph, 𝑑𝐺𝑖 (𝑥𝑖 , 𝑦𝑖 ) = 1 if 𝑖 = 𝑘. Also, for 𝑖 ≥ 𝑘 + 1,
min{𝑑𝐺𝑖 (𝑥𝑖 , 𝑦𝑖 ), 2} ≤ 2. Thus eccentricity of no vertex is more
than two and we get ecc(𝑥) = 2 for every 𝑥 ∈ 𝐺. Hence 𝐺 is a
2-self-centered graph.
Advances in Advances in Journal of Journal of
Operations Research
Hindawi Publishing Corporation
Decision Sciences
Hindawi Publishing Corporation
Applied Mathematics
Hindawi Publishing Corporation
Algebra
Hindawi Publishing Corporation
Probability and Statistics
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

The Scientific International Journal of

World Journal
Hindawi Publishing Corporation
Differential Equations
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

Submit your manuscripts at

http://www.hindawi.com

International Journal of Advances in

Combinatorics
Hindawi Publishing Corporation
Mathematical Physics
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

Journal of Journal of Mathematical Problems Abstract and Discrete Dynamics in

Complex Analysis
Hindawi Publishing Corporation
Mathematics
Hindawi Publishing Corporation
in Engineering
Hindawi Publishing Corporation
Applied Analysis
Hindawi Publishing Corporation
Nature and Society
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

International
Journal of Journal of
Mathematics and
Mathematical
Discrete Mathematics
Sciences

Journal of International Journal of Journal of

Hindawi Publishing Corporation Hindawi Publishing Corporation Volume 2014

Function Spaces
Hindawi Publishing Corporation
Stochastic Analysis
Hindawi Publishing Corporation
Optimization
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014