International Journal of Neutrosophic Science (IJNS) Vol. 23, No. 03, PP.

140-147, 2024

Selection of the best process for desalination under a Treesoft set

environment using the multi-criteria decision-making method
G. Dhanalakshmi1 , S. Sandhiya1*, Florentin Smarandache2
1
Department of Mathematics, Vels Institute of Science, Technology and Advanced Studies, Chennai, Tamilnadu,
India
2
Department of Mathematics, University of New Mexico, Gallup, NM 87301, USA.
Emails: [email protected]; [email protected]; [email protected]

Abstract
Multi-criteria decision-making (MCDM), which has been called a revolution in the field, is one of the most exact
methods for making decisions. Multicriteria decision-making (MCDM) is the process of selecting options by
considering multiple criteria to determine which is best. A multitude of applications in engineering, design, and
finance are possible with the tools and methods derived from MCDM. Application-oriented problems with multiple
criteria involve ambiguous and more inaccurate options, to deal with this ambiguity Smarandache introduced Treesoft
sets, which are an extension of hypersoft sets. So, in this paper, we will consider a real-life application-oriented
problem “Desalination process” under the treesoft sets environment and find the best method for desalination using
one of the MCDM methods.
Keywords: Desalination; Multi-criteria Decision Making; Treesoft Set; Neutrosophic set
1. Introduction
One of the most valuable resources on the earth is water. More than 40% of people worldwide are already impacted
by its scarcity, according to the UN a figure that prompts concern and motivates the pursuit of remedies. Desalination
is one of these, and it is not novel either. It involves using physical and chemical methods to extract the minerals
mostly salt from saltwater. In the upcoming years, increasing desalination plant capacity while lowering environmental
impact will be crucial. Since 70% of our world is covered in water, it is easy to assume that there is an abundance of
it. Nevertheless, fresh water is limited in the world (it makes up only 3% of total water), and two-thirds of that is either
frozen solid or unstable. Approximately 2.7 billion people experience scarcity at least one month of the year, while
1.1 billion people worldwide lack access to clean water. Desalination can be beneficial in these situations because,
ironically, a lot of places where fresh water scarcity occurs are near the sea.The method of desalination is used to get
rid of the dissolved mineral salts in water. This currently one of the most popular methods for extracting fresh water
for use in agriculture or human consumption when applied to seawater. Reverse osmosis, solar distillation,
electrodialysis, nanofiltration, and gas hydrate are some of the commonly used methods in this desalination process.
Decision-making is a multifaceted intellectual process that involves solving problems while taking into account
various factors to achieve a desired outcome. This process may be rational or irrational and can use implicit and
explicit assumptions, which may be impacted by biological, cultural, physiological, and social factors. The complexity
of decision-making is influenced by these factors, as well as by the level of risk and authority involved. In today’s
world, computer equipment can assist in automatically calculating and estimating solutions to decision-making
problems by using mathematical equations, manifold statistics, mathematics, and economic theories Multicriteria
decision-making (MCDM) aims to find the best option by considering several criteria during the selection process.
Various tools and techniques from MCDM can be applied in different sectors, such as finance, engineering, and
robotics. The scientific tools for managing incomplete data are believed to be periodic mathematics (PM), rough set
theory (RST), fuzzy set theory (FST), and probability set theory (PST), which are particularly relevant to these issues,
regardless of the missing data range.

140
DOI: https://doi.org/10.54216/IJNS.230311
Received: August 14, 2023 Revised: November 28, 2023 Accepted: January 29, 2024
 International Journal of Neutrosophic Science (IJNS) Vol. 23, No. 03, PP. 140-147, 2024

Fuzzy sets were first proposed by Zadeh to manage uncertainty, leading to their usefulness in various tasks like pattern
detection, medical diagnosis, and decision-making. As a result of their popularity, numerous generalizations of fuzzy
sets, such as rough sets, soft sets, intuitionistic fuzzy sets, bipolar fuzzy sets, bipolar soft sets, m-polar soft sets,
hypersoft sets, and many others, have been introduced. Soft sets (Ss) are an intriguing expansion of fuzzy sets with
additional vagueness and ambiguity features at the highest level of incompleteness. Soft sets deal with uncertain,
fuzzy, and unspecified elements. In every decision-making scenario, the role of parameters is relevant, but when
parameters are unclear, experts must evaluate them. In such cases, uncertainty is assessed by assigning a fuzzy
membership grade, or a fuzzy soft set, which is an extension of soft sets.
There are various circumstances in which attributes need to be further classified into attribute-valued disjoint sets, and
fuzzy soft sets are seen to be appropriate parametrization technique to deal with vagueness and uncertainty.
Smarandache originated the idea of hypersoft sets (HSSs) from the notion of soft sets by substituting a multiparameter
(sub-attribute) function defined on the cartesian product of n attributes for the one-parameter function. Compared to
SS, the established HSS is more adaptable and better suited for contexts where decisions are made. Additionally, he
introduced sophisticated HSS extensions, such as fuzzy HSS, intutionistic fuzzy HSS, and crisp HSS. HSS theory and
its extensions have been gaining surprising traction in recent years. Numerous researchers examined advanced
distinguishing operators in addition to traits derived from HSS and its expansions.
The hypersoft sets, a soft set extension, used a multi-argument approximation function to solve the shortcomings of
the current structures for attribute-valued disjoint sets. Eventually, MultiSoft sets were added to soft sets to address
ambiguous circumstances in real-world issues. Also, treesoft sets, which are extremely close to hypersoft sets, have
been an extension of MultiSoft sets in recent times. Treesoft sets are concerned with parameters, sub-parameters, sub-
sub-parameters, and so on, while hypersoft sets deal with parameters and sub-parameters. One of the unique
characteristics of treesoft sets is their ability to have numerous sub-sub-parameters, denoted as level 0, level 1, level
2,…,level n. Uncertainties in real-world problems will be greatly diminished with the use of treesoft sets. In this paper
One of the MCDM method, the ELECTRE method is used. ELECTRE ( Elimination and choice Translating Reality)
approach has been used to solve issues in many different fields because how well it ranks the alternatives in terms of
precedence among the criteria. Benayoun et al. first introduced the ELECTRE method, which has then expanded into
many other methods. Numerous domains, including risk assessment, supplier selection, and multiple-criteria decision
making have effectively used ELECTRE approaches.
2. Preliminaries :
2.1 Linguistic variable: A linguistic variable is defined in terms of a base variable, whose values are real numbers
within a specific range. A base variable in the usual sense, as exemplified by any physical variable( e.g., temperature,
pressure, electric current, magnetic flux, etc.) as well as any other numerical variable (e.g., interest rate, blood count,
age, performance, etc.)
Table 1: Linguistic variable
Linguistic Terms Numerical range
Very Small [0.0-0.25]
Small [0.25-0.45]
Medium [0.45-0.65]
Large [0.65-0.85]
Very Large [0.85-1.0]

2.2 Fuzzy set: Let 𝑈 be the universe whose generic element be denoted by 𝑢. A fuzzy set 𝐹 in 𝑈 is a function 𝐹: 𝑈 →
[0,1]. We frequently use 𝜇𝐹 for the function 𝐹 and say that the fuzzy set 𝐹 is characterized by its membership function
𝜇𝐹 : 𝑈 → [0,1] which associates with each 𝑢 in 𝑈, a real number 𝜇𝐹 (𝑢) in [0,1]. The value 𝜇𝐹 (𝑢) at 𝑢 represents the
grade of membership of 𝑢 in 𝐹 and is interpreted as the degree to which 𝑢 belongs to 𝐹. Thus, the closer the value of
𝜇𝐹 (𝑢) is to 1, the more 𝑢 belongs to 𝐹.
2.3 Soft set: Let 𝑈 be a universal set, 𝑃(𝑈) be the power set of 𝑈 and 𝐴 be the set of all parameters. A soft set (𝛿, 𝐴)
on 𝑈 is defined by:

(𝛿, 𝐴) = {(𝑎, 𝛿(𝑎)): 𝑎 ∈ 𝐴, 𝛿(𝑎) ∈ 𝑃(𝑈)} where 𝛿: 𝐴 → 𝑃(𝑈)

141
DOI: https://doi.org/10.54216/IJNS.230311
Received: August 14, 2023 Revised: November 28, 2023 Accepted: January 29, 2024
 International Journal of Neutrosophic Science (IJNS) Vol. 23, No. 03, PP. 140-147, 2024

2.4 Hypersoft sets: Let 𝑈 be the universal set and 𝑃(𝑈) be the power set of 𝑈. Consider 𝑎1 , 𝑎2 , 𝑎3 , … , 𝑎𝑛 , for 𝑛 ≥ 1
be 𝑛 well defined attributes, whose corresponding sub-attributes of 𝑎𝑖 are respectively the sets 𝐴1 , 𝐴2 , 𝐴3 , … , 𝐴𝑛 with
𝐴𝑖 ∩ 𝐴𝑗 = ∅ for 𝑖 ≠ 𝑗 and 𝑖, 𝑗 ∈ {1,2,3, … , 𝑛} then the pair (𝛿, 𝐴1 × 𝐴2 × … × 𝐴𝑛 ) is said to be hypersoft set over
𝑋 where 𝛿: 𝐸1 × 𝐸2 × … × 𝐸𝑛 → 𝑃(𝑈).
2.5 MultiSoft sets: Let 𝑈 be the universal set and 𝐻 be a non-empty subset of 𝑈. And 𝑃(𝐻) is the power set of 𝐻. Let
𝐴1 , 𝐴2 , … , 𝐴𝑛 be 𝑛 ≥ 2 sets of attributes (parameters) whose intersection 𝐴1 ∩ 𝐴2 ∩ … ∩ 𝐴𝑛 = ∅. Let 𝐴 = 𝐴1 ∪ 𝐴2 ∪
… ∪ 𝐴𝑛 and 𝑃(𝐴) be the powerset of 𝐴. Then 𝐹: 𝑃(𝐴) → 𝑃(𝐻) is a Multisoft set over 𝐻.
2.6 Treesoft sets: Let 𝑈 be the universal set and 𝐻 a non-empty subset of 𝑈, with 𝑃(𝐻) the powerset of 𝐻. Let 𝐴 be
the set of attributes, 𝐴 = {𝐴1 , 𝐴2 , … , 𝐴𝑛 }, for integer 𝑛 ≥ 1, where 𝐴1 , 𝐴2 , … , 𝐴𝑛 are attributes of first level (since they
have one-digit indexes). Each attribute 𝐴𝑖 , 1 ≤ 𝑖 ≤ 𝑛, is formed by sub-attributes:

𝐴1 = {𝐴1,1 , 𝐴1,2 , … }

𝐴2 = {𝐴2,1 , 𝐴2,2 , … }

.
.
.

𝐴𝑛 = {𝐴𝑛,1 , 𝐴𝑛,2 , … }

Where 𝐴𝑖,𝑗 are sub-attributes ( or attributes of second level) (since they have two-digit indexes). Again, each sub-
attribute 𝐴𝑖,𝑗 is formed by sub-sub-attributes (or attributes of third level): 𝐴𝑖,𝑗,𝑘 and so on, as much refinement as
needed into each application, up to sub-sub-…-sub-attributes or attributes of 𝑚 − 𝑙𝑒𝑣𝑒𝑙 (or having 𝑚 digits into
indexes): 𝐴𝑖1,𝑖2 ,…,𝑖𝑚 Therefore, a graph-tree is formed, that we denote as Tree(A) , whose root is 𝐴 ( considered as
level zero), then nodes of 𝑙𝑒𝑣𝑒𝑙 1, 𝑙𝑒𝑣𝑒𝑙 2, … , 𝑙𝑒𝑣𝑒𝑙 𝑚, we call it as leaves of the graph-tree, all terminal nodes (nodes
that have no descendants). Then the treesoft set is:

𝐹: 𝑃(𝑇𝑟𝑒𝑒(𝐴)) → 𝑃(𝐻)

Tree(A) is the set of all nodes and leaves (from level 1to level m) of the graph-tree, and 𝑃(𝑇𝑟𝑒𝑒(𝐴)) is the powerset
of the Tree(A). All node sets of the TreeSoft set of 𝑙𝑒𝑣𝑒𝑙 𝑚 are:

𝑇𝑟𝑒𝑒(𝐴) = {𝐴𝑖1 |𝑖1 = 1,2, … } ∪ {𝐴𝑖1,𝑖 |𝑖1 , 𝑖2 = 1,2, … } ∪ … ∪ {𝐴𝑖1,𝑖2 ,…,𝑖𝑚 |𝑖1 , 𝑖2 , … , 𝑖𝑚 = 1,2 … }
2

The first set is formed by the nodes of 𝑙𝑒𝑣𝑒𝑙1, second set by the nodes of 𝑙𝑒𝑣𝑒𝑙 2, third set by the nodes of 𝑙𝑒𝑣𝑒𝑙 3,
and so on, the last set is formed by the nodes of 𝑙𝑒𝑣𝑒𝑙 𝑚. If the graph-tree has only two levels (𝑚 = 2), then the
TreeSoft set is reduced to a MultiSoft set.
Example 1: Consider the set 𝑆 = {𝑠1 , 𝑠2 , 𝑠3 , 𝑠4 , 𝑠5 , 𝑠6 , 𝑠7 , 𝑠8 } be the cancer diagnosed patients, and 𝑃(𝑆) the powerset
of 𝑆 with the corresponding attributes 𝐻 = {𝐻1 , 𝐻2 , 𝐻3 }
Now lets’s assume that the function F gets the following values:
𝐹(𝐹𝑎𝑡𝑖𝑔𝑢𝑒, 𝑇𝑢𝑚𝑜𝑟 𝑙𝑜𝑐𝑎𝑡𝑖𝑜𝑛, 𝐵𝑙𝑜𝑜𝑑 𝑐𝑒𝑙𝑙 𝑐𝑜𝑢𝑛𝑡, 𝐴𝑏𝑛𝑜𝑟𝑚𝑎𝑙) = {𝑠4 , 𝑠5 , 𝑠8 , }
𝐹(𝐹𝑎𝑡𝑖𝑔𝑢𝑒, 𝑇𝑢𝑚𝑜𝑟 𝑙𝑜𝑐𝑎𝑡𝑖𝑜𝑛, 𝐵𝑙𝑜𝑜𝑑 𝑐𝑒𝑙𝑙 𝑐𝑜𝑢𝑛𝑡, 𝑛𝑜𝑟𝑚𝑎𝑙) = {𝑠1 , 𝑠2 }
𝐹(𝐹𝑎𝑡𝑖𝑔𝑢𝑒, 𝑇𝑢𝑚𝑜𝑟 𝑙𝑜𝑐𝑎𝑡𝑖𝑜𝑛, 𝐵𝑙𝑜𝑜𝑑 𝑐𝑒𝑙𝑙 𝑐𝑜𝑢𝑛𝑡)
= 𝑎𝑙𝑙 𝑝𝑎𝑡𝑖𝑒𝑛𝑡𝑠 ℎ𝑎𝑣𝑖𝑛𝑔 𝑡ℎ𝑒𝑠𝑒 𝑎𝑙𝑡𝑒𝑟𝑛𝑎𝑡𝑖𝑣𝑒𝑠 𝑐𝑜𝑚𝑒 𝑖𝑢𝑛𝑑𝑒𝑟 𝑡ℎ𝑖𝑠 𝑐𝑎𝑡𝑒𝑔𝑜𝑟𝑦
𝐹(𝐹𝑎𝑡𝑖𝑔𝑢𝑒, 𝑇𝑢𝑚𝑜𝑟 𝑙𝑜𝑐𝑎𝑡𝑖𝑜𝑛, 𝐵𝑙𝑜𝑜𝑑 𝑐𝑒𝑙𝑙 𝑐𝑜𝑢𝑛𝑡, 𝑎𝑏𝑛𝑜𝑟𝑚𝑎𝑙)
∪ 𝐹(𝐹𝑎𝑡𝑖𝑔𝑢𝑒, 𝑇𝑢𝑚𝑜𝑟 𝑙𝑜𝑐𝑎𝑡𝑖𝑜𝑛, 𝐵𝑙𝑜𝑜𝑑 𝑐𝑒𝑙𝑙 𝑐𝑜𝑢𝑛𝑡, 𝑛𝑜𝑟𝑚𝑎𝑙) = {𝑠4 , 𝑠5 , 𝑠8 , 𝑠1 , 𝑠2 }

142
DOI: https://doi.org/10.54216/IJNS.230311
Received: August 14, 2023 Revised: November 28, 2023 Accepted: January 29, 2024
 International Journal of Neutrosophic Science (IJNS) Vol. 23, No. 03, PP. 140-147, 2024

CANCER
DIAGNOSIS

𝑯𝟏
𝑯𝟐 =IMAGING 𝑯𝟑 =LAB TESTS
=SYMPTOM
RESULTS RESULTS
ANALYSIS

𝑯𝟐𝟏 =TUMOR 𝑯𝟐𝟐 =TUMOR 𝑯𝟑𝟏 =TUMOR 𝑯𝟑𝟐 =BLOOD

𝑯𝟏𝟏 =PAIN 𝑯𝟏𝟐 =FATIGUE
SIZE LOCATION MARKER LEVEL CELL COUNT

𝑯𝟐𝟐𝟏 = 𝑯𝟑𝟏𝟏 𝑯𝟑𝟐𝟏

𝑯𝟏𝟏𝟏 =LOW 𝑯𝟏𝟐𝟏 =LOW 𝑯𝟐𝟏𝟏 =SMALL
PERIPHERAL =NORMAL =NORMAL

𝑯𝟏𝟏𝟐 𝑯𝟏𝟐𝟐 𝑯𝟐𝟏𝟐 𝑯𝟐𝟐𝟐 𝑯𝟑𝟏𝟐 = 𝑯𝟑𝟐𝟐 =

=MEDIUM =MEDIUM =MEDIUM =CENTRAL ELEVATED ABNORMAL

𝑯𝟏𝟏𝟑 =HIGH 𝑯𝟏𝟐𝟑 =HIGH 𝑯𝟐𝟏𝟑 =BIG

3. MCDM Method
ELECTRE (Elimination and Choice Translating reality) MCDM method: The ELECTRE approach is notable
outranking models that are broadly applicable for comparing performances across a range of factors for various
MCDM issues. The Engineering and Management departments have already started using ELECTRE as a decision-
making tool. Benayoun et al.’s ELECTRE techniques, which was first developed, and then rooted out to ELECTRE
I, II, III, IV, and IS methods are a popular family of outranking techniques for MCDM issues, which have seen lot of
positive and beneficial developments.

Step 1: Develop the Decision matrix 𝑋 = [𝑥𝑖𝑗 ]𝑛×𝑚

Step 2: Normalized matrix and weighted normalized matrix can be determined using the below formulae:
𝑥𝑖𝑗
𝑟𝑖𝑗 = 𝑚
(1)
√∑𝑖=1 𝑥𝑖𝑗 2
𝑚

𝑉𝑖𝑗 = 𝑤𝑖𝑗 ∑ 𝑟𝑖𝑗 (2)

𝑖=1

Step 3: Find the concordance and discordance interval sets

The concordance interval set is used to describe the dominance query

𝑐𝑎𝑏 = {𝑗|𝑥𝑎𝑗 ≥ 𝑥𝑏𝑗 }, where 𝑗 = 1,2,3 … 𝑛 (3)

The discordance interval sets is given by

𝑑𝑎𝑏 = {𝑗|𝑥𝑎𝑗 < 𝑥𝑏𝑗 }, where 𝑗 = 1,2,3 … 𝑛 (4)

Step 4: Calculation of concordance interval matrix

𝑐𝑎𝑏 = ∑ 𝑤𝑗 𝑓𝑜𝑟 𝑗 = 1,2,3 … 𝑛 (5)

𝑗∈𝑐𝑎𝑏

Step 5: Determine the discordance interval matrix

143
DOI: https://doi.org/10.54216/IJNS.230311
Received: August 14, 2023 Revised: November 28, 2023 Accepted: January 29, 2024
 International Journal of Neutrosophic Science (IJNS) Vol. 23, No. 03, PP. 140-147, 2024

𝑗 ∈ 𝐷𝑎𝑏 𝑚𝑎𝑥 |𝑣𝑎𝑗 − 𝑣𝑏𝑗 |

𝑑(𝑎, 𝑏) = (6)
𝑗 ∈ 𝐽; 𝑚, 𝑛 ∈ 𝐼 𝑚𝑎𝑥 |𝑣𝑚𝑗 − 𝑣𝑛𝑗 |

Step 6: Calculation of concordance index matrix

𝑚 𝑚
𝑐(𝑎, 𝑏)
𝑐̅ = ∑ ∑ (7)
𝑚(𝑚 − 1)
𝑎=1 𝑏=1

Based on the threshold value 𝑐̅ concordance dominance matrix E can be determined as follows
𝑒(𝑎, 𝑏) = 1 𝑖𝑓 𝑐(𝑎, 𝑏) ≥ 𝑐̅
𝐸={ (8)
𝑒(𝑎, 𝑏) = 0 𝑖𝑓 𝑐(𝑎, 𝑏) < 𝑐̅
Step 7: Determine the discordance index matrix
The preference of dissatisfaction can be measured by discordance index:
𝑚 𝑚
𝑑(𝑎, 𝑏)
𝑑̅ = ∑ ∑ (9)
𝑚(𝑚 − 1)
𝑎=1 𝑏=1

Based on the discordance index mentioned above, the discordance matrix F is given by :

𝑓(𝑎, 𝑏) = 1 𝑖𝑓 𝑑(𝑎, 𝑏) ≤ 𝑑̅
𝐹={ (10)
𝑓(𝑎, 𝑏) = 0 𝑖𝑓 𝑑(𝑎, 𝑏) > 𝑑̅

Step 8: Calculate the net superior and net inferior value

Let 𝑐𝑎 and 𝑑𝑎 be the net superior and net inferior value
The 𝑐𝑎 is given by:
𝑛 𝑛

𝑐𝑎 = ∑ 𝑐(𝑎, 𝑏) − ∑ 𝑐(𝑏, 𝑎) (11)

𝑏=1 𝑏=1

On the contrary 𝑑𝑎 is used to determine the number of inferiority ranking of the alternatives:
𝑛 𝑛

𝑑𝑎 = ∑ 𝑑(𝑎, 𝑏) − ∑ 𝑑(𝑏, 𝑎) (12)

𝑏=1 𝑏=1

Step 9: Calculate the ranking of the alternatives and find the best alternative.
4. Formulation of the problem: Let 𝐴 and 𝐵 respectively be the Environmental head and Technical Manager; they
are the decision makers, and they will decide the best desalination process from the below methods for converting
sea water into drinking water, with the following weight criterions given by {0.2,0.13,0.16,0.11,0.21,0.19}
𝐷 = {𝑑1 , 𝑑2 , 𝑑3 , 𝑑4 , 𝑑5 } be the desalination process, were,
𝑑1 = 𝑅𝑒𝑣𝑒𝑟𝑧𝑒 𝑂𝑠𝑚𝑜𝑠𝑖𝑠, 𝑑2 = 𝑆𝑜𝑙𝑎𝑟 𝐷𝑖𝑠𝑡𝑖𝑙𝑙𝑎𝑡𝑖𝑜𝑛, 𝑑3 = 𝐸𝑙𝑒𝑐𝑡𝑟𝑜 𝑑𝑖𝑎𝑙𝑦𝑠𝑖𝑠,
𝑑4 = 𝑁𝑎𝑛𝑜 𝑓𝑖𝑙𝑡𝑟𝑎𝑡𝑖𝑜𝑛, 𝑑5 = 𝐺𝑎𝑠 ℎ𝑦𝑑𝑟𝑎𝑡𝑒 𝑓𝑜𝑟𝑚𝑎𝑡𝑖𝑜𝑛
and the corresponding attributes be
𝑒1 = 𝐶𝑜𝑠𝑡𝑠, 𝑒2 = 𝐶ℎ𝑒𝑚𝑖𝑐𝑎𝑙 𝑟𝑒𝑎𝑐𝑡𝑖𝑜𝑛𝑠, 𝑒3 = 𝑃𝑟𝑜𝑐𝑒𝑠𝑠 𝑜𝑓 𝑑𝑒𝑠𝑎𝑙𝑖𝑛𝑎𝑡𝑖𝑜𝑛,
𝑒4 = 𝑅𝑒𝑠𝑜𝑢𝑟𝑐𝑒𝑠, 𝑒5 = 𝐻𝑒𝑎𝑙𝑡ℎ 𝑎𝑠𝑝𝑒𝑐𝑡𝑠, 𝑒6 = 𝑃𝑢𝑏𝑙𝑖𝑐 𝑆𝑢𝑝𝑝𝑜𝑟𝑡
with the corresponding sub-attributes
𝑒11 = 𝑒𝑐𝑜𝑛𝑜𝑚𝑖𝑐 𝑐𝑜𝑠𝑡𝑠, 𝑒21 = 𝑎𝑑𝑣𝑒𝑟𝑠𝑒 𝑒𝑓𝑓𝑒𝑐𝑡𝑠,
𝑒1 = 𝐸1 = { } ; 𝑒2 = 𝐸2 = { }
𝑒12 = 𝑒𝑛𝑣𝑖𝑟𝑜𝑛𝑚𝑒𝑛𝑡𝑎𝑙 𝑐𝑜𝑠𝑡𝑠 𝑒22 = 𝑛𝑜𝑛 − 𝑎𝑑𝑣𝑒𝑟𝑠𝑒 𝑒𝑓𝑓𝑒𝑐𝑡𝑠

144
DOI: https://doi.org/10.54216/IJNS.230311
Received: August 14, 2023 Revised: November 28, 2023 Accepted: January 29, 2024
 International Journal of Neutrosophic Science (IJNS) Vol. 23, No. 03, PP. 140-147, 2024

𝑒 = 𝑚𝑒𝑚𝑏𝑟𝑎𝑛𝑒 𝑝𝑟𝑜𝑐𝑒𝑠𝑠, 𝑒41 = 𝑚𝑎𝑛𝑝𝑜𝑤𝑒𝑟,

𝑒3 = 𝐸3 = { 31 } ; 𝑒4 = 𝐸4 = { 𝑒 = 𝑒𝑞𝑢𝑖𝑝𝑚𝑒𝑛𝑡 }
𝑒32 = 𝑡ℎ𝑒𝑟𝑚𝑎𝑙 𝑝𝑟𝑜𝑐𝑒𝑠𝑠 42

𝑒 = ℎ𝑖𝑔ℎ 𝑓𝑟𝑒𝑠ℎ 𝑤𝑎𝑡𝑒𝑟 𝑠𝑐𝑎𝑟𝑐𝑖𝑡𝑦

𝑒5 = 𝐸5 = {𝑒51 = 𝑖𝑜𝑑𝑖𝑛𝑒 𝑑𝑒𝑓𝑖𝑐𝑖𝑒𝑛𝑦}; 𝑒6 = 𝐸6 = { 61 }
𝑒62 = 𝑙𝑜𝑤 𝑓𝑟𝑒𝑠ℎ 𝑤𝑎𝑡𝑒𝑟 𝑠𝑐𝑎𝑟𝑐𝑖𝑡𝑦
And the corresponding sub-sub-attributes are given below:
𝑒111 = 𝑒𝑛𝑒𝑟𝑔𝑦 𝑐𝑜𝑠𝑡, 𝑒 = 𝑑𝑒𝑠𝑡𝑟𝑢𝑐𝑡𝑖𝑜𝑛 𝑜𝑓 𝑓𝑜𝑜𝑑 𝑐ℎ𝑎𝑖𝑛,
𝑒11 = { } ; 𝑒12 = { 121 };
𝑒112 = 𝑙𝑎𝑛𝑑 𝑐𝑜𝑠𝑡 𝑒122 = 𝑟𝑒𝑠𝑖𝑑𝑢𝑎𝑙 𝑠𝑎𝑙𝑡
𝑒21 = {𝑒211 = 𝑚𝑜𝑟𝑒 𝑜𝑓 𝑐ℎ𝑒𝑚𝑖𝑐𝑎𝑙𝑠}; 𝑒22 = {𝑒221 = 𝑝𝑢𝑟𝑖𝑡𝑦 𝑜𝑓 𝑤𝑎𝑡𝑒𝑟};
𝑒31 = {𝑒311 = 𝑡ℎ𝑖𝑛 𝑠ℎ𝑒𝑒𝑡 𝑜𝑓 𝑓𝑖𝑙𝑡𝑒𝑟}; 𝑒32 = {𝑒321 = ℎ𝑒𝑎𝑡 𝑒𝑛𝑒𝑟𝑔𝑦};
𝑒 = 𝑟𝑒𝑠𝑒𝑎𝑟𝑐ℎ𝑒𝑟𝑠, 𝑒 = 𝑒𝑙𝑒𝑐𝑡𝑟𝑖𝑐𝑎𝑙,
𝑒41 = { 411 } ; 𝑒42 = { 421 };
𝑒412 = 𝑡𝑒𝑐ℎ𝑛𝑖𝑐𝑖𝑎𝑛𝑠 𝑒422 = 𝑒𝑙𝑒𝑐𝑡𝑟𝑜𝑛𝑖𝑐𝑎𝑙
𝑒51 = {𝑒511 = 𝑟𝑒𝑚𝑜𝑣𝑎𝑙 𝑜𝑓 𝑖𝑜𝑑𝑖𝑛𝑒 𝑑𝑒𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦};
𝑒61 = {𝑒611 = ℎ𝑖𝑔ℎ 𝑝𝑢𝑏𝑙𝑖𝑐 𝑠𝑢𝑝𝑝𝑜𝑟𝑡}; 𝑒62 = {𝑒621 = 𝑙𝑒𝑠𝑠 𝑓𝑟𝑒𝑠ℎ 𝑤𝑎𝑡𝑒𝑟 𝑠𝑐𝑎𝑟𝑐𝑖𝑡𝑦}
Let the function be 𝜑: 𝐸1 ∪ 𝐸2 ∪ 𝐸3 ∪ 𝐸4 ∪ 𝐸5 ∪ 𝐸6 → 𝑃(𝑆)
Now let us assume the relation
𝜑(𝐸1 ∪ 𝐸2 ∪ 𝐸3 ∪ 𝐸4 ∪ 𝐸5 ∪ 𝐸6 )
𝑒𝑛𝑣𝑖𝑟𝑜𝑛𝑚𝑒𝑛𝑡𝑎𝑙(𝑒12 ), 𝑝𝑢𝑟𝑖𝑡𝑦 𝑜𝑓 𝑤𝑎𝑡𝑒𝑟(𝑒221 ), 𝑚𝑒𝑚𝑏𝑟𝑎𝑛𝑒 𝑝𝑟𝑜𝑐𝑒𝑠𝑠 (𝑒31 ), 𝑒𝑙𝑒𝑐𝑡𝑟𝑖𝑐𝑎𝑙(𝑒421 ),
= 𝜑( )
𝑖𝑜𝑑𝑖𝑛𝑒 𝑑𝑒𝑓𝑖𝑐𝑖𝑒𝑛𝑦 (𝑒51 ), ℎ𝑖𝑔ℎ 𝑝𝑢𝑏𝑙𝑖𝑐 𝑠𝑢𝑝𝑝𝑜𝑟𝑡 (𝑒611 ).
These are some of the actual requirements for selecting the best method from desalination process, the decision makers
give their opinions in matrix form separately as:
𝜑(𝑑1 ) = {𝑒12 (0.7), 𝑒221 (0.8), 𝑒31 (0.7), 𝑒421 (0.6), 𝑒51 (0.5), 𝑒611 (0.6)}
𝜑(𝑑2 ) = {𝑒12 (0.7), 𝑒221 (0.3), 𝑒31 (0.4), 𝑒421 (0.5), 𝑒51 (0.1), 𝑒611 (0.2)}
𝐴 = 𝜑(𝑑3 ) = {𝑒12 (0.1), 𝑒221 (0.4), 𝑒31 (0.6), 𝑒421 (0.2), 𝑒51 (0.3), 𝑒611 (0.6)}
𝜑(𝑑4 ) = {𝑒12 (0.5), 𝑒221 (0.3), 𝑒31 (0.4), 𝑒421 (0.2), 𝑒51 (0.7), 𝑒611 (0.4)}
{𝜑(𝑑5 ) = {𝑒12 (0.8), 𝑒221 (0.1), 𝑒31 (0.3), 𝑒421 (0.2), 𝑒51 (0.6), 𝑒611 (0.2)}}
𝜔(𝑑1 ) = {𝑒12 (0.9), 𝑒221 (0.7), 𝑒31 (0.5), 𝑒421 (0.6), 𝑒51 (0.3), 𝑒611 (0.5)}
𝜔(𝑑2 ) = {𝑒12 (0.2), 𝑒221 (0.5), 𝑒31 (0.3), 𝑒421 (0.2), 𝑒51 (0.7), 𝑒611 (0.4)}
𝐵 = 𝜔(𝑑3 ) = {𝑒12 (0.6), 𝑒221 , (0.3)𝑒31 (0.4), 𝑒421 (0.1), 𝑒51 (0.3), 𝑒611 (0.4)}
𝜔(𝑑4 ) = {𝑒12 (0.5), 𝑒221 (0.4), 𝑒31 (0.2), 𝑒421 (0.6), 𝑒51 (0.5), 𝑒611 (0.8)}
{𝜔(𝑑5 ) = {𝑒12 (0.1), 𝑒221 (0.8), 𝑒31 (0.4), 𝑒421 (0.3), 𝑒51 (0.4), 𝑒611 (0.5)}}
Step 1: Construct the decision matrix
0.7 0.8 0.7 0.6 0.5 0.6
0.7 0.3 0.4 0.5 0.1 0.2
𝐴 = 0.1 0.4 0.6 0.2 0.3 0.6
0.5 0.3 0.4 0.2 0.7 0.4
[0.8 0.1 0.3 0.2 0.6 0.2]
0.9 0.7 0.5 0.6 0.3 0.5
0.2 0.5 0.3 0.2 0.7 0.4
𝐵 = 0.6 0.3 0.4 0.1 0.3 0.4
0.5 0.4 0.2 0.6 0.5 0.8
[0.1 0.8 0.4 0.3 0.4 0.5]
Step 2: By using an accuracy function (average of the matrices) both A and B has been reduced to a single matrix as
follows:

145
DOI: https://doi.org/10.54216/IJNS.230311
Received: August 14, 2023 Revised: November 28, 2023 Accepted: January 29, 2024
 International Journal of Neutrosophic Science (IJNS) Vol. 23, No. 03, PP. 140-147, 2024

0.8 0.75 0.6 0.6 0.4 0.55

0.45 0.2 0.35 0.35 0.4 0.3
𝑄 = 0.35 0.35 0.5 0.15 0.3 0.4
0.5 0.35 0.3 0.35 0.6 0.6
[0.45 0.45 0.35 0.25 0.5 0.35]
Step 3: The normalized and weighted normalized matrix are found using equation (1) and (2) mentioned above:
0.134 0.095 0.099 0.077 0.083 0.103
0.076 0.025 0.058 0.052 0.083 0.056
𝑄 = 0.059 0.044 0.082 0.019 0.062 0.075
0.084 0.044 0.049 0.045 0.125 0.112
[0.076 0.057 0.058 0.032 0.104 0.065]
Step 4: Based on the above conditions in equation (3) and (4) and applying the equations (5) and (6) the concordance
and discordance interval matrices are calculated and obtained as follows:
Concordance interval Matrix
0 1 1 0.6 0.79
0.21 0 0.52 0.27 0.47
𝑄1 = 0 0.48 0 0.29 0.35
0.4 0.73 0.84 0 0.71
[0.21 0.89 0.65 0.29 0 ]
Discordance interval matrix
0 0 0 0.819 0.354
1 0 0.766 1 1
𝑄2 = 1 1 0 1 1
1 0.147 0.528 0 0.272
[1 0.61 0.594 1 0 ]
Step 5: Again, by applying equation (7), (8), (9) and (10) we obtain the concordance index matrix and discordance
index matrix
Concordance index matrix
0 1 1 1 1
0 0 1 0 0
𝑄1 = 0 0 0 0 0
0 1 1 0 1
[0 1 1 0 0]
Discordance index matrix
1 1 1 0 1
0 1 0 0 0
𝑄2 = 0 0 1 0 0
0 1 1 1 1
[0 1 1 0 1]
Step 6: Finally, the net superior and net inferior values are found using the equations (11) and (12), then the ranking
is found:
DESALINATION PROCESS NET SUPERIOR RANK NET INFERIOR
RANK
VALUE VALUE
𝑑1 (𝑅𝑒𝑣𝑒𝑟𝑠𝑒 𝑂𝑠𝑚𝑜𝑠𝑖𝑠) 2.57 1 −2.827 1
𝑑2 (𝑆𝑜𝑙𝑎𝑟 𝐷𝑖𝑠𝑡𝑖𝑙𝑙𝑎𝑡𝑖𝑜𝑛) −1.63 4 2.009 4
𝑑3 (𝐸𝑙𝑒𝑐𝑡𝑟𝑜 𝑑𝑖𝑎𝑙𝑦𝑠𝑖𝑠) −1.89 5 2.112 5
𝑑4 (𝑁𝑎𝑛𝑜 𝑓𝑖𝑙𝑡𝑟𝑎𝑡𝑖𝑜𝑛) 1.23 2 −1.872 2
𝑑5 (𝐺𝑎𝑠 𝐻𝑦𝑑𝑟𝑎𝑡𝑒 𝑓𝑜𝑟𝑚𝑎𝑡𝑖𝑜𝑛 −0.28 3 0.578 3

146
DOI: https://doi.org/10.54216/IJNS.230311
Received: August 14, 2023 Revised: November 28, 2023 Accepted: January 29, 2024
 International Journal of Neutrosophic Science (IJNS) Vol. 23, No. 03, PP. 140-147, 2024

From the net superior value of Concordance matrix and net inferior value of Discordance matrix and by ranking of
the alternatives we can conclude that Reverse Osmosis method is ranked in the first position when compared to all
other desalination methods.
4. Conclusion
In an effort to discuss the Treesoft set environment, a real-world problem involving selecting the optimal desalination
procedure using one of the MCDM approaches is taken up in this study. When choosing the process, the related
attributes are also given priority. The ELECTRE approach has already been used to enhance performance in problems
with multiple criteria outranking. Additionally, we may solve more ambiguous and multi-attribute problems by
utilising Treesoft sets. Treesoft sets can be used in all other fuzzy environments Intutionistic, Pythagorean,
Neutrosophic, and so forth in future research.
References
[1] Abbas, M, Murtaza.G & Smarandache,F. “Basic operations on hypersoft sets and hypersoft point”.
Neutrosophic sets and systems, vol.35, no. 1, pp. 407-421 (2020).
[2] Atanassov, K. Intuitionistic fuzzy sets. “Fuzzy sets and systems”, vol. 20,0 no.1, pp. 87-96 (1986).
http://dx.doi.org/10.1016/S0165-0114(86)80034-3
[3] Cagman, N., Enginoglu, S., & Citak,F. “Fuzzy soft set theory and its applications”, Iranian journal of fuzzy
systems, vol.8, no.3, pp.137-147(2011). 10.22111/IJFS.2011.292
[4] Khalid Abd, Kazem Abhary, Romeo Marian. “Application of Fuzzy Logic to Multi-Objective Scheduling
Problems in Robotic Flexible Assembly Cells”. Automation Control and Intelligent Systems. vol. 1, No. 3,
2013, pp. 34-41. 10.11648/j.acis.20130103.11
[5] Kong, Z., Wang., L., Wu, Z. “Application of fuzzy soft set in decision making problems based on grey theory”
. J. Comput. Appl. Math.,vol. 236, pp. 1521-1530(2011). https://doi.org/10.1016/j.cam.2011.09.016
[6] Maji, P.K., Biswas, R., & Roy, A.R. “Fuzzy soft sets”. Journal of fuzzy mathematics. vol. 9, pp. 589-602
(2001).
[7] Maji, P.K, Biswas R and Roy A.R. “Soft set theory”, Computer. Math. Appl., vol. 45, pp. 555-562(2003).
https://doi.org/10.1016/S0898-1221(03)00016-6
[8] Smarandache, F. ‘ Extension of soft sets to hypersoft set, and then to plithogenic hypersoft set, “ Neutrosophic
Sets and system, vol. 22, pp. 168-170,(2018). 10.5281/zenodo.1408740
[9] Smarandache, F. “ Practical applications of IndetermSoft sets and IndetermHypersoft sets and Introduction to
Treesoft set as an extension of Multisoft set, “ Neutrosophic Sets and System, vol.51, pp. 941-947, (2022).
10.5281/zenodo.7154238
[10] Smarandache, F. “New types of Soft sets, HyperSoft sets, IndetermSoft sets, IndetermHyperSoft set, and
TreeSoft set”, Neutrosophic Sets and System, vol.8 (2023) https://doi.org/10.61356/j.nswa.2023.41.
[11] Yildiray Celik and Sultan Yamak, “Fuzzy soft set theory applied to medical diagnosis using fuzzy arithmetic
operations” vol.1, no. 82 (2013). https://doi.org/10.1186/1029-242X-2013-82
[12] Zadeh L.A. “Fuzzy sets, Information and Control” vol. 8, pp. 338-353 (1965). https://doi.org/10.1016/S0019-
9958(65)90241-X

147
DOI: https://doi.org/10.54216/IJNS.230311
Received: August 14, 2023 Revised: November 28, 2023 Accepted: January 29, 2024