Journal of

Marine Science
and Engineering

Article
The Development of a Transdisciplinary Framework to
Overcome Energy Efficiency Barriers in Shipbuilding:
A Case Study for an Iranian Shipyard
Seyed Vahid Vakili * , Aykut I. Ölçer and Alessandro Schönborn

Maritime Energy Management (MEM), World Maritime University (WMU), 211 18 Malmö, Sweden;
aio@wmu.se (A.I.Ö.); as@wmu.se (A.S.)
* Correspondence: svv@wmu.se

Abstract: Through a systematic literature review and a holistic perspective, the study proposes a
conceptual transdisciplinary framework to overcome energy efficiency barriers during the ship-
building phase. The process of the proposed transdisciplinary framework consists of five steps
of “goal formation”, “system analyzing”, “scenario construction”, “multi-criteria decision making
assessment” and “strategy building” to identify and rank the energy efficiency barriers during ship
construction based on decision makers’ priorities. The framework categorizes the barriers into five
disciplines—operations, policies and regulations, technology and innovation, the human elements,
and economics—and the framework is applied to an Iranian shipyard. The results show that the





 economic barriers have the greatest impact, while the human barriers have the least impact on the
shipyard’s energy performance. Due to the generalized structure of the framework, it categorizes
Citation: Vakili, S.V.; Ölçer, A.I.;
not only energy efficiency barriers according to the importance and priorities of stakeholders in the
Schönborn, A. The Development of a
shipbuilding industry, but can also be applied to other phases of the ship life cycle and even to other
Transdisciplinary Framework to
Overcome Energy Efficiency Barriers
industries. By applying the framework, decision-makers can make rational and optimal decisions to
in Shipbuilding: A Case Study for an be able to invest in energy efficiency measures based on their priorities.
Iranian Shipyard. J. Mar. Sci. Eng.
2021, 9, 1113. https://doi.org/ Keywords: air emissions; energy efficiency barriers; energy policy; life cycle management; sustainable
10.3390/jmse9101113 energy management system; transdisciplinary

Academic Editors: Menglan Duan,

Hassan Karampour and
Hema Wadhwa 1. Introduction
Industry contributes to 24% of global greenhouse gas emissions (GHG) and is respon-
Received: 22 September 2021
sible for approximately one third of the world’s energy consumption [1]. To meet the UN
Accepted: 11 October 2021
Sustainable Development Goals (SDGs), industrial emissions must be reduced by 1.2%
Published: 13 October 2021
per year, from 8.5 gigatons (Gt) in 2018 to 7.4 Gt of CO2 in 2030 [1]. To achieve this goal,
improving energy efficiency is crucial. Energy efficiency plays a key role in energy security
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
and climate change, and is at the heart of solutions to reduce CO2 emissions, tackle climate
published maps and institutional affil-
change [2,3], and increase energy productivity and business competitiveness [4,5].
iations.
The transport sector as a whole contributes to 36% of energy consumption in 2017 [6]
and the maritime sector, which transports 80% of the world’s freight volume [7], contributes
to 2.89% of global emissions [8]. It is predicted that the related emissions from shipping
will increase by 50% compared to 2018 levels by 2050 [8]. The IMO, which regulates interna-
tional shipping, has proposed multidisciplinary measures to improve energy efficiency and
Copyright: © 2021 by the authors.
thereby reduce GHG emissions. The Energy Efficiency Design Index (EEDI) and the Energy
Licensee MDPI, Basel, Switzerland.
This article is an open access article
Efficiency Existing Ship Index (EEXI) are technical measures, while the Enhanced Ship
distributed under the terms and Energy Efficiency Management Plan (ESEEMP)1 is an operational measure. In addition,
conditions of the Creative Commons with the introduction of the European Emissions Trading Scheme (ETS) in the maritime
Attribution (CC BY) license (https:// cluster, the IMO has been pressed to adopt market-based measures.
creativecommons.org/licenses/by/ Studies and research about improving energy efficiency and related barriers in ships
4.0/). have been conducted in other sectors. However, the associated regulations, actions and

J. Mar. Sci. Eng. 2021, 9, 1113. https://doi.org/10.3390/jmse9101113 https://www.mdpi.com/journal/jmse

J. Mar. Sci. Eng. 2021, 9, 1113 2 of 25

studies are focused on the operation cycle of the ships, and less attention is paid to im-
proving energy efficiency in the other cycles of ships’ life-cycles, which are construction,
maintenance and dismantling. Shipbuilding, as one of the vital global industries, is a
significant energy-consuming and polluting industry that produces a substantial CO2
emissions (4% of ships’ life cycle emission), as well as air pollutants such as CO (29% of
ships’ life-cycle emission) [9]. This means that the total amount of the ships’ life cycle
CO2 emissions that occur in the building, maintenance and dismantling stages is higher
than the ports’ contribution, being responsible for approximately 2% [10] of operational
CO2 emissions. Due to motivation in using zero-carbon fuels, electricity and sail and
solar power, the fraction of shipyard operations in a ship’s life cycle might become larger
than their operational phase, e.g., if a car ferry is being propelled by batteries and using
electricity from the Norwegian electricity grid, its construction will have a more significant
life-cycle climate impact than its operation cycle [11].
Currently, the problem is that energy efficiency measures are not always implemented,
as there are a number of barriers that prevent implementation. Most commonly, economic
researchers consider market failures (imperfections), such as incomplete information,
client-contractor relationships, adverse selection and split incentives [12–14], to be barriers
to improving energy efficiency. However, non-economic researchers strive to identify
other types of barriers by considering different perspectives. Based on these perspectives,
different solutions have been proposed. To improve energy efficiency in the shipping
cluster, more attention is paid to technology [15] and operational measures [16]. Therefore,
safety and reliability, technical uncertainty, behavior, market constraints, financial and
economic constraints and complexity [17] are identified as types of barriers in the ship
operation cycle. However, to the authors’ knowledge, there are no studies that identify
the barriers to energy efficiency in shipping in the context of the ship construction and
maintenance phases of the life cycle.
In addition to the lack of studies that consider energy efficiency during the operational
and manufacturing cycles of the vessel, barriers are treated as solitary and, if they are part
of a group, their relationship and interaction is ignored. In order to support sustainable
shipping and sustainable energy efficiency improvement in the shipping cluster, a holistic,
systematic and transdisciplinary approach from a life cycle perspective must be consid-
ered. This approach identifies the relationship and interaction of barriers with each other,
different stakeholders and policy measures [18,19].
This contributes to the development of a holistic, systematic and interdisciplinary
conceptual framework to address barriers to energy efficiency in shipping clusters and
within manufacturing cycles. To design and develop such a framework, it is important to
review related papers on energy efficiency barriers in different industries and maritime
operations. In the absence of attention to the relationship and interplay between barriers to
energy efficiency and the life cycle perspective within the shipping cluster, this study has
paid particular attention to how barriers interact across disciplines in the manufacturing
life cycle.
In light of the above, this study aims to provide a framework for identifying barriers
to energy efficiency in the shipbuilding industry and overcoming them from a life-cycle
perspective within the maritime cluster. The framework is holistic, systematic and trans-
disciplinary and takes into account the interrelationship and interaction between different
types of barriers. Developing such a framework and implementing it during the construc-
tion phase has the potential to improve energy efficiency and reduce the carbon footprint
of the shipbuilding industry, and also has the same effect on reducing emissions in the life
cycle perspective of offshore and subsea projects. Offshore and subsea projects play a key
role in countries’ blue economic growth. However, deepwater projects are a challenging
activity and such projects require appropriate logistical support from onshore facilities [20].
Specialized vessel types, such as diving vessels, DP vessels, jack-up barges, cable and pipe-
laying vessels and PSVs, improve the efficiency and effectiveness of offshore operations,
J. Mar. Sci. Eng. 2021, 9, 1113 3 of 25

and these specialized vessels, as well as offshore platforms and infrastructure, are built at
shipyards, which have an important role in supporting the projects [21].
By implementing the framework within the organization, decision makers (DMs) will
be able to have a transparent view of the energy barriers in their organization. In this
way, managers and DMs can adopt appropriate energy strategies and policies to overcome
these barriers. The flexibility of the framework means that it is generic and can be applied
to all other sectors and organizations. To validate the proposed framework, a case study
was conducted at an Iranian shipyard which is active in ship repair and conversion and
offshore infrastructure construction.
In light of the above, Section 2 discusses, through a systematic literature review, vari-
ous barriers to energy efficiency with different categorizations in both different industries
and the ship construction cycle. Section 3 explains the methods and methodology, i.e., a
systematic literature review and systematic and transdisciplinary approaches. Section 4
presents the results of the case studies and the related discussion. Finally, Section 5 draws
the conclusion.

2. Literature Review
Since the oil crisis, energy efficiency and the related gap between potential energy
efficiency and actual energy efficiency has become an important topic for both policy
makers and end-users. Although industry considers energy as a vital resource, the demand
for energy efficiency measures and technologies depends on the marginal cost of both
energy and energy efficiency measures due to its inelasticity [22,23]. Removing these
barriers will reduce the discount rate and the risk of energy efficiency [24,25] and increase
the enthusiasm for energy efficiency investments.
The gap in energy efficiency is due to the existence of barriers, and one of the most
important factors is the cost-effectiveness of the technology [26–28]. Barriers are mech-
anisms that prevent investment in technologies that are both energy and economically
efficient [12,29]. Barriers prevent cost-effective energy measures from always being used,
and the inconsistency between optimal implementation and real-time implementation is
called the “energy efficiency gap” [30,31].
Blumstein et al. [32] conducted one of the first studies to systematically classify bar-
riers to energy efficiency into six categories: misplaced incentives, lack of information,
regulation, market structure, financing and adaptation. According to Hirst and Brown [33],
barriers are divided into structural barriers (beyond the control of the individual) and be-
havioral barriers (dependent on end-user decision-making). Fuel price distortion, fuel price
uncertainty, limited access to capital, government fiscal and regulatory policies, codes and
standards, and infrastructure provision are classified as structural barriers, and attitudes
towards energy efficiency, perceived risk of energy efficiency in investment, information
gaps and misplaced incentives are classified as behavioral barriers [33]. Howarth and
Andersson [34] examined the theory of the energy-using equipment market to show that
incomplete information and transaction costs can affect rational consumers and they noted
a lack of information, principal-agent problems, consumer uncertainty, and lack of access
to credit to be barriers in the energy-using equipment market. Eyre [35] conducted psy-
chological, social and institutional analyses and classified barriers as information barriers,
cost-benefit separation problems, barriers to accessing capital, tariff barriers, externalities
and limited rationality.
In addition, in the Third Assessment Report (2001), the Intergovernmental Panel
on Climate Change (IPCC) classified the sectors and technological barriers to reducing
greenhouse gas emissions into eight categories, namely technological innovation, prices,
finance, trade and environment, market structure, institutional framework, information
provision, and social, cultural and behavioral norms and standards. While Weber [36]
categorized the barriers into institutional, market, organizational and behavioral barriers,
Fleiter et al. [37] introduced twelve different characteristics of energy efficiency measures
from the areas of relative advantage, technical context and information context. In addition,
J. Mar. Sci. Eng. 2021, 9, 1113 4 of 25

there are different types of methods for empirical studies, e.g., Thollander and Palm [38]
classified and described different barriers that may not fully explain the existence of the
“energy efficiency gap” as follows: distortions in energy prices, different perspectives on
energy, government fiscal and regulatory policies, constraints in the supply infrastructure,
codes and standards, and structure.
The energy efficiency gap arises mainly from market failures. Market failure means
that the market does not have the characteristics of the perfect market [13,39]. According
to Adam Smith’s theory, cited in [40], the four types of market failures are imperfect mar-
kets, imperfect competition, imperfect information and asymmetric information. Market
failures are the main barriers preventing the introduction of energy efficient technologies
and related investments [41] and may lead to public intervention [12,42–44]. However,
Backlund et al. [45] distinguish between market failure and barriers, as they consider mar-
ket failure to be any deviation from the characteristics of the perfect market, while barriers
are broader and include economic, organizational and behavioral barriers.

2.1. The Economic Perspective

2.1.1. Market Failure Barriers
Information problems are one of the most critical barriers to market failure. While
Gillingham et al. [13] and Chai and Yeo [46] classified the information problems as infor-
mation deficiency, asymmetric information and principal-agent, Cagno et al. [47] consider
asymmetric information as an incomplete information problem that leads to the following
additional barriers: split incentives, adverse selection and principal-agent relationships.
The lack of information on the energy efficiency of measures and technologies has
a negative impact on industry decision support systems (DSS) [48]. Since investments
in energy efficiency measures in industries (especially energy-intensive industries) are
rare and are due to rapid technological development and change [49,50], as well as due to
the difficulty of quantifying and measuring energy savings [51,52], distortion of decision-
making in the selection of appropriate measures and technologies can lead to a more
serious problem.
Asymmetric information is a particular form of incomplete information and occurs
when the parties involved in the transaction have different levels of information [53,54]
and can lead to inappropriate energy efficiency decisions. Asymmetric information leads
to split incentives, adverse selection and principal-agent relationships [55].
Divided incentives are related to the fair distribution of investor benefits from energy
efficiency investments [56,57], e.g., one department in a company may not be responsible
for energy use [58]. In addition, a typical case in industry may be that managers who
have only a short period of employment are reluctant and hesitant to invest in projects,
measures and technologies with a long payback period [59]. However, by considering and
emphasizing the life-cycle costs of investments, greater energy savings and benefits can
be achieved.
Negative selection occurs when the buyer does not have sufficient information about
the quality, efficiency and effectiveness of the technology and measure [60,61]. Since energy
efficiency characteristics are not tangible and sensitive, most industrial decision-makers
pay more attention to other factors, such as the original price and visible aspects, such as
color and design [62]. This type of approach leads to a negative selection of the required
technologies and measures.
The principal–agent relationship arises when principals do not have sufficient knowl-
edge, experience and training to strictly monitor the energy efficient activities of the agent.
This type of approach can lead to energy efficiency measures being overlooked [63]. For
example, the principal–agent relationship was found to have a negative impact on the
energy efficiency of passenger cars in the Netherlands [64].
J. Mar. Sci. Eng. 2021, 9, 1113 5 of 25

2.1.2. Non Market Failure Barriers

The barriers that are not linked to market failures are categorized as hidden costs,
capital costs, risks and heterogeneity.
Hidden costs are those costs that are usually overlooked when accounting for the bene-
fits associated with energy efficient technologies [65,66]. The hidden cost includes overhead
costs related to investments, data collection and analysis, and production disruptions [67].
Access to capital costs is one of the most important barriers to non-market failure.
However, according to Golove and Eto [68], it is better to classify this as an information
problem, because access to capital costs is a factor that examines the creditworthiness and
feasibility study of a project. If an energy efficiency project has a high rate of return due
to the high risk, it is an increased risk to implement the projects. Many consumers have
access to capital only at costs above the average return on capital in the economy [37,69]
and this poses a major challenge for SMEs, low-income consumers and end-users [70,71].
Risks can prevent the adoption of energy efficient measures and the use of energy
efficient technologies. Risks fall into three categories: technical risks, business risks and
external risks [72,73]. Taking into account the high discount rate and rejecting a certain
type of technology or measure may justify rational action in view of the risks [74,75].
Heterogeneity is about the fact that even some cost-effective energy-efficient technolo-
gies may not be suitable for certain industries or regions [76]. This type of barrier is more
common for industries and firms that have a particular or single type of production, or for
technologies that have limitations due to temperature or pollution [38,77].

2.2. The Behavioral Perspective

In addition to economic barriers, behavioral barriers can increase the energy efficiency
gap. These barriers are classified in terms of information, credibility and trust, values,
inertia and limited rationality [78], and are discussed below.
Information plays a crucial role in reducing the energy efficiency gap. Information
must be personal, specific, vivid, clear and simple, and provided in a timely manner [79].
Credibility and trust in the provider of the information are crucial to closing the energy
efficiency gap. If the industrial user cannot access appropriate information on energy and
cost efficient technologies, they may approach the most accessible and credible types of
information [80].
Norms and habits lead to behavior and behavior shapes values. Norms, habits and
values are very closely related to culture [81]. If values, such as support for others and
environmental awareness, are highlighted in the group, it will influence individuals to take
energy efficiency measures [82]. However, these values and mindsets need to be translated
into everyday behavior and practices to promote culture.
Inertia represents the combination of the effect of treating gains differently from
losses, giving greater weight to specific outcomes compared to uncertain outcomes, and
reducing regret [83,84]. Habits and routines shape the behavior of individuals and orga-
nizations [85,86] and it is usually difficult to change behavior because individuals strive
to prevent changes in the environment [87]. If the behavior and culture of the group has
an environmental commitment, there is less inertia in using energy-efficient technologies
and implementation takes less time and costs less [38,88]. Limited rationality is another
barrier that can increase the gap in energy efficiency. Policy makers are surrounded by
many constraints, such as time and information resource constraints, leading them to
replace imprecise decisions and rules of thumb with optimized decisions [89]. Due to
the conflicting interests and goals of individuals, DMs find themselves in a complicated
situation of optimizing decisions to get the lowest price, but with the highest efficiency and
effectiveness [90,91].

2.3. The Organization Perspective

Companies and firms are composed of different individuals and departments that have
different cultures, priorities, interests and cross-sector interaction and integration [65,73].
J. Mar. Sci. Eng. 2021, 9, 1113 6 of 25

As explained when discussing the constraints on bounded rationality, optimized decision

making in such an ambiguous situation becomes complicated. It is important to note
that the organizational formation can lead to barriers to the adoption of energy efficient
technologies. The related barriers are categorized into power and culture.
In terms of power, it is due to the lack of resources and different interests between
different departments within the company that there is a hidden conflict on how to achieve
and use the limited resource. Based on the structure, perception and priorities of the
company, managers have different levels of power. If the company’s energy manager has
a low status, energy measures and related efficient measures may be given low priority
accordingly [52].
Culture is formed by norms, knowledge and ideology and is closely related to val-
ues [92]. Developing organizational culture by raising awareness of environmental issues,
related problems and solutions can promote energy efficiency within the company [93].
Moreover, it can justify the failure to introduce energy efficient technologies [94].

2.4. Energy Efficiency Gap and Energy Policies

Since the energy system is a socio-technical system [95], not only the technology, but
also the environment, such as behaviors, regulations, organizations and individuals, are
crucial to the development of the system, which leads policy makers to understand indus-
trial energy efficiency to be a multidimensional subject [96]. To improve energy efficiency, it
is important to take a multi-faceted view of the barriers to energy efficiency and to consider
policy measures to address both market failures and barriers. This type of approach leads
to the classification of barriers into different areas, such as policy, management, financing
and information. Some other studies seek to rank barriers based on the area in which the
study is conducted, but the validity and importance of barriers in this type of study is
limited to the location and time in which the study is conducted [46].
To overcome the barriers to energy efficiency, the European Commission proposed
white certificates, voluntary agreement (VA) and long-term agreement (LTA) programs,
financial support programs, and promotion of the energy services market [97]. The main
objective of the measures is to provide funds to support energy efficiency measures and
programs and to increase the demand for energy efficient technologies by raising the price
of energy. However, similar to the carbon leakage phenomenon [98], an increase in the
price of energy may lead to a relocation of activities to areas with lower energy prices [99].
In addition to the above measures, education and information programs, such as energy
labelling [100], energy audits [101], energy management systems [102] and training and
certification of energy managers, are being implemented in order to raise awareness [103].

Energy Management System

Energy contributes to the socio-technical system and the environment influences the
development of the system [104]. To address the problems of energy efficiency, a systematic
approach must be considered. This means considering different perspectives and angles to
address the barriers [105]. Energy conservation is a necessary measure for addressing the
barriers, and plays a crucial role in preventing the energy gap from widening [44,106].
Although less attention has been paid to holistic and systematic analysis and intercon-
nectedness of barriers to energy efficiency, energy management aims to improve energy
efficiency in industry by analyzing how barriers are interconnected, interlinked and inter-
act. It also involves the design, development and implementation of long-term strategies
by a committed and educated energy manager [107,108].
Nagesha and Balachandra [109] suggested that barriers may have multiple structural
levels. Interpretive structural modeling (ISM) is a type of decision-making framework
used in energy conservation to identify and rank the barriers to energy efficiency and
analyze the interactions between them [110,111]. In the paper “Barriers to energy efficiency
improvement and decision-making behavior in Thai industry”, a decision-making frame-
work for energy efficiency investment is proposed [112], while the Motivation, Capability,
J. Mar. Sci. Eng. 2021, 9, 1113 7 of 25

Implementation and Result (MCIR) framework was proposed by Chai and Yeo [46] to
overcome barriers to energy efficiency through a systems approach. Cagno et al. [47] con-
sidered external and internal barriers and the actors affected by the barriers. They aimed
to overcome and determine the overlap between barriers, the impact of barriers on firms’
decision-making, implicit responses and the relationship between barriers. Prashar, [113]
by implementing a PDCA (Plan, Do Check, Act) cycle, encouraged a continuous investi-
gation of energy saving opportunities and associated energy saving activities (ESAs) to
promote energy optimization, and by developing a management energy efficiency ma-
turity (EEM) framework, promoted a systematic and continuous integration of energy
efficiency practices.

3. Methodology
There is a lack of studies that address a systematic approach to removing barriers
to improving energy efficiency in the maritime cluster. In disciplines where established
methodologies and research are lacking, the development of frameworks and conceptual
designs are more useful, efficient and effective than robust and rigorous models [114,115].
The validity of the proposed conceptual framework can be supported by mathematical
methods.
Figure 1 shows the research design and sequences of the study. First, a systematic
literature review was conducted from journals, conference proceedings, industry and
company reports in different sectors and the maritime business cycle. Barriers to energy
efficiency improvement from different industries and sectors were listed. In the second
phase, various energy barriers from the list (the list created in the previous phase) were
analyzed and the barriers related to energy efficiency improvement in the shipbuilding
industry were identified. In order to overcome the energy barriers, it is important to take
into account the surrounding characteristics of the energy sector. Therefore, in the third
phase of the study, the authors conducted an extensive literature review and identified and
classified the most important and crucial disciplines for solving the energy efficiency prob-
lems in shipbuilding into five main disciplines, namely, operations, policy and legislation,
technology and innovation, human factors (consisting of behavioral and organizational
systems) and economics. The barriers (from the list in phase two) were categorized under
these five disciplines (see Figure 2).
In order to reduce the energy gap, overcome the problem and prioritize the barriers
within the five disciplines, four2 criteria were selected: how severe the barriers are, how
easy it is to remove the barriers and how much impact they have on energy efficiency
and economic performance. However, it is important to note that the barriers within the
categorized framework are interlinked, interrelated and interact with each other. In the last
step, the interactions between the barriers from different disciplines have been clarified
using two3 multi-criteria decision making (MCDM)4 methods, namely, fuzzy analytical
hierarchy process (FAHP) and fuzzy multiple attribute group decision making technique
(FMAGDM)5 . The framework helps managers in a fuzzy atmosphere to identify and rank
the barriers according to their preferences through a transdisciplinary approach and to
make rational investment decisions to improve the energy efficiency of their portfolio.

3.1. Systematic Literature Review

As part of step one of this research, a total of 1142 works were collected that concerned
the energy sector and related barriers to energy efficiency in different clusters. A title
screening excluded 369 duplicated and unrelated studies. Furthermore, the abstracts and
conclusions of the articles were reviewed and 583 academic articles that did not discuss the
topic of the review and those with a weak context were excluded. The filtered academic
articles, i.e., 190, were carefully read and, after excluding articles with repeated findings,
the barriers were categorized into different disciplines, as shown in Figure 2. As a result,
the five disciplines became the corners and pillars of the “diamond”, as a new and holistic
form of barriers to energy efficiency in construction phase of ships (see Figure 2).
 J. Mar. Sci. Eng. 2021, 9, 1113 8 of 25
J. Mar. Sci. Eng. 2021, 9, 1113 8 of 26

Figure 1. Research design and sequences of the study.

Figure 1. Research design and sequences of the study.

3.1.
3.2.Systematic
The SystematicLiterature Review
and Transdisciplinary Approach to Solve the Problem
As part of step one of this
Systematic and the transdisciplinaryresearch, a total of 1142 have
approaches worksbeenwereusedcollected that con-
to develop the
cerned
proposed theframework.
energy sector Theand related barriers
systematic approachtoisenergy
helpfulefficiency in different
in addressing clusters.and
large, complex A
title screening excluded 369 duplicated and unrelated studies. Furthermore,
transdisciplinary issues, and it promotes a shift from linear to circular thinking [116]. A the abstracts
and conclusions
systematic of theinvolves
approach articles were
lookingreviewed and 583
at problems academic
from differentarticles that did[117]
perspectives not dis-
and
cuss the topic of the review and those with a weak context were
defining a boundary for the system to identify its components [118,119]. In addition, excluded. The filtered
the
academic
relationshipsarticles,
and i.e., 190, werebetween
interactions carefully read
the and,components
system after excluding articles
should with repeated
be emphasized.
findings,
Theretheare
barriers
variouswere categorized
methods, into different
perspectives disciplines,
and theories thatasareshown
usedin toFigure
analyze 2. As
the
aenergy
result,system
the fiveanddisciplines became
solve related the corners
problems, but it and pillars oftothe
is important “diamond”,
synthesize as a new
the technology
and holistic
through an form of barriers toapproach
interdisciplinary energy efficiency in construction
in a life cycle perspective phase of ships
to explore (see Figure
the relationship
2).
and interaction between the barriers and different stakeholders [18,19]. Energy system
is a socio-technical system. This means that in order to solve energy problems, not only
3.2.
theThe Systematic
technical and Transdisciplinary
solution and technology Approach to Solvebut
are important, thealso
Problem
the environment, e.g., the
human factor, social
Systematic and the aspects, activities, policies
transdisciplinary approaches and have
legislation, economic
been used and financial
to develop the pro-
factors must be considered, and, instead of one-dimensional thinking,
posed framework. The systematic approach is helpful in addressing large, complex and the interdisciplinary
approach must be
transdisciplinary takenand
issues, intoitaccount
promotes [106]. Multidisciplinary
a shift and transdisciplinary
from linear to circular thinking [116].are A
two levels of interdisciplinarity [120–122]. While interdisciplinary
systematic approach involves looking at problems from different perspectives researchers cling to their
[117] and
languages,
defining conceptsfor
a boundary andthemythologies, transdisciplinary
system to identify its components researchers
[118,119].integrate methods,
In addition, the
approaches, concepts
relationships and languages
and interactions betweenand the create
systema components
new language, method
should and concept to
be emphasized.
solve the problem
There are various[123,124]. A transdisciplinary
methods, perspectives and approach
theories allows
that are theused
researcher to break
to analyze the
through silos and cross the boundaries between disciplines [125,126].
energy system and solve related problems, but it is important to synthesize the technology
through an interdisciplinary approach in a life cycle perspective to explore the relation-
ship and interaction between the barriers and different stakeholders [18,19]. Energy sys-
tem is a socio-technical system. This means that in order to solve energy problems, not
 factors must be considered, and, instead of one-dimensional thinking, the interdiscipli-
nary approach must be taken into account [106]. Multidisciplinary and transdisciplinary
are two levels of interdisciplinarity [120–122]. While interdisciplinary researchers cling to
their languages, concepts and mythologies, transdisciplinary researchers integrate meth-
J. Mar. Sci. Eng. 2021, 9, 1113 ods, approaches, concepts and languages and create a new language, method and concept 9 of 25
to solve the problem [123,124]. A transdisciplinary approach allows the researcher to
break through silos and cross the boundaries between disciplines [125,126].

Figure
Figure2.2.Categorization
Categorizationof
ofkey
keybarriers
barriersin
inthe
theshipbuilding
shipbuildingindustry.
industry. Resource;
Resource; [12,22,32,33,38,44,69,71,74,106,127–132].
[12,22,32,33,38,44,69,71,74,106,127–132].

3.3. Develop a Holistic, Systematic and Interdisciplinary Framework

A conceptual holistic, systematic and transdisciplinary framework is proposed, based
on DMs’ priorities and on the feedback process. Figure 3 shows the process of identifying
and ranking the energy efficiency barriers based on the DMs’ preferences, which consist
of five key steps, namely “goal formation”, “system analyzing”, “scenario construction”,
“multi-criteria decision making (MCDM) assessment” and “strategy formation”.
J. Mar. Sci. Eng. 2021, 9, 1113 10 of 25
J. Mar. Sci. Eng. 2021, 9, 1113 11 of 26

Figure 3.
Figure 3. Transdisciplinary
Transdisciplinary approach
approach to
to design
design and
and develop
developthe
theframework.
framework.

3.3.1.The
Fuzzy Analyticalstarts
framework Hierarchy Process
with “goal (FAHP) At this stage, based on the priorities
formation”.
and interests of the
As discussed indecision
Section 3makers in terms
and shown of energy
in Figure 3, theefficiency
FAHP methodmeasures, the barriers
was used to de-
are classified
termine into five
the weight of disciplines. However,
main disciplines depending
and criteria. on stakeholders’
Triangular interests,
fuzzy number (TFN)the
disciplines
demonstrates may three numbers of A = (𝑎, 𝑏, 𝑐); as a, b and c are lowest potential value,
bychange.
higherIn potential
the second step,and
value “system analysis”,
the highest the importance
potential number of of the number
fuzzy disciplines will be deter-
ñ respectively.
A is a triangular fuzzy number, where 𝑎 ≤ 𝑏 ≤ 𝑐, has the following function [134];
mined using the FAHP method on the basis of the priorities of the DMs.
 J. Mar. Sci. Eng. 2021, 9, 1113 11 of 25

“Scenario construction” is the third step in the creation of the framework, where
DMs, using the FAHP, determine the importance of the criteria based on their preferences.
Here, the question is whether the four criteria have the same importance across the five
disciplines. If the criteria have different importance in each discipline, the DMs can create
a matrix of comparison weights for each criterion in each discipline.
The fourth step is the “MCDM assessment”. In this step, by referring to the results of
steps two and three and using fuzzy multiple attribute group decision making (FMAGDM),
the DMs can rank the barriers with respect to criteria in each discipline.
“Strategy building” is the final step in the framework. Based on the results of the
previous step, managers have a better understanding of the types of barriers and their
importance from different perspectives. There is no silver bullet for overcoming energy
efficiency [133]. The results of the framework help administrations to make rational
decisions on the selection and adoption of energy efficiency measures and to trade-off
between different solutions, as well as to select and design appropriate energy efficiency
policies. The framework is supported by the PDCA cycle to make the framework dynamic
so that it can be updated with new technologies and DMs’ priorities.
J. Mar. Sci. Eng. 2021, 9, 1113 12 of 26
3.3.1. Fuzzy Analytical Hierarchy Process (FAHP)
J. Mar. Sci. Eng. 2021, 9, 1113 12 of 26
As discussed in Section 3 and shown in Figure 3, the FAHP method was used to
determine
sdisciplinary approach to design and develop the framework.the weight of main disciplines and criteria. Triangular fuzzy number (TFN)
demonstrates by three numbers of A 0 = (a, b, c); as a, b and𝑋c<are 𝑎 lowest potential value,
⎧𝑋 − 𝑎
zzy Analytical Hierarchy Process (FAHP) higher potential value and the highest ⎪ potential number𝑎of≤fuzzy 𝑋 ≤ number 𝑏 ñ respectively. A
is a
J. Mar. Sci. Eng. 2021, 9, 1113
triangular fuzzy number, where0 ⎪a𝑏≤ − 𝑎
b ≤ c, has the 𝑋 following
<𝑎 function [134];
𝜇ñ⎧9,(𝑋used
discussed in Section 3 and shownJ.in Figure 3, the FAHP method was ( )
Mar. J.Sci.
Mar.Eng.Sci.2021,
Eng. 9,
J.2021,
Mar.
11139,Sci.
1113Eng. 2021, − 𝑎 to𝑐 de-
) =
1113 −𝑋
the weight of main disciplines and criteria. Triangular fuzzy number ⎪
⎪ 𝑏 − 𝑎 ⎨(TFN) 𝑎 ≤ 𝑋 𝑏≤≤𝑏𝑋 ≤ 𝑐
𝑐0− 𝑏 X<a
rates by three numbers of A = (𝑎, 𝑏, 𝑐); as a, b and c are lowest ( ) potential
𝜇ñ( ) = 𝑐 − 𝑋 ⎪
⎪ X0−a
 value, 𝑋>𝑐
otential value and the highest potential number of fuzzy numberµ ñ⎨ 𝑏 ≤≤𝑋X≤ ≤
a 𝑐 b

( X )
⎩
respectively.
𝑐 =− 𝑏
b− a
⎪
( A) c − X
b≤X≤ c 0 𝑋<𝑎
angular fuzzy number, where 𝑎 ≤ 𝑏 ≤ 𝑐, hasThe thegeometric
followingmean function ⎪ 0 
[134];
technique [135] was used𝑋to>calculate
 c−b
𝑐 the0⎧fuzzy𝑋 − 𝑎weights in<this
⎩ 𝑚𝑒𝑎𝑛 𝑜𝑓 0 X>c 0 ⎪ 0 𝑋 𝑎 𝑋
𝑎≤ 𝑋< 𝑎 ≤
study. While Ƒ𝑖 = 𝑔𝑒𝑜𝑚𝑒𝑡𝑟𝑖𝑐 𝑡ℎ𝑒 𝑖 th row, and Ꞷi ⎧𝑋 =−⎧ 𝑎 𝑋
fuzzy ⎪ 𝑏𝑎weight
− − 𝑎 ⎧𝑋of − the 𝑖 th
𝑎
( )
The geometric
event; mean mean technique [135] was [135]used was to calculate ⎪
⎪𝜇ñ
the
𝑏
⎪
)𝑎=
fuzzy
− ⎪ 𝑏−
𝑐−𝑋⎪
weights ⎪in𝑏𝑎this
≤ 𝑋𝑎 ≤ 𝑋 𝑏 ≤ 𝑏𝑎
⎨ 𝑎( )ofweights − 𝑎in this
The geometric technique used to calculate ( the fuzzy
Ƒ𝑖 = Ƒ= 𝑔𝑒𝑜𝑚𝑒𝑡𝑟𝑖𝑐 𝑚𝑒𝑎𝑛 𝑜𝑓 𝑡ℎ𝑒 𝑖 Ꞷi( ) = fuzzy ( ) 𝑖 𝑏≤𝑋≤ 𝑐
study.study.
WhileWhile i= (ñi,1 ⊗ ñi,2 ⊗mean
geometric ... ñi,n) o f 1/n th row,
the ith row, and and 𝜇ñři ( )= =fuzzy
𝜇ñ ( ) = weight
weight
𝑐 − 𝑋𝑐 ⎪ −
𝜇ñ𝑋 𝑐(of− 𝑏 ith
=
) the
the 𝑐− th 𝑋
event;
event; = ((𝑎 i,1X 𝑎 i,2 X 𝑎 i,3 … X 𝑎 i,n) 1/n, ⎨ ⎨ ⎪ 0 ⎨ 𝑏 ≤ 𝑋𝑏≤≤ 𝑐𝑋 ≤𝑋 𝑐> 𝑐𝑏
𝑐 − 𝑏 𝑐 ⎩− 𝑏 𝑐 − 𝑏 (1)
Ƒ==(ñi,1 ⎪
⎪ 0⎪ ⎪ 0 ⎪
⎪ 0 𝑋 > 𝑐𝑋 > 𝑐
(ñi,⊗1 ñi,2
⊗ ñi, ⊗ ...2 ñi,n)
⊗ . . 1/n. ñi, n) 1/n
(𝑏 i,1X 𝑏 i,2 X 𝑏,i3 … X 𝑏 i,n) 1/n, (𝑐 i,1 X 𝑐 i,2 X ⎩ 𝑐 i,3 ⎩ … X 𝑐 i,n) ⎩1/n)used to
==((𝑎 i, 1X𝑎a i,2
(( ai,1X i, 2XX 𝑎a i,i,33… . . .XX 𝑎a i,i,n)
The geometric
n) 1/n,
1/n, mean technique [135] was
(1) (1)calculate
(( b𝑏 i,i,2
1XXb i,𝑏,i3 b, Xi3𝑏 . The study.
geometric
bThe While
i, n(𝑐)geometric mean TheƑ𝑖mean =X 𝑔𝑒𝑜𝑚𝑒𝑡𝑟𝑖𝑐
technique
geometric
c𝑐i,technique
2 X…c[135] mean 𝑚𝑒𝑎𝑛[135]
was X𝑜𝑓
.techniquei,𝑡ℎ𝑒
cused
was 𝑖1/n
n)used
to th
[135]row,
calculate
)to was and
calculatethe
used Ꞷi to= c
fuzzy
the
(𝑏 i,1X 2 X… . . X1/n,
i,n) i,1 X (𝑐c i,i,21 X
1/n, i,3 i,
X3𝑐. .i,n) 1/n)
study.study. event;
While Ƒ𝑖 =
While 𝑔𝑒𝑜𝑚𝑒𝑡𝑟𝑖𝑐
Ƒ𝑖
study.
Ƒ
= 𝑔𝑒𝑜𝑚𝑒𝑡𝑟𝑖𝑐
While 𝑚𝑒𝑎𝑛
Ƒ𝑖 = 𝑚𝑒𝑎𝑛
𝑜𝑓
𝑔𝑒𝑜𝑚𝑒𝑡𝑟𝑖𝑐
𝑡ℎ𝑒 𝑜𝑓 𝑖 𝑡ℎ𝑒
th 𝑚𝑒𝑎𝑛
𝑖
row, th 𝑜𝑓
row,
and 𝑡ℎ𝑒
Ꞷi
and 𝑖 =
th Ꞷi
fuzzy
(2) row,
= fuzz
we
an
event;event; Ꞷi = event; Ƒ=i (ñi,1 ⊗ ñi,2 ⊗ ... ñi,n) 1/n
Ƒ ⊕Ƒ …….⊕Ƒ
ři = 1/n, (2)
Ƒ= 1⊕
Ƒ Ƒ=
(ñi,1 .=ñi,2
. ((𝑎
2 ⊗.(ñi,1 ... Ƒ=
. . .⊗⊗. i,1X
ñi,2
⊕ n⊗𝑎(ñi,1
ñi,n) ...i,2
1/n X
ñi,n) 𝑎1/n
⊗ ñi,2 i,3 X 𝑎 1/n
…ñi,n)
⊗ ... i,n)(2)
After determining the Ꞷi fuzzy = (weights) for disciplines and criteria (Ꞷi), the geometric
𝑎i,1X
mean After
method determining
was usedthe to fuzzy
Ƒ ⊕Ƒ= …….⊕Ƒ
obtain(weights)
((𝑎(𝑏=i,1X
the defuzzified
((𝑎
i,1X
for 𝑏(DF)
disciplinesi,2𝑎
i,2 XX=i,2𝑎((𝑎
𝑏,i3
mean
i,3
X i,1X
and 𝑎
…
of the
X𝑎𝑏…
…criteria
i,3
X 𝑎i,2
i,n)X𝑎1/n,
i,n)
X(ři),
weights
𝑎i,n)
1/n, i,3
the(𝑐1/n,
[136]:
…i,1XX 𝑎𝑐 i,n)
geometric i,2 X 1/n,
𝑐 i,3
mean method was
After determining the fuzzy (weights)used to obtain the
(𝑏 i,1X for𝑏i,1X
(𝑏 i,2 𝑏
defuzzified X i,2
disciplines
[(
𝑏,i3
(DF)
(𝑏X…
) (
𝑏,i3
mean
i,1X
and 𝑏…
𝑏 i,n)
X criteria
)]
X 𝑏X1/n,
of
i,2 the 𝑏,i3
i,n)
(Ꞷi),weights
(𝑐1/n,
…i,1X(𝑐
the X𝑏 i,1
𝑐i,n)
[136]:
geometric i,2 𝑐 i,2
X 1/n,
X 𝑐(𝑐i,3
Xi,1 𝑐 Xi,3
… X 𝑐…𝑐 i,2
i
mean method was used to obtain the defuzzified DF Ꞷi = (DF) mean of the weights [136]: (3)
[(ci − ai ) + (bi − ai )] Ꞷi =
Ƒ
DF ři = [( ) ( )] Ꞷ Ƒ ⊕Ƒ …….⊕Ƒ (3)
DF Ꞷi = Then Wi = ∑ Ꞷ 3 + ai Ƒ Ƒ (3) Ƒ (4)
Ꞷi = Ꞷi = Ꞷi =
Ƒ ⊕Ƒ …….⊕ƑƑ ⊕Ƒ …….⊕Ƒ Ƒ ⊕Ƒ …….⊕Ƒ
After determining
ꞶDF ři the fuzzy (weights) for disciplines and cr
Then Wi
Thenmethod =
Wi ∑= Ꞷwas used to obtain the defuzzified (4)(DF) (4)mean of t
After mean
After
determiningdetermining ∑the
AfterDFři fuzzy
determining
the fuzzy (weights) (weights)
thefor fuzzy
disciplines
for(weights)
disciplinesandfor criteria
and
disciplin
criter
(Ꞷ
3.3.2. Fuzzy Multiple Attribute Group Decision Making (FMAGDM)
meanmean method method wasmean used
wasmethod used
to obtain to wasobtain
theused defuzzified
thetodefuzzified
obtain (DF)
the [( mean
(DF)
defuzzified
) mean
(of the of
)] weigh
(DF)
the m
w
3.3.2. The
Fuzzy fuzzy TOPSIS
Multiple (FTOPSIS)
Attribute Group method Decisionwas used Making to handle
(FMAGDM) FMAGDM. DF ꞶiThe = fuzzy TOP-
3.3.2. Fuzzy Multiple Attribute Group Decision Making (FMAGDM) [( ) (
[( ) )]
( )]
[( ) (
SIS method is a suitable method for solving group decision DFproblems
ꞶiDF Ꞷi =[137].DF
=FMAGDM. To Ꞷirank = Ꞷthe
The The
fuzzy TOPSISfuzzy TOPSIS
(FTOPSIS) (FTOPSIS)
method was method was
used to handle used to handle
FMAGDM. TheThen fuzzyWi The
TOP- fuzzy
= solution
alternatives, the method determines the distances to both the fuzzy positive-ideal ∑ Ꞷ
TOPSISismethod
SIS method a suitable is amethod
suitable for method solving for solving
group group decision
decision problems problems
[137]. [137].
To Ꞷ To rank
rank theꞶ the Ꞷ
(FPIS) and the fuzzy negative-ideal solution (FNIS) simultaneously Then Then Wi
[138].= Wi = Then Wi =
alternatives, the method determines the distances to both the fuzzy ∑ Ꞷ ∑ solution
positive-ideal Ꞷ ∑ Ꞷ
alternatives,Assumingthe method thatdetermines
each discipline the distances to both the the
has K alternatives, fuzzy positive-ideal
weight of each criterion solutionand the
(FPIS) and the fuzzy negative-ideal solution
3.3.2. Fuzzy (FNIS)
Multiple simultaneously
Attribute Group[138].Decision Making (FMAG
(FPIS) score
and the of thefuzzy negative-ideal
alternatives with solution
respect to(FNIS) simultaneously
each criterion [138].
can be calculated as follows:
Assuming that each discipline 3.3.2. has
3.3.2.
Fuzzy K Fuzzy
Multiple
The3.3.2.
alternatives,Multiple
fuzzy Attribute
Fuzzy
TOPSIS
the Attribute
Multiple
weight Group ofGroup
(FTOPSIS) Decision
Attribute
each Decision
method
criterion Making
Groupwas
andMaking
Decision
(FMAGDM)
used
the (FMAGDM
to Making
handle FM
score of the alternatives with respect Ծij
The = SIS [Ծij
Themethod
tofuzzy
each 1
fuzzy (+)
criterion
TOPSIS Ծij
isThe
TOPSIS 2
acan (+)
suitable
(FTOPSIS) be
fuzzy ···(+) Ծij
method
calculated
(FTOPSIS)
TOPSIS
method K ] as
methodfor solving
follows:
(FTOPSIS)
was used
wasmethodto
usedgroup
handle
to handle
was (5)
decision
FMAGDM.
used FMAG pro
to h
SIS alternatives,
SIS method methodis a suitable
SIS themethod
is a method
suitable method is
method determines
a forsuitable
solving
for solving
methodthe distances
group group
for
decision
solvingto problems
decisionbothgroupthe fuz
proble
dec[1
Ծij = Ꞷj[Ծij = (FPIS) [ ꞶjԾij
1 (+) 1 2(+) (+)Ꞷj ···(+)2 (+) Ծij K ] Ꞷj K ]
···(+) (6)
(5) simultaneou
alternatives,
alternatives, theand thethe
method
alternatives,
methodfuzzy
determines negative-ideal
determines
the method the distances
the solution
determines
distances
to both (FNIS)
theto the
both
distances
fuzzy
the fuzzypositiv
to both p
where Ծij K and Ꞷj KꞶjare (FPIS)
the (FPIS)
and
rating the
and
and Assuming
fuzzy
the
(FPIS)
fuzzy
importance that
negative-ideal
and each
negative-ideal
the discipline
fuzzy
weight solution has
negative-ideal
of solution
the (FNIS)
Kth Kdecision
alternatives,
(FNIS)
simultaneously
solution
simultaneously
maker. the weigh
(FNIS) [138].
sim
= [ Ꞷj 1 (+) Ꞷj 2 (+) ···(+) Ꞷj K ] (6)
However, in this study Ꞷj Assuming K, the score
Assuming of
that
weight theeachalternatives
that
forAssuming
discipline
eachmain
the with
discipline
that has respect
eachKhas
disciplines K to
alternatives,
discipline eachcriteria
alternatives,
and hascriterion
theKweight can
alternatives,
the
were ofbe
weight eachcalc
ofthc
Ծij K andthrough
where achieved Ꞷj K arethe theFAHPrating and
score method
score
of thethe importance
ofalternatives
the alternatives
score of weight
with
the
(see Section 3.3.1). Continuing respectof
alternatives
with the
respect
to Kth
each decision
with
to criterion
eachrespect
on from maker.
criterion
can
to each
be cancalculated
criterion
be calcula ca
as
Ծij = [Ծij 1 (+) this, Ծij 2a(+) fuzzy···(+) Ծij K
 After determining theƑ= =𝜇ñ
fuzzy
((𝑎 (ñi,1
=(weights)
( ) i,1X
⊗⎨ñi,2 𝑎𝑐 − i,2 𝑋
=...
⊗Ƒ= for
X((𝑎 (ñi,1
ñi,n) i,1X⎩⊗1/n
𝑎 disciplines
i,3 …
ñi,2 𝑎(Xi,2 𝑏)⊗𝑎≤
Then Xand ⎪𝑎𝑏 −
i,n)
𝑋Wi ≤ criteria
1/n,
i,3 𝑎Then
=𝑐1/n … XꞶ(Ꞷi), Wi𝑎 i,n) =Then the 1/n, Ꞷgeometric
Wi = Ƒ= = ((𝑎 Ꞷ i,1X 𝑎 i,2 X 𝑎
Then ⊗ ñi,2 Wi ⊗=(1)
Ꞷ ⎩ …X 𝑎 i
(4)Ꞷ i,31/n
mean method wasThe usedgeometric
to =obtain the 𝑐 −
defuzzified 𝑏 (DF) 𝜇ñ( mean ) = ñi,n)
... 𝑐−
of the 𝑋∑weights Ꞷ ∑ Ꞷ
[136]: ∑ Ꞷ
(ñi,1 ...∑ ñi,n) (4)(1)
(𝑏 i,1X((𝑎 𝑏 i,2
mean
i,1X ⎪
X⎪(𝑏 𝑎
𝑏,i3 0i,2
technique
=… X((𝑎 𝑏X𝑎𝑏i,2 i,1X
i,3 X…𝑎
[135]
𝑏,i3 Xi,2𝑎… was X𝑋 ⎨
i,n) used
𝑎>𝑏 1/n, i,3 to
…1/n, calculate
XX 𝑎(𝑐𝑐i,n) i,31/n,
Thethe 𝑏 ≤ fuzzy𝑋
geometric ≤
𝑐 =Xi,n) ((𝑎𝑐
weights i,1X mean
X 𝑎
in
X i,2
technique
this X X𝑎1/n) 𝑏 i,3
[135]
…1/n, X was 𝑎(𝑐i
study. While Ƒ𝑖 = 𝑔𝑒𝑜𝑚𝑒𝑡𝑟𝑖𝑐
i,1X
𝑚𝑒𝑎𝑛
i,n)
𝑜𝑓
1/n,
𝑡ℎ𝑒
(𝑐
𝑖
Xi,1 𝑐X 𝑐− i,n) 𝑏𝑐 and i,2
Ꞷi
i,1 X…𝑐(𝑏 Xi,2i,1X
Ƒ𝑖
𝑐𝑏 1/n)
=
i,3
i,2 …
𝑔𝑒𝑜𝑚𝑒𝑡𝑟𝑖𝑐
𝑏,i3 𝑐 (1)
𝑖
… i,n)
𝑚𝑒𝑎𝑛
i,n)
𝑜𝑓 𝑡ℎ𝑒 (1)𝑖t
⎩ [( ) ( )] th⎪
⎪ row,
0
study. = fuzzy While
𝑋
weight
> 𝑐
of the th
3.3.2.(𝑏 Fuzzy i,1X 3.3.2.𝑏 i,2
Multiple X
FuzzyDF 𝑏,i3
(𝑏
3.3.2. Ꞷi i,1X
Attribute
Multiple=Fuzzy
… 𝑏X 𝑏i,2Group X 𝑏,i3
i,n)
Attribute
Multiple 1/n,Decision
3.3.2. … (𝑐Attribute
Xi,1
Fuzzy
Group 𝑏 Xi,n) 𝑐Decision
Making
Multiple 1/n,i,2
Group X(𝑐event; 𝑐Attribute
(FMAGDM) i,1i,3X…𝑐(𝑏
Making
Decision Xi,2 𝑐Making
i,1X
(FMAGDM)
Group Xi,n)(3)𝑐𝑏 1/n) i,3
i,2
Decision … X 𝑏,i3
(FMAGDM) X 𝑐 Making … i,n) X 1/n) 𝑏 i,n) (FMAGD 1/n, (𝑐
event;
The geometric mean technique [135] was used to calculate ⎩ the fuzzy weights in this
J. Mar. Sci. Eng. 2021, 9, 1113
study. While Ƒ𝑖 = 𝑔𝑒𝑜𝑚𝑒𝑡𝑟𝑖𝑐 The Ƒ=
fuzzy (ñi,1
The The
𝑚𝑒𝑎𝑛 ⊗
TOPSIS ñi,2
fuzzy
geometric𝑜𝑓Then ⊗
𝑡ℎ𝑒Wi ...
(FTOPSIS)
TheTOPSIS ñi,n)
𝑖meanfuzzy
th=Ꞷi 1/n Ꞷ
(FTOPSIS)
method
TOPSIS
technique The
Ƒ
Ꞷi was
(FTOPSIS)
method
fuzzy
[135] used TOPSIS to
was method
handle Ƒ used (FTOPSIS)
to FMAGDM.was
to
of handle used Ƒ= method
𝑖(4) (ñi,1
to
FMAGDM.The
handle ⊗ fuzzy ñi,2FMAGDM.
was ⊗(2)
used ... ñi,n)
The
TOP- 12tofuzzy in1/n
ofhandle
25 The TOP- FMA(2)Ƒ
fuzz
row, ∑ Ꞷ = and =
Ꞷi fuzzy=was used weight calculate the th the fuzzy weights Ꞷi =this
event; SIS method =SIS((𝑎is i,1X
method
a suitable
study. While Ƒ𝑖 = 𝑔𝑒𝑜𝑚𝑒𝑡𝑟𝑖𝑐 𝑚𝑒𝑎𝑛𝑎
SIS i,2
is method
a X
method 𝑎
suitable i,3 is …
for
a ƑX
method
SIS
suitable ⊕Ƒ
solving 𝑎method …….⊕Ƒ
i,n) for 1/n,
method
group
Ƒ 𝑜𝑓 𝑡ℎ𝑒 𝑖 th row,
solving
is a Ƒ decision⊕Ƒ
suitable
for group
Ƒ
…….⊕Ƒ
solving method
problems
decision group
and Ꞷi = fuzzy weight =for ((𝑎 problems
[137].
decision i,1X
solving To 𝑎 i,2
problems
rank
group
[137]. X the 𝑎
(2)of the 𝑖 th (2)Ƒ…ri
To i,3
decision [137].
rank … ƑX ⊕Ƒ
the
prob
To 𝑎
Ꞷithe =determines Ꞷi =to (1) Ꞷi = fuzzy
alternatives,
Ƒ= (ñi,1 ⊗(𝑏ñi,2 alternatives,
⊗ ... the method alternatives,the determines
method alternatives,
method
the
Ƒ ⊕Ƒ …….⊕Ƒ distances determinesthe the distances method
both
Ƒ ⊕Ƒ …….⊕Ƒ the the distances
to
determines
fuzzy both positive-ideal
the to fuzzy
both
the distances
the
positive-ideal fuzzy
solution topositive-ideal
both solution
the Ƒ ⊕Ƒ … s
After event;
i,1X
determining 𝑏 ñi,n) X 1/n
i,2After 𝑏,i3
the … X 𝑏(weights)
determining
fuzzy i,n) the 1/n, (𝑐
fuzzyfori,1 X 𝑐 i,2
disciplines
(weights) Xandfor 𝑐 disciplines
i,3 criteria
After…(𝑏 𝑐 i,n)
X determining
i,1X
(Ꞷi), and 𝑏the 1/n)
i,2
criteria X 𝑏,i3
geometric the(Ꞷi), … Xthe
fuzzy 𝑏(weights)
i,n)
geometric 1/n, (𝑐
fo
3.3.2. Fuzzy Multiple (FPIS) Attribute
and (FPIS)
the
𝑎 Assuming Group
fuzzy
X 𝑎used and (FPIS)
Ƒ= Decision
negative-ideal
the (ñi,1 fuzzyand
𝑎was ⊗i,n)ñi,2 Making
the
negative-ideal
fuzzy (FPIS)
solution
⊗defuzzified
... (FMAGDM)
negative-ideal
and (FNIS)solution
the fuzzy simultaneously
(FNIS)
solution
negative-ideal simultaneously
(FNIS) [138]. simultaneously
solution [138]. (FNIS) [138].
simultaneousl
=mean
((𝑎 i,1Xmethod
After
i,2was
determiningmean i,3
method
After
…
that
the
X
todetermining
obtain
each
fuzzy used
the
discipline 1/n,
(weights) the toñi,n) obtain
has
fuzzyfor K 1/n the (DF)
alternatives,
disciplines
(weights) defuzzified mean and
for mean of
the the
disciplines (DF)
method
weight weights mean ofand was each[136]:
ofcriteriaused
the criterion weights to(Ꞷi), obtain and [136]: the defuzzi
The fuzzy TOPSIS Assuming
(FTOPSIS) Assuming
thatmethod each
= ((𝑎 i,1X Assuming
discipline
that
was each
used𝑎 respect discipline
that
has
i,2 X 𝑎 to
to handleK eachAssuming
alternatives,
i,3 …discipline
has
FMAGDM. X K 𝑎 that
alternatives,
i,n) thehaseach The
1/n, Kcriteria
weight After
discipline
alternatives,
fuzzy the ofdetermining
(Ꞷi),
TOP-each
weight (1)
has the
criterion
the K ofgeometric
alternatives,
weight
each the and fuzzy
criterion
ofthe the
each (weights)
the geometric
and
criterion
weight thefoo
score of the alternatives with each criterion can be calculated as follows:
(𝑏 mean
i,1X
SIS method is a suitable 𝑏 i,2method
X 𝑏,i3
score of method the was
…mean
score X 𝑏
used
alternatives
offormethod
i,n)
thescore to 1/n,obtain
alternatives
solving with was
(𝑐
of the i,1 used
the
respect
group X
alternatives
DFwith 𝑐
defuzzified
Ꞷi
to
score
decision i,2 obtain
to=respecteachX[( 𝑐 i,3
Ƒof with the the)(DF)
criterion
problems DF to (… defuzzified
alternatives
respect
each
Ꞷi =canX mean )]𝑐 [( mean
i,n)
criterion
[137].
of
tobeeach the
1/n)
with )(DF) ( weights
method
calculated
To can
rankcriterionmean
respect )] was
be the
[136]:
of
calculated
asto can used
the
follows:
each weights
be criterion to obtain
calculated
as(2) (3)follows: [136]: the
can DFas be defuzzi
(1)
follows
Ꞷicalcu =(3)Ƒ
[(

(𝑏 i,1X 𝑏 i,2 X Ꞷi
𝑏,i3 = … X 𝑏[( i,n) 1/n, ) ( positive-ideal (𝑐 i,1
)] [( X 𝑐 i,2
) ( solution X 𝑐 )] i,3 … X 𝑐 i,n) 1/n) Ꞷi =
alternatives, the method determines the distancesDF to 1 both Ƒ ⊕Ƒ the …….⊕Ƒ fuzzy Ƒ ⊕Ƒ [( …
Ծij ij == Then Ꞷi
[Ծij[ Ծij =ij1 =1(+) (+) Ծij
[Ծij DF
Ծij 2ijꞶThen Ꞷi
=2(+)
1(+) (+) = [Ծij
···(+)Ծij ···(+) 2=Ծij
1(+) (+) KԾij Ꞷ ]Ծij
···(+)ij =K2]Ծij (+) [Ծij K···(+) ]1 (+)ԾijԾij K (3) ]2 (+) (5)···(+) DF Ꞷi
(5) Ծij(5) =K(3)]
(FPIS) and the fuzzy negative-ideal (FNIS)Ksimultaneously Wi = Wi (4) Then (4)W
After determiningsolution the fuzzy Ƒ (weights) for disciplines ∑ Ꞷ [138].
Ꞷ and criteria ∑ Ꞷ determining
After Ꞷ (Ꞷi), the geometric
(2) the fuzzy (weights) fo
Assuming that each discipline Ꞷihas = K alternatives, the weight of each criterion
mean method was used toƑ obtain ⊕Ƒ Ꞷj …….⊕Ƒ =the defuzzifiedThen
1[ Ꞷj Ꞷj1= (+)[ Ꞷj Wi =
∑ Ꞷj2Ꞷ
Ꞷj
(DF) Then
1=(+) Ƒ (+)
mean Wi
···(+)
[ mean Ꞷj
Ꞷj =
of1Ꞷj 2the (+)
∑ K Ꞷj
(+)
method ]and
···(+)
weights
Ꞷ Ꞷj
= 2Ꞷj the [ Ꞷj
(+)
was K ···(+)
[136]: ]used
1 (+) ꞶjtoꞶj (4)
Kobtain ]2 (+) (6) the ···(+)defuzzifThen
Ꞷj(6)
(2) K(4)W ]
score of the alternatives with respect to each criterion řj = [can řj 1Ꞷi be(+) = řj 2 (+)as
calculatedƑ ⊕Ƒ …….⊕Ƒ ···(+) follows: řj K ] (6)
3.3.2.
After determining Fuzzy
where ԾijMultiple
the 3.3.2.
fuzzy
where
K Fuzzy
Attribute
Ծij
Ꞷj
and (weights) where
KKare andMultiple
theԾij
Ꞷj
for Group disciplines
rating
K are
and Attribute
KDecision
where
the
and Ꞷj=rating Kthe[(and
are Group
Ծij Making
importance
the ) ( Decision
criteria
and
K and ratingtheꞶj (FMAGDM)
)](Ꞷi), 3.3.2.
importance
and K
weight are theMaking
the Fuzzy
the geometric
of
importancerating
the
weight (FMAGDM)
Multiple
Kth and of
decision weight
the Attribute Kth
importance maker.
of theGroup
decision Kth weight
maker. Decisio
decision [(
DF Ꞷi (3) DF Ꞷi = of
mean method3.3.2. was The where
Fuzzy
used
However, fuzzy Ծij
toMultipleij
3.3.2.=
K
TOPSIS
After
obtain
However,
in this and
The [Ծij
Fuzzy řj
(FTOPSIS)
fuzzy
Attribute
determining
the study 1 K
defuzzified
However,
in (+)
thisare
Ꞷj
Multiple Ծij
TOPSIS the
Group
study
K,in 2
method
the (+)
rating
Attribute
(DF)
the fuzzy
this Ꞷj···(+)
(FTOPSIS)
Decision
However,
weight and
was
mean Ծij
K,(weights)
study the the
used
Groupof K
Ꞷj
for method
Making in ]
importance
the
weight to
K,
the this forhandle
Decision
weights the was
(FMAGDM)
mainstudy
for 3.3.2.
disciplinesweight usedweight
FMAGDM.
the The
Making
[136]: Ꞷj
Fuzzy
disciplines to
main forfuzzy
K, and of
handle the
(FMAGDM)
Multiple
the
the (5)TheTOPSIS
criteria
disciplines
and Kth
FMAGDM.
weight
main fuzzy decision
Attribute
(Ꞷi),
criteria (FTOPSIS)
TOP-
disciplines
forand the were
the The maker.
Group
geometric
criteria fuzzy
method
main andwere TOP-
Decisio
discip
criter wa
However, in this study řj K, the weight for the Ꞷmain disciplines and criteria were achieved
SISachieved
method mean
The fuzzy TOPSIS is a
SIS
through
achievedsuitable
method
method the
The(FTOPSIS) method
was
achieved
throughFAHP
fuzzy TOPSIS is aused suitable
the
[( method
for
method
through to FAHP solving
obtain
Then
) ((FTOPSIS) method
achieved(see
the method
was)]Wi thegroup
FAHP
Section =
used for
defuzzified
through
method (see solving
decision
method
to 3.3.1). Section
handle the
was usedSIS group
(DF) problems
Continuing
(see FAHP method
3.3.1).
FMAGDM. mean
Section
Thetofuzzy decision
method [137].
is
of
Continuing
handle on a
3.3.1).the
The suitable
from
TOPSIS problems
(see To
weights
FMAGDM.
fuzzy rank
Continuing
this,
Section
on method
(FTOPSIS) from
TOP-a the
[137].
[136]:
fuzzy 3.3.1).
(4)The fuzzy this,
on Tofor from
method rank
solving
Continuin
a fuzzy
Then this,
TOP- the
wa W
through
alternatives, Ꞷj
the = FAHP
the
alternatives,
method [DF Ꞷjdetermines
Ꞷi 1method
the=(+)method Ꞷj (see 2the (+)determines ···(+) Ꞷj 3.3.1).
Section
distances K∑to]the Ꞷ
both Continuing
distances
the alternatives,
fuzzy to on
positive-ideal
both fromthe
the (6)
(3)method this,
fuzzy a
solution fuzzy
positive-ideal
determines multi- the
solution distan
multi-criteria
SIS method is aSIS multi-criteria
group
suitable
method multi-criteria
decision-making
method group
is a suitable decision-making
for solving group multi-criteria
problem
method decision-making
group matrix
for solving problemgroup
decision [( can decision-making
SIS matrix
)groupproblem
be
problems
( method as follows:
can
decision
)] matrix be
[137].
is a suitable as problem
can
follows:
problems
To rank be as
method matrix
follows:
the
[137]. To can
for rank be
solving as the fo
(FPIS) criteria
and the group
fuzzy
(FPIS) decision-making
negative-ideal
and the fuzzy solution problem
negative-ideal DFmatrix
(FNIS) Ꞷisolution = can be
simultaneously (FPIS)
(FNIS) as follows: andsimultaneously
[138]. the fuzzy negative-ideal
[138]. solution (3) (
where Ծij K and Ꞷj K are the
alternatives, thealternatives,
method
rating and determines
Then thetheWi method
importance = theDecision
Ꞷ determines
distances weight of tothe both
the distances
Kth the alternatives,
fuzzy
decision to positive-ideal
both maker. the
the (4) method
fuzzysolution positive-ideal
determines the solution distan
3.3.2.Assuming
Fuzzy Multiple that Attribute
Assuming
each discipline Group
that each ∑ Ꞷ
has K discipline
alternatives, Making
Criteria has
Criteria K
the (FMAGDM) 3.3.2.
alternatives,
weight
(Attributes)
(Attributes)
Criteria Fuzzy
Assuming
of
Ꞷ (Attributes) each
Criteria the Multiple
weight
criterion that
(Attributes) each
ofAttribute
and each discipline
the criterion
Criteria (Attributes) Group has and Decisio
K alter
the
However, in (FPIS) this studyand the
Alternatives Ꞷj fuzzy
Alternatives
(FPIS)
K, thenegative-ideal
Alternatives
and weight the fuzzy
Alternatives for the solution
negative-ideal main (FNIS)
Alternatives disciplines
Then solution
simultaneously
Wi = (FPIS)
and (FNIS) criteria andsimultaneously
[138]. thewere fuzzy negative-ideal
[138]. solution (4) (
score of
The the
fuzzy alternatives
score
TOPSIS of the with
(FTOPSIS) alternatives
respect method to witheach was respect
criterion
used to can each
handle be ∑calculated
score
criterion Ꞷ
FMAGDM. The of the fuzzycan alternatives
as be The follows:
TOPSIS calculated
fuzzy with
(FTOPSIS)
TOP- as respect
follows: method to each wa c
Assuming
achieved through the FAHP method (see C1 that Assuming
each disciplineC1 that each
has
Section 3.3.1). Continuing C1 K C2
discipline
alternatives,
C2 C1 has C2 K
the ….alternatives,
on .from weight
.. C2 C1 Assuming
of….
this, a fuzzy each the …. weight
criterion
... ….
that C2 ….
each
of and each Cm
discipline
theCmcriterion…. …. Cmhas and K alter
the C
SIS method is a suitable method Ծ11 for solving
Ծ11 Ծ12 group decision SIS problems
method [137].
is a suitableTo rank method the for solving
3.3.2. Fuzzy Multiple
multi-criteria score Attribute
groupofdecision-making
theA1 alternatives
A1score of
Group A1the Decision
with
problem alternatives
Ծij 11 respect
= matrix
A1 Making
[Ծij towith 1eachcan(+) Ծij(FMAGDM)
12 be
Ծij=Ծ11
respect
criterion
A1 as 2 (+) Ծ12
follows:
[Ծij to···(+) can1each (+).be
…. .Ծijscore
Ծij Ծ12
.criterion K Ծ11
calculated2]of (+)…. the can
···(+) …. alternatives
as.be .Ծij.follows:
…. Ծ12
K ] …. Ծ1m
calculated with as
Ծij
(5) 1m respect
follows:
….
= …. Ծ1m [Ծij to 1each (+) (5)ԾcԾ
J.J.Mar.
Mar.Sci.
Sci.Eng.
Eng.2021,
2021,9,
9, 1113
1113 alternatives, 3.3.2.
A2 the method
Fuzzy A2 determines
Multiple
The fuzzy TOPSIS (FTOPSIS) method was used to handle FMAGDM. The fuzzy TOP- Ծ J. Mar.
12 A2 Attribute
Sci. the
Ծ
Eng. 12 distances
Ծ
2021, Group
22 9, Ծ 1113
A2 12 to
Decision
Ծ both 22 …. the alternatives,
Making Ծ fuzzy
Ծ 22 12 …. positive-ideal
(FMAGDM) …. the method
…. Ծ 22 …. solutionԾ determines
2m 13
…. 13 ….ofԾ
of 26 the
262m distan Ծ
A2fuzzy negative-ideal Ծij 12= [Ծij 1 (+) Ծij 22 Ծij= 2 (+) [Ծij ···(+)1 (+). .Ծij .ԾijK2](+) ···(+) . .Ծij . K] Ծij
(5) 2m = [Ծij 1 (+) (5)(Ծ
SIS method is(FPIS) and the
a suitable method
…. The fuzzy for
….solving Ꞷj. Criteria
TOPSIS =…. group [solution
(FTOPSIS) 1 Ꞷj
Ꞷj(Attributes)
.decision
(FNIS)
.(+)method=Ꞷj …. . 2[ (+)
problems Ꞷj simultaneously
.was ···(+)
1 .used (+)
[137].(FPIS)
ꞶjꞶj K.to2To
and
.]handle
(+) [138].
.rank
theFMAGDM.
···(+) . the
fuzzy
Ꞷj.K. ]negative-ideal . The . Ꞷj (6)fuzzy = . . [TOP- solution
.Ꞷj 1 (6) (+)
J. Mar.J.
Alternatives
Sci.
Mar. Eng.
Sci.2021,
Eng.
J. Mar.
2021,
9, Assuming
1113
Sci.9,
Eng.
1113 2021, . .
J.
9,. that
Mar.
1113 J. each
Sci.
Mar. Eng. discipline
Sci. 2021, .
Eng.
J. Mar. 9,
2021,
1113 Sci.has 9, Eng.
1113 K alternatives,
2021,
J. Mar.. 9, Sci.
1113 Eng. the
2021, weight
.
9, 1113 Assumingof each criterion
. that each and discipline
the . has 13 K
of alter
26
13 of
alternatives, the methodSIS C1
…. method….is
determines the C2
a distances
suitable
Ꞷj . rating
=…. method
to .both
Ꞷj …. Ꞷj.(+) for
the =Ꞷj …. solving
fuzzy
. 2[ and Ꞷj.…. group
positive-ideal . (+) ꞶjꞶj .decision
2.] the Cm
. solution .problems ꞶjꞶj .Kof .K [137].
.are . Ꞷj To = . rank . [ and .Ꞷj the
. Mar. Sci. Eng. 2021, 9,(FPIS)
1113 and whereofԾij
score the K and where
alternatives Ꞷj K Ծij are with Kthe and respect Ꞷj [Kand are
to
1the
eachthe importance
rating
criterion
(+)
can
···(+)
1the weight be importance
where
score
K
calculatedof of
(+)
Ծijthe
···(+)
Kth
Kweight and decision
alternatives
as follows:
] the
13 maker. Kth
with
of the
26
(6) decision
rating
respect to
1the(6)
maker.each
(+)imc
A1 the fuzzytransformation Ծ11
alternatives,
negative-ideal
... .
transformation
An An the
was
was Ծ12
solution Ծmethod
used
used .n1 An (FNIS)
toto determines
Ծweight
transform
transform simultaneously
….
n1Ꞷj …..K, Ծ
the
the
Anthe n1 distances
various
various …. …. [138].…. . to….
criteria
transformation
criteria Ծboth n1 Ծ1m
scales
scales
…. the…. fuzzy
into
into
was . …. positive-ideal
aa….
used ….Ծ
comparable
comparable towere transform
nm . …. solution
scale.Ծ nm
scale.
…. the were vario Ծf
However, in
ԾijԾ(FPIS) this
However,
Ꞷj study Ծij Ꞷj
in this
K, the study
Ꞷjfuzzy for the the weight main However,
disciplines
for the Ծij2m main in and this disciplines
criteria
Ꞷjof study Ꞷj and K, criteria
the weight
A2 where
Assuming that each Kdiscipline
Therefore,
Therefore,Ꞷ12and where and
the
the Knormalized
Ꞷ
are
the
has Ծfuzzy
normalized Kthe
K22 and
Ծij
Ꞷ1
rating
alternatives,
=Ꞷ
Kand
negative-ideal
fuzzy are
Ꞷ1
the
the
the
1decision
…. decision
Ꞷ2
importance
rating
. solution
.weight
Ծij Ꞷ2matrix
Ꞷ1 matrixand
(+)Ꞷ2 of…. each the
weight
(FNIS) importance
Ծij
ŔŔTherefore,
. was
.was
where
.criterion
. Ꞷ2
of Ծ
simultaneously
achieved
achieved
Ꞷ1
the
….the andKth
Kweight andthe
by: decision
normalized
.by: Ꞷ2
Kthe
[138]. are maker. Kth
the decision
ꞶnԾij
fuzzy ratingdecision Ꞷn
and maker. the im
matri Ծ
achieved
However, J. through
Mar. J.in
Anachieved
Mar.
Sci.this
However,
Eng.
Sci. the
study
Eng.
2021, FAHP through
2021,
9, Ꞷj
in
1113 9,
n1
method
this
K,
1113 thethe
study
[Ծij FAHP
(see
weight Ꞷj
(+) . .method
Section K, for the the 3.3.1). (see
weight
···(+)
main
….Continuing
Section achieved
However,
disciplines
for
K ]3.3.1).
the on
through
main
….
in and
. . .….
Continuing
from this disciplinesthis,
the
criteria
study
…. FAHPa on fuzzy
Ꞷj
were
(5)
from
and
nm
method
K,
=…. ….
this,
criteria
the
[Ծij a 1fuzzy
(see
weight
(+)
were SecꞶ f
score of…. the alternatives with . Assuming
respect
transformation
transformation to .each that
transformation
was each
criterion
was
used discipline
usedto . can
transform
was to be has
transformation
transform
used K
calculated
transformation alternatives,
the
to . transform
various
the as
transformation
was follows:
various the
criteria
used
was the weight
criteria
. used
various
to transformation
scales
transform
was
to of
scales each
transform
criteria
intoused into
the
a criterion
to
comparable
scalesvarious
transform
was
athe comparableand
used
various
into the
criteria
a to
scale.
the
compa trans
criter
var
sca
sc
multi-criteria
achieved through řgroup
multi-criteria
achieved the decision-making
FAHP through ř1group
method the decision-making
FAHPproblem
(see ř2
Section method matrix
ŔŔ =[ŕij]m×n,
=[ŕij]m×n,
3.3.1). problem
(see can . be
. multi-criteria
Continuing
Section . . as
achieved matrix follows:
3.3.1). on can
through . group
be
. . . as
Continuing
from this,
the follows:
decision-making
FAHPa onfuzzy řn
frommethod this, (7)
(7) problem
(see
a Ŕ fuzzy=[ŕij]m
Sec
…. transformation score
Ծij, was ∀i; ofTherefore,
Therefore, the
used Ծij, alternatives
to
the
Ꞷj,∀i; Ꞷj
.transform
Therefore, =1,2,…,
normalized
jthe Ծij, [ Ꞷj
with
normalized
∀i; the
the respect
1fuzzy
..various (+)
normalized fuzzy Ꞷj
Therefore, jto
decision
Ծij, each
=2linguistic
criteria
Therefore,(+)
decision
∀i; j···(+)
fuzzy the criterion
scales
matrix Ꞷj
nTherefore,
normalized
matrix
decision
the ŔKinto ]can
normalized
was = Ŕ1,2,…, be
a.matrix comparable
was
achieved
the fuzzycalculated
Therefore,
normalized
achieved fuzzy
Ŕaredecision
was
by: as
scale.
decision
theachieved
by: fuzzyꞶj
follows:
(6)=described
matrix
normalized matrix Ŕ[was
decision
by: ꞶjbeŔ 1descr
fuzzy achie
was
matr (+)d
where J. Mar.
where
multi-criteria
where
J..Mar.
Sci.Ծij Eng.=j∀Eng.
ij,Sci.
group
and
2021,
multi-criteria
where
2021,
j[Ծij 9,
i;decision-making
and 1113
1 (+) 9,
řj, =and
1113
group jԾij=Ꞷj, 21,2, (+)jjnwhere
and
decision-making.=···(+)
problem
are , Ꞷj,
.1,2,…, nlinguistic
Ծij
are n
matrix K are
1,2,…,
] problem .linguistic
can
variables,
and
be
Ꞷj,
are variables,
multi-criteria
asmatrix
jlinguistic
variables,
which
follows: can
can
which n
variables,
(5)
group
be
which
as
be candescribed
linguistic
follows:
can
decision-making bewhich be
described by
variables,
can
problem
bywhi
Therefore, the
where
where 𝐵𝐵
normalized𝑎𝑛𝑑
𝑎𝑛𝑑 𝐶 𝐶 are
are fuzzy the
the setdecision
set
Ծijfuzzy of
of benefit
benefit matrix Ծij criteria
𝑏ij,(aij, criteria
=𝑐ij), Ŕ was Ծijand
and
𝑏ij, achieved cost
cost
𝑐ij),where criteria,
criteria,
𝑏ij, by: 𝐵 𝑎𝑛𝑑
𝑐ij),
Ծij 𝐶
respectively,
respectively, are the𝑏ij, set and
𝑐ij), and of benefit criteria and
An where ԾijԾtriangular
triangular
by Kn1 triangular
andfuzzy Ꞷj K numbers,
triangular
fuzzy
are fuzzy
…. the numbers, numbers,
rating = and (𝑎ij,
Ծij triangular
….ijnumbers,
the == Criteria (𝑎ij,
[Ծij
importance bij,fuzzy1and (+)
cij), Ŕ=Ꞷj
….
(Attributes) numbers,
(𝑎ij,
Ծij
and =Criteria
=[ŕij]m×n,
weight Ŕ (Ꞷj1,
2whereand (+)
=[ŕij]m×n,
řj =of Ꞷj
···(+) Ꞷj2,
Ծ Ծij
(řj1,
theŔ =nm
(Attributes) and
Ծij =Ꞷj3).
(Ꞷj1,
=[ŕij]m×n,
Kth
K řj2, (𝑎ij,
Kdecision
and Ꞷj] Ꞷj2,
řj3). The
Ꞷj =K (Ꞷj1,
TheꞶj3).
linear
are maker. Ꞷj2,
and
Ŕlinear
the Thescale
=[ŕij]m×n, ŔꞶj
rating Ꞷj3).
linear
scale
=[ŕij]m×n,= and (Ꞷj1,
The scale
(5)Ŕthe (7) Ꞷj2
line
=[ŕij C(
im
Alternatives Ꞷj AlternativesꞶj Ꞷj transformation transformation Ꞷj waswas usedused to
Alternativestransform
to transform the various
the various criteria criteria scales scales int
Ꞷ However, Ꞷ1
transformation
in this study = [ was Ꞷ2
C1 1
Ꞷjused (+)
K, the to transform2 (+)
ŔC2
weight ….
C1 ···(+)
=[ŕij]m×n, for
Criteriathe K
the ]
various …. ….
main
C2
(Attributes) criteria
However,
disciplines
Criteria Ꞷn
scales
…. ….
(Attributes) in and into (6) criteria
this a comparable
study Cm …. C1 Ꞷj
(7)
were K, scale.
the weight
Cm C2 Cf
Alternatives where where 𝐵 𝑎𝑛𝑑
Alternatives 𝐵 where 𝐶
𝑎𝑛𝑑 are𝐶the 𝐵 𝑎𝑛𝑑
are Therefore,
set the
Ꞷj of 𝐶set
Therefore,
= benefit
are where
of[the the
benefit
Ꞷj where 1 𝐵
normalized
the
criteria
set 𝑎𝑛𝑑
of
(+) criteria 𝐵Ꞷj
benefit𝐶𝑎𝑛𝑑
normalized
andwhere are
2 cost
and
(+) 𝐶fuzzy
the
criteria
are
Alternatives 𝐵criteria,
cost
···(+) set 𝑎𝑛𝑑
fuzzy the decision
Ꞷj of 𝐶K
criteria,
and
whereset decision
benefit
arerespectively,
]cost
of the 𝐵respectively,
matrix
benefit 𝑎𝑛𝑑
criteria,
criteria
set matrix 𝐶criteria
of Ŕare was
benefit
and
and Ŕthe
respectively, was
achieved
and
cost
and set achieved
criteriacriteria,
cost
of bene
(6) by:
and
cri
an
Ծ11 Ծ12Ծ11 ∗∗ ,, of Ծ12 𝑗𝑗 transform
∈∈to 𝐵;𝐵; Ծ1m Ծ11 ∗ , Ծ1m ∗Ծ12,scales
achieved Therefore,
through the
the normalized
FAHP method fuzzy (see
ŕij= decision
transformation Section
(( weight
∗∗,, cost
transformation matrix3.3.1). ),),was Ŕwas was
Continuing
used achieved used achieved
to on
through
transform by:
from this,
the FAHP
a fuzzy method (see Sect
where Ծij K and Ꞷj
where 𝐵K A1
𝑎𝑛𝑑are𝐶the areratingthe set A1
and C1
of the benefit importance criteria ŕij=C2 C1 and ∗∗criteria,….
….
the C2Kth decision
respectively, ….
A1
…. ….
…. maker. andthe various the Cm ….
…. various
C1 ŕij= criteria (criteria scales
Cm C2 ∗ ),int
where Ծij, ∀i;multi-criteria
j and Ꞷj, j =group1,2,…,decision-making
nA2 Ծare12linguisticTherefore, variables,
Ծproblem
Ծ Therefore, which
matrix
the Ծ
normalized
the can canbe
normalized beasdescribed
multi-criteria follows:
fuzzy fuzzy decision by
group
decision Ŕ matrix =[ŕij]m×n,
Ծ Ŕ….
decision-making
Ծ =[ŕij]m×n,
matrix Ŕ was Ŕ was Ծ
achieved Ծproblem
achieved
However, in this study A2
A1where Ꞷj K,Ծij the KA1 Ծ11Ꞷj Kfor
and
weight arethe the Ծ12Ծ11 22 12
rating
main and
Ŕdisciplines the ….
Ծ12
…. ×
22
importance and criteria ….
A2
weight
….
A1 ….
…. were of the Kth Ծ1m …. 2m Ծ11 12
decision maker. Ծ1m Ծ12 by
2m 22
triangular fuzzy numbers, Ծij = (𝑎ij, 𝑏ij, 𝑐ij), and Ꞷj = (Ꞷj1, ŕij= =
( ŕij= [ŕij]m
Ꞷj2, , ( Ꞷj3)., , n,
ŕij=
), , The ( ), , linear , 𝑗 scale
∈
), 𝐵;𝑗ŕij= ∈ 𝐵;
( ŕij= , 𝑗( ∈ , , 𝐵; ),ŕij= , ( ),(7) , ∗. 𝑗 , ŕij= ∈),𝐵
achieved through theA2 ….
FAHPHowever, method ….
inA2 Ծ(see .12 Section
this study Ꞷj ԾԾ.K,
3.3.1).
where .12Continuing
where
22 the𝐵 𝑎𝑛𝑑 𝐵 𝑎𝑛𝑑
weight∗ 𝐶 are Ծ ….∗ ∗.𝐶.on 22for∗ ∗
arefrom
the the
set
theof ∗ ∗ main
this,
set ∗ ….
benefit
of
….
A2 …. .a. benefit
∗
disciplines
fuzzy criteria Ŕ criteria ԾŔ
∗ and
=[ŕij]m×n, .2m
and
…. Ծ . .and
∗ ∗
=[ŕij]m×n,cost
12 ∗ ∗
criteria cost ∗
criteria, Ծwere
∗
criteria, Ծ.2mrespec22 ∗res
ŕij= ( ∗ , ,
∗ method ∗ ), Criteria 𝑗
(Attributes) ∈ 𝐵; .. , C)
multi-criteria group ….achieved
….
decision-making
where
Alternatives B and C through
….
….problem
are ..the the set of FAHP matrixbenefit ŕij= .. ..can
ŕij= ( 𝐵be
(criteria ,,𝑎𝑛𝑑 (see
as ,𝑎𝑛𝑑follows:
,and Section
),..),𝐶..cost 3.3.1).
criteria, 𝑗𝑗Alternatives
∈ ∈ C; C;Continuing
…. .. ..
respectively,
…. on and ..from.. .. this, ŕij= a( fuzzy , .. respec
where where 𝐵 𝐶 are the
are set
the ofset benefit
of benefit criteria criteria and and
cost cost
criteria, criteria, res
….multi-criteria
An ԾC1
An
…. group.n1 decision-making ԾC2
…. . .n1 problem ….
…. ….. . matrix can
….
An….
…. …. . .be(ŕij= as follows: ԾCm …. Ծ.C1
.nm n1
.), Ծ .nm C2
…. . 𝐵;
Ծ11 Criteria ŕij= , ( , , ), , 𝑗 ∈ 𝑗
𝐵; ∈
A1 ŕijԾ12 ∗Ծ1m
∗ Ծ11 Ծ12
bij cij…. ….
A1
Ꞷ
An Ꞷ
ԾꞶ1
An n1 Ծ Ꞷ2
…. (Attributes)
Ꞷ1=n1 ( aijc∗jŕij= , c∗(ŕij= , ,c∗…. ….(Ꞷ2
), ,, ŕij= ),j, ∈ ( ), B;, An Ꞷ
….𝑗….
…. ,∈ C;
∗
𝑗 ),∈ŕij=
∗
C; (Ծŕij= Ꞷn𝑗….…. ,Ծ∈ Ꞷ1
∗
(n1C;, , ŕij= ),, (Ծ),Ꞷn ,nmꞶ2𝑗 ∈, C; 𝑗
ŕi
Alternatives j …. …. nm ….
A2 C1 Ծ C2 12 ŕij= ( , Ծ …. ∗∗22
j
−…. Criteria (Attributes) ….
A2 Ծ Ծ
2m 12 Ծ 22𝐵;
a j 𝑗…. ∈ ∗ , ( ∗ ,, ∗∗Ꞷn , Ꞷ1 𝑗 𝑀𝑎𝑥 ∈Ꞷn 𝑗𝐵; ∈
Ꞷ Alternatives ꞶꞶ1 𝑐Ꞷ2𝑐Ꞷ1 ,= = a𝑀𝑎𝑥−
), a j 𝑐ij
j𝑀𝑎𝑥
−
𝑐ij ….
Ꞷ2 if if C; 𝑗𝑗 ∈∈ 𝐵; 𝐵; Cm
ŕij=
Ꞷ
…. …. (ŕij= ),…. ∗ ), 𝑐 ∗
= Ꞷ2 𝑐ij i
A1 where Ծij, ….Ծ11 ∀i; wherej and Ꞷj, Ծij,Ծ12 j. =∀i;1,2,…, j andC1 ŕijare
nꞶj, =…. ( cij
.j =linguistic 1,2,…, , bij C2 , naij )….
variables,
are ., linguistic j ∈
where which …. C;variables,
Ծij,
….
Ծ1m . can ∀i;be j and ….Ꞷj,
described
which . can j. = by be Cm
1,2,…, described n are. ling by
∗ = Max cij
Ծij c𝑏ij, 𝑐ij),
Ծij if
𝑏ij, 𝑐ij), ∗j ∈ B; Ծ Ծij ∈𝑏ij,
A2 where triangular ….Ծfuzzy
Ծij, ∀i; triangular
12where
j andnumbers,
A1Ꞷj, Ծij, Ծj.fuzzy =∀i;
22 1,2,…,
j and =numbers, Ծ11
(𝑎ij, nꞶj, are j ….
−
.j =linguistic 1,2,…,𝑐 ∗=and Ծ12
=(𝑎ij, 𝑐𝑀𝑎𝑥 ∗ Ꞷj
=are
nvariables, .𝑀𝑎𝑥
…. 𝑐ij = linguistic
(Ꞷj1,
𝑐𝑐ij triangular
if=and
where Ꞷj2,
which ….
𝑀𝑎𝑥 Ꞷj….
Ծij, Ꞷj3).
if variables, .ŕij=
𝑗𝑐ij 2m
can =fuzzy
∈ ∀i; (Ꞷj1,
(ŕij=
𝐵;𝑗be The
ifj∈,𝑐and 𝐵; Ꞷj2,
(∗described
numbers,
….
=
whichlinear
, 𝑀𝑎𝑥𝑐 ∗𝑗.Ꞷj3).
Ꞷj, ,= j.scale
),∈can =𝐵;),
𝑐ij
𝑀𝑎𝑥 by The
1,2,…,
be =Ծ1m
∗𝑗 (𝑎ij,
𝑐𝑐ij
if = ∈n𝑀𝑎𝑥
linear
described 𝑗C; ifare 𝑗.C; 𝑐ij∈𝑐ij)
scale
ling 𝑐𝑗∗
𝐵;
by
…. triangular fuzzy An . triangular A2
numbers,fuzzy Ծ n1 . Ծij 𝑐 ==∗ Ծ𝑀𝑎𝑥𝑏ij,
numbers,
(𝑎ij, 12 a ….𝑎 𝑎
j 𝑐ij =
. 𝑐ij),=
= Min 𝑀𝑖𝑛
Ծijif=and 𝑀𝑖𝑛 aij
Ծ 𝑎ij𝑎ij
(𝑎ij,Ꞷj 22 ….
𝑗𝑏ij,if
if if ∈ .= 𝐵; 𝑐ij),
(Ꞷj1, j
triangular ∈ 𝑗
and𝑗 C.
∈
….∈
Ꞷj2,Ꞷj 𝐶.𝐶.An
…. Ꞷj3). . =fuzzy (Ꞷj1, Thenumbers, ….
Ꞷj2,
linearԾ nmԾ Ꞷj3). n1
scale Ծij The 𝑎 Ծ =
= (𝑎ij, 2m
linear𝑀𝑖𝑛 …. 𝑎ij
𝑏ij,scale 𝑐ij)i
Ꞷ Ꞷ ŕij= (ŕij=, ( , ,), ), 𝑗 ∈ 𝑗C;∈ C;
…. . …. Ꞷ1. . Ꞷ2. . ….. . ….. . ꞶnꞶ1 . Ꞷ2
Byconsidering
By consideringthe thedifferent differentimportance importance
importance 𝑎 = 𝑎 𝑀𝑖𝑛 of of
= each each
𝑎ij
𝑀𝑖𝑛 𝑎 criterion,
criterion,
𝑎ij By𝑀𝑖𝑛
= considering 𝑗 𝑎ij
∗∈the
the 𝐶. 𝑗∗authors
authors∈ 𝑎𝐶. the = 𝑎different
could
could
𝑀𝑖𝑛
𝑗 ∈ = 𝐶.𝑎ij
𝑀𝑖𝑛construct
construct
construct importance
𝑎 𝑎ij = 𝑀𝑖𝑛 𝑗𝐵; 𝑎ijof
∈ 𝐶. 𝑎𝑗
An Ծweighted
n1 …. …. . …. . …. if if
. Ծ nm 𝑐 = 𝑐 𝑀𝑎𝑥 if = 𝑀𝑎𝑥 𝑐ij . 𝑐ij if if if 𝑗 . ∈ 𝑗
𝐵;if ∈
where the
the
Ծij, weighted
∀i; j and normalized
normalized
Ꞷj, j = 1,2,…,𝑎 fuzzy
fuzzy
= 𝑀𝑖𝑛
n are decision
decision
𝑎ij linguistic if matrix
matrix
matrix 𝑗
variables, ∈ as:
as:
as: 𝐶.the where weighted
which Ծij, can normalized
∀i; be j anddescribed Ꞷj, fuzzyj = by1,2,…, decision n are matrix lingu
Ꞷ Ꞷ1ṼṼ=== [ṽij]mxn, By Ꞷ2
Anconsidering i𝑖considering
= 1,2, Ծ.considering
n1 𝑗𝑗…. ,…. …. …. Ꞷn …. 𝑗Ծ nm
𝑖= = 1,2, =… .… ,.,𝑚,,𝑚, = 1,2,
=jdifferent
1,2, … 2,,By
… .𝑛𝑛.different Ṽṽij=ṽij= 𝑖By = 1,2, … , 𝑚, = 1,2, …∈ , 𝐵; 𝑛coul
∗(.) ∗(.)
triangular fuzzy By
[[ṽij]mxn,
ij]mxn,
numbers, Ծij 1, By
2, the
(𝑎ij, different
the
m,
𝑏ij, =
𝑐ij), 1,importance
the
and importance
considering
.Ꞷj(8),
,(8), By
n=(8), considering
where
where oftriangular
(Ꞷj1, importance
whereeach of By
Ꞷj2, =the each
criterion,
ŕij
[ṽij]mxn,
ŕij
ij= 𝑐the
considering
different
Ꞷj3). (.) =of
ŕij
fuzzy 𝑐Ꞷj.
criterion,Ꞷj.
𝑀𝑎𝑥
different
The =the
each řj.𝑀𝑎𝑥 𝑐ij
importance
numbers, the
authors
considering
criterion,
linear 𝑐ij
theimportance
different
authors
ifscale could
ifthe
Ծij of 𝑗importance
=each
the
could∈ofdifferent
authors 𝑗𝐵;
construct
(𝑎ij, criterion
each constru
𝑏ij, crit
(8)
𝑐ij) oi
By considering thefuzzy
The
The fuzzyꞶ
theweighted
weighted
the different
positive
positive the
normalized importance
weighted
normalized Ꞷ1
idealsolution
ideal solutionfuzzy normalizedof fuzzy
the each
decision
(FPIS,
(FPIS, Ꞷ2
weighted
the criterion,
decision A 𝐴
fuzzy ∗
∗
𝐴matrix
weighted ∗
)) and andnormalized
the
matrix the
decision fuzzy as:
The
fuzzy ….
authors
weighted
normalized as:
fuzzymatrix
negative
negative fuzzy thecould
normalized
positive fuzzy
as:
weighted ….
decision
ideal
ideal construct decision
solution
ideal
solution fuzzy
matrix
normalized
solution Ꞷn
matrix
decision
(FNIS,as:
(FNIS, (FPIS,as:
fuzzy matri 𝐴 d ∗
𝑎− =𝑎 𝑀𝑖𝑛 = 𝑀𝑖𝑛 𝑎ij ∗𝑎ij if if 𝑗 ∈ 𝐶. 𝑗 ∈ 𝐶.
where Ծij, the ∀i; weighted
j and𝐴 A𝐴Ꞷj,
− j = 1,2,…,
normalized
)) are:
are: 𝐴
A ∗
𝐴Ṽ === (ṽ
∗ nṼfuzzy
∗∗
1are
([ṽij]mxn,
(ṽ , ṽṽlinguistic
=11,,[ṽij]mxn, ∗∗ ∗2decision
2, 2 ,.Ṽ 𝑖,.…,
= 1,2,
𝑖ṽṽ=
.= ,[ṽij]mxn,
…, ∗ ∗ variables,
∗n
n…1,2,
nmatrix ),),),, and
𝑚,
and
and …𝑗𝑖,as: =𝑚,
A 𝐴𝐴Ṽ
− which
1,2,𝑗=== …… 1,2,
==([ṽij]mxn,
(ṽ (ṽ −
Ṽ, 𝑛 can
=𝑚,
1,1,…
1,𝐴
[ṽij]mxn,
(8), −
𝑗ṽ,ṽ)= be
𝑛2, where
2,
are:
2, .described
𝑖1,2,Ṽ=
(8), …,
.…, .=1,2,
…
𝐴 𝑖∗ṽṽ
where
,[ṽij]mxn,,ṽij=
=𝑛=… n),
n),1,2,
, ŕij
(8),
(ṽ 𝑚, by
∗where
ṽij=
1…
where 𝑗,Ṽ𝑖,=
(.)
where ṽ𝑚,=
=Ꞷj.
ŕij ∗1,2, 𝑗1,2,
2[ṽij]mxn,
(.) ∗
ṽṽ, = ∗Ꞷj.
ṽij=jj…
…, j==1,2,
…
=,(1,1,1)
𝑛 ,ṽ𝑚,
ŕij ∗…
(1,1,1) (8),
n(.) 𝑖𝑗,),=
𝑛Ꞷj.
=and 1,2,
where
andand1,2,…
(8), 𝐴…ṽij=
where , ,𝑚,𝑛= 𝑗(ṽ ŕiṽ
(8
triangular fuzzyṼnumbers, − where
ṽṽ j = (0,0,0),The
= [ṽij]mxn,
j = Ծij
(0,0,0), 𝑖 =Ծij,
= j(𝑎ij,
1,2,
= ∀i;
j == fuzzy
The …
1,2,
1,2,…,n.𝑏ij,
1,2,…,n. j, and
𝑚,
fuzzy
. . 𝑐ij),
𝑗
positive
. ,n. = Ꞷj,
The
The and
1,2,
positive
The j =
fuzzy… Ꞷj 1,2,…,
,
idealideal
distance
distancedistance 𝑛 By= (Ꞷj1,
(8),
positive
solutionof n
considering
By where
of are
solution
The
of alternatives Ꞷj2,
considering
ideal linguistic
(FPIS,
alternatives
alternatives ṽij=
fuzzy
The Ꞷj3). the
(FPIS,ŕij𝐴from
ṽfuzzy
solution ∗(.)
The
positive
jfrom )= and variables,
different
the from 𝐴
Ꞷj.
The
(0,0,0),∗linear
different
positive
(FPIS,
) fuzzy
FPIS
FPISand fuzzy
FPIS 𝑎
ideal jand scale
importance=
=fuzzy
and 𝑎which
∗importance
𝐴1,2,…,n.
ideal
and 𝑀𝑖𝑛
solution =
positive
negative
)FNIS
FNIS The
and FNIS 𝑀𝑖𝑛
𝑎ij can
solution
negative fuzzy
fuzzy
are
areTheof 𝑎ij
(FPIS,
ideal
ideal
are ifbe
each of described
ifeach
positive
(FPIS,
solution
negative
ideal
solution
calculated
calculated
calculated
distance ∗ 𝑗 criterion,
𝐴 )solution
criterion, ∈
and 𝐴
as
of 𝐶.
𝑗∗∈
ideal
as by
)(FPIS,
(FNIS,
fuzzythe𝐶.
and(FN
alternat solu au
th
fuzn𝐴
∗ Ծij ∗ 𝑏ij, 𝑐ij), ∗∗ Ꞷj
The as fuzzy triangular
follows:
follows:
follows: 𝐴positive 𝐴 ) fuzzy
) are: 𝐴∗ =𝐴(ṽ
ideal
are: ∗numbers,
∗
solution
)=1are: , ṽ∗ 1∗ 2𝐴(FPIS,
(ṽ ,the, ṽ…, (ṽ,=ṽ∗…,
=∗ 2weighted
the 𝐴1∗𝐴 n∗
(𝑎ij,
weighted
,) ), ṽare:
)and 𝐴
and2 normalized
n ,),fuzzy
)…, 𝐴 𝐴∗normalized
are:
and =and
∗negative
n𝐴(ṽ
ṽ=follows:
𝐴
(ṽ ),1)==and fuzzy
1,
,are:
(ṽ(ṽ ṽ∗ṽ∗=12fuzzy
𝐴𝐴(Ꞷj1,
ideal
1, 2, ∗ ∗ solution
ṽṽ…,
,, decision
…, =𝐴22, (ṽ,ṽꞶj2,∗
ṽ)…,
decision n
1are:
1,
n), ṽꞶj3).∗∗ (FNIS,
ṽand
,),matrix 𝐴
where
nn),
22, ∗ The
matrix
),, …, 𝐴
=…,
where
and (ṽṽ∗ṽṽ=∗1as:
as: linear
j∗𝐴(ṽ,= ṽ),
n),
n ∗∗ scale ∗
(1,1,1)
1,
=where
j2and
=(ṽ,ṽ(1,1,1)
…, 1,
2, 𝐴ṽ
and ṽ j=nan
…, 2,
=(
By=dconsidering Byij, = considering the different
the different importance ∗importance of each of each criterion, criterion, the au
th
𝐴 ) are: 𝐴∗ = (ṽṽ∗ 1j ,=ṽṽ(0,0,0), ∗
…,j =ṽṽ1,2,…,n.
j2=, (0,0,0), ∗ dand
jnj==),(0,0,0), ∗
i ∗∗=The
1,2,…,n.
𝑑𝑑 =
∑
= ∑
𝐴j =nj=
∑
Ṽ Ṽ
=1 ṽ(ṽ([ṽij]mxn,
distance
The
1,2,…,n.
𝑑(
𝑑(
j 1,
distance
ṽij,
ṽij,fuzzy
ṽof ∗
ṽ[ṽij]mxn,
(0,0,0),
Thej i2,=∗)alternatives
…,
(0,0,0),
ṽṽ∗i)i)normalized
𝑖
of =
i
distance
j = = ṽ1,2,
𝑖
alternatives =
1,2,…,n.
𝑖𝑖 =
1, jn),
j = =…
1,2,
2,
=fuzzy
.,
where
(0,0,0),
of 𝑚,
.
1,2,…,n.
1,2,
1,2,
from…
. , 𝑗,
m
alternatives
…
The
…decision
𝑚,
=
from
,,𝑚
ṽ
FPIS
j ṽ= 1,2,
𝑗 =
distance
The
𝑚 decision
1,2,…,n.
j = …
1,2,
FPIS
and ,
(1,1,1)
(0,0,0), 𝑛
distance …
from FNIS
andof ,
(8), The𝑛 and where
(8),
alternatives
j FPIS
=
FNIS
∗ are
of
1,2,…,n.
distance where
alternatives
and ṽij=
calculated
are
∑ (8)
(8)
FNIS ṽij=
ŕij
calculated
The
of
from (.)
alterna ŕij
are
dista
as Ꞷj.
from
FPI(.)
ca
ṽ∗ i)
ṽ j = (0,0,0), j = follows: 1,2,…,n. follows: The distance follows:−of alternatives
the weighted
the Theweighted
n follows:
Thefuzzy normalized
from
follows: positive positive
FPIS follows: ideal
andideal FNIS
fuzzy
solution solution
are
follows: (FPIS,
calculated
matrix(FPIS, 𝐴∗𝑑)as:
matrix 𝐴=∗ )as:
and
as and
fuzzy 𝑑(negative
(8)
fuzzy ṽij,nega
di =
𝑑𝑑 𝐴= =∑) ∑𝐴are: j∑ Ṽ1)=dare: Ṽ
([ṽij]mxn,
𝑑(𝐴 = =[ṽij]mxn,
∗ij,
∗𝑑(ṽij, 𝐴(ṽ
−
∗ ∗j) 𝑖 ∗=
ṽṽ=1j) ∗(ṽ , ṽi1∗= 1,2,
2𝑖, 𝑖=
∗𝑖∑
ṽ1,
,==…,
∗… 1,2,
22,
1,2, ,,.𝑚,ṽ.…
…, …
.∗ n ,𝑗,m
…,1,2,
𝑚,=
,ṽ𝑚
),𝑚 ∗ 1,2,
∗n
𝑗=
and ), and …
1,2,
∗𝐴 =
, 𝑛…𝐴(8), ,(ṽ 𝑛 =where (8),
1,∗(ṽ= 1,2,ṽij=
ṽwhere
∗∑
ṽ…, (9) 2,ṽij=
ŕij…,ṽ (.) ∗n),
ŕijṽꞶj. (.)
w n
=ṽij, ṽ∗1,2, ∗𝑑 ∗ 𝑑( ṽij,
=𝑖ṽṽ=∗…
follows: = j)𝑑(
𝑑 ∗The =𝑑The ∑
fuzzy ∑𝑑(positive
fuzzy 𝑑ṽij, =ṽṽij,
positive i)ideal i)𝑑(
ideal𝑖solution
ṽij,
= 𝑖𝑑ṽ=
solution ∗
i)=
…1,2, ,𝑑𝑚
(FPIS,∑ …=(FPIS,
𝑖, = 𝑚 ∑
𝑑(𝐴 1,2, ) 𝑑and
ṽij, 𝐴 𝑑(…∗= ṽ,ṽij,
) 𝑚
andi)∑ ṽ(9)
fuzzy
(9)i)𝑑(
fuzzy 𝑖𝑑 =ṽij,
negative 1,2,
(8)
nega ∑(
The closeness∗ coefficient ṽ j
of each = ṽ (0,0,0),
j = (0,0,0),
∗ alternative j = 1,2,…,n.
j = 1,2,…,n.
∗ to ∗ ,determine
The The
distance distance of
∗the ranking order
alternatives
of alternatives from of the from FPIS FPIS and an
F
The
The closeness
closeness 𝑑 =
coefficient
coefficient ∑ 𝑑( 𝐴 ofṽij, )
of follows:𝐴are:
each ṽ ) i)
each alternative 𝐴
are: ∗
= 𝐴
alternative to 𝑖 ∗=
(ṽ ∗ 1,2,
= 1 (ṽ , ṽ 1…to∑ 2 , ,𝑚
ṽ ∗
…,
determine
determine2
The closeness, ṽ
…, ∗
n ṽ
), n
and
the ),
the ranking and 𝐴
ranking
coefficient 𝐴
= (ṽ (8)
=
order
order1,(ṽ ṽ
of of 1, 2,
each
of ṽ …,
the 2,
thealternati ṽ
…, n), ṽ w n
alternatives is calculated by: follows: 𝑑 =𝑑 ∑= ∑ 𝑑(𝑑ṽij,𝑑( =ṽṽij, j) ṽ 𝑑( j)𝑖 =ṽij, 1,2,𝑖𝑑= ṽ …1,2, j),𝑑𝑚∑
= …= 𝑖, =𝑚𝑑( ∑1,2, ṽij,
𝑑… 𝑑(= ,ṽ𝑚 ṽij, j)∑ ṽ 𝑖j)𝑑( =𝑑ṽij, 1,2,
(9) 𝑖==ṽ…(
alternatives
alternatives is is calculated
calculated∑by: by: ṽ j =ṽ (0,0,0), j = (0,0,0), j = 1,2,…,n. j = 1,2,…,n. alternatives The The distance distance
is calculated of alternatives of alternatives by: fromfrom FPISFPIS andaF
𝑑 = 𝑑( ṽij, ṽ j) 𝑖 = 1,2, … , 𝑚 𝑑 ∗ =𝑑 ∗ ∑= ∑𝑑( ṽij, 𝑑(of ṽij, ∗ (9)
ṽranking i) ṽ ranking ∗
𝑖oforder
i)alternative = 1,2,
𝑖 to = …1,2, , the
𝑚… ,𝑚
The The closeness closeness Thecoefficient follows:
closeness
coefficient follows: dof − coefficient
each of
The eachalternative
closeness
Thealternative ofcloseness each The
coefficient
toalternative
determine
closeness
to coefficientdetermine ofThe to
each
thecoefficient
determine
closeness
the each
alternative coefficient
theeach order
ranking
of determi
alternat
to ofofde o
t
The closeness coefficient
alternativesalternatives isalternativesofiseach
calculated CC
calculated = C𝐶C𝐶
alternative
i by: isdcalculated= =by:
i
∗∗ − to
alternatives , , , determine
alternatives by: i 𝑖 𝑖= ==
is 1,2,
1, 1,2,
calculated2,
alternatives is ….
the… . ., ,
calculated𝑚
,𝑚 m
ranking by: isalternatives
calculated
by: order of isthe
by: C𝐶
calculated =(10)
(10)
(10) ∗ by: ,
∗
i + di 𝑑∗ = =𝑑∗ ∑∑ = = ∑∑ 𝑑( 𝑑(ṽij, ṽij,
𝑑( 𝑑(ṽij, ṽṽij,
ṽ∗ i)j)ṽṽ∗ i)j)𝑖𝑖 = = 1,2, 𝑖1,2,
𝑖= =… 1,2,
…
1,2, ,,𝑚 𝑚 ……,,𝑚 𝑚
alternatives is calculated by:
Figure
Figure 44 shows shows the the hierarchical
hierarchical C𝐶 prioritizing
prioritizing
= C𝐶 = , of
of
C𝐶 , the 𝑖
the= =
Figure
𝑑 ∗ =𝑑1,2,𝑖
barriers =
barriers ofto∑ …,1,2,
4 ,
=of 𝑚
shows …
to to
∑ 𝑖 ,
C𝐶
𝑑( =
𝑚
energy
energy 1,2,
ṽij, =the
𝑑( C𝐶 …
ṽij, , 𝑚
efficiency
hierarchical
ṽ∗ j) ṽ ∗to=efficiency
, C𝐶 ,
𝑖 determine
=to 𝑖 ==
1,2,
by
by 1,2, 𝑖 =
prioritizin
𝑖 determine
=… ∗1,2, …
,(10)𝑚… 1,2,
, , 𝑚C𝐶𝑚…
, (1
Figure 4 shows the hierarchicalThe The
prioritizing closeness closeness
∗
of∗coefficient
thecoefficient barriers each energy each alternative alternative
efficiency j)by AHP the
AHP
AHP for for aa shipyard.
shipyard. C𝐶 = ∗alternatives ,alternatives 𝑖 =is1,2, is…calculated
calculated ,𝑚 AHP for
by: by: a shipyard. (10)
for a shipyard.
Figure Figure 4 shows 4 Figure
shows the hierarchical 4the shows
The hierarchical
The the
closenessFigure prioritizing
closeness hierarchical
Figure prioritizing
4coefficient
shows 4 shows
coefficient Figure
ofprioritizing
thethe of ofhierarchical
the
barriers
4the
each ofshows hierarchical
each barriers
Figure
ofto
alternative the energy
prioritizing
alternative 4to
hierarchical
barriers
shows
energy
prioritizing
to efficiency toto the
determine efficiency
ofenergy
prioritiz
determinehierarch
theof by barr
the theef
Figure 4 shows AHPAHP theafor
for hierarchical
shipyard. AHP
a shipyard. for aprioritizing shipyard.
alternatives alternatives AHPof AHP
isfor the a for
calculated
is barriers
shipyard. aAHP
calculated shipyard. by: toforby: energy
aC𝐶 shipyard. =C𝐶efficiency
AHP ∗=
for ∗,
a shipyard. , 𝑖by = 1,2, 𝑖=… 1,2, , 𝑚… , 𝑚
AHP for a shipyard.
Figure Figure 4 shows 4 shows the the hierarchical C𝐶 =C𝐶prioritizing
hierarchical ∗= ∗,
prioritizing , 𝑖 =of1,2, 𝑖= the of… 1,2, , 𝑚…barrier
the
barriers , 𝑚 to
AHPAHP for afor shipyard.a shipyard.
J. Mar. Sci. Eng. 2021, 9, 1113 13 of 25
J. Mar. Sci. Eng. 2021, 9, 1113 14 of 26

Figure 4. The
Figure 4. The hierarchical
hierarchical prioritizing of the
prioritizing of the barriers
barriers to
to energy
energy efficiency
efficiency by
by AHP
AHP for
for aa shipyard.
shipyard.

4.
4. Results
Results and
and Discussions
Discussions
4.1. Case Study
To prove
prove that
that thethe framework
framework works,
works, interviews were conducted with five senior
managers (DMs) at an Iranian ship repair
(DMs) at an Iranian ship repair yard andand
yard an offshore infrastructure
an offshore builder
infrastructure (ISY).
builder
The interviewees had five main responsibilities for energy aspects in the shipyard.
(ISY). The interviewees had five main responsibilities for energy aspects in the shipyard. ISY
is anisactive
ISY shipship
an active repair yardyard
repair in Iran and builds
in Iran many
and builds offshore
many and subsea
offshore infrastructures.
and subsea infrastruc-
Figure Figure
tures. 3 shows3 the implementation
shows of the framework
the implementation for ISY. Allfor
of the framework steps and
ISY. Allsequences
steps andhave
se-
been summarized
quences have beenand the resultsand
summarized are the
presented
results in
areTables 1–4. in Tables 1–4.
presented

Table 1. Ranking
Table 1. Ranking of
of disciplines.
disciplines.

Disciplines Disciplines Technology

Technology & &Policy &
Policy &
Human Element Operation
Human Element Operation Innovation Regulation Economics
Economics
Companies
Companies Innovation Regulation
ISY ISY 0.031 0.031 0.140 0.140 0.144 0.144 0.125
0.125 0.560
0.560

Table 2. Ranking
Table 2. Ranking of
of criteria.
criteria.

Criteria Severity of Simplicity of Impact of Barriers Removal Impact of

Impact of Barriers Barriers
Removal to
to Energy
Impact of Barriers
Company Criteria
Barriers Barriers Removal Simplicity of Efficiency Removal to
Economic Performance
Severity of Barriers Removal to Energy
CompanyISY 0.102 0.118 Barriers Removal
0.267 Economic
0.513
Efficiency
Performance
ISY 0.102 0.118 0.267 0.513
J. Mar. Sci. Eng. 2021, 9, 1113 14 of 25

Table 3. Ranking of the ISY barriers in each discipline.

Barriers
T1 T2 T3 E1 E2 E3 E4 E5 E6 E7 E8 E9 E10 E11 O1 O2 O3 O4 O5 H1 H2 H3 H4 H5 H6 H7 H8 H9 H10 H11 H12 H13 P1 P2 P3 P4
ISY
Ranking 2 1 3 4 10 2 7 8 1 5 3 6 6 9 1 2 5 4 3 6 7 8 9 11 10 5 2 5 4 1 7 3 4 2 1 3

Table 4. Final ranking of ISY.

Barriers
T1 T2 T3 E1 E2 E3 E4 E5 E6 E7 E8 E9 E10 E11 O1 O2 O3 O4 O5 H1 H2 H3 H4 H5 H6 H7 H8 H9 H10 H11 H12 H13 P1 P2 P3 P4
ISY
Ranking 6 2 11 4 12 3 9 10 1 7 5 8 8 13 14 19 22 21 20 28 29 30 31 33 32 27 24 27 26 23 29 25 18 16 15 17
J. Mar. Sci. Eng. 2021, 9, 1113 15 of 25

4.1.1. Goal Formation

The taxonomy developed in the study is shown in Figure 2. The barriers were clas-
sified into five disciplines for ISY. These are: human factors, operations, technology and
innovation, policy and regulation, and economics.

4.1.2. System analysis

Table 1 shows ISY’s priorities with respect to five main disciplines. In the second
step of “system analyzing”, the weight of the disciplines for the case yard was determined
based on the prioritized disciplines by applying the FAHP. For calculating the determined
weight of each discipline, Equations (1)–(4) were used (see Section 3.3.1). As shown in
Table 1, finance (0.560) was the most important discipline for ISY, and technology and
innovation (0.144) was the second highest priority for ISY. Policies and regulations (0.125)
were identified as the third highest priority of the company and the human factor (0.031)
was identified as the least important discipline for the yard in the present case. There
was a serious imbalance between the disciplines. Interestingly, the importance of the first
priority (economics) in the ISY was 18 times higher than the least priority (human element)
in the ISY.

4.1.3. Scenario Construction

Scenario construction was the third step in the implementation of the framework. At
this stage, four criteria were introduced: the severity of the barriers, the ease of removing
the barriers, the impact of the barriers on energy efficiency and the economic performance.
The aim here was to see whether the four criteria have the same importance across the five
disciplines. If the criteria have different importance in each discipline, the managers can
create a matrix of comparison weights for each criterion in each discipline. Interviewees
were asked which criterion is most important to them in overcoming the barriers. The
FAHP method (Equations (1)–(4), see Section 3.3.1) was used to determine the importance
of the criteria. Table 2 shows the ranking of the criteria.
As shown in Table 2, the effect of removing barriers to economic performance (0.513)
was the most significant. This criterion was five times more important than the least
important criterion (severity of barriers (0.102)) for the ISY. The second highest priority
for ISY was the effect of removing barriers on energy efficiency (0.267) and the simplicity
of removing barriers (0.118) was the third highest priority for ISY. Finally, the severity of
barriers was the least important priority for the yard (0.102).

4.1.4. MCDM Assessment

Tables 3 and 4 show the results of the ranking of the barriers in each discipline and
the final ranking of the barriers for ISY. The fourth step was the “MCDM assessment”. In
this step, by applying the weight of both disciplines and criteria, the yard DMs were able
to rank the barriers within each discipline and overall. The FTOPSIS methodology was
used to manage the FMAGDM and rank the barriers6 .

Ranking of ISY’ Barriers

Table 3 shows the ranking of the barriers in each discipline for the ISY. After applying
the weight of criteria (see Section 4.1.3) and utilizing the FMAGDM method, the authors
could rank the barriers in each disciplines based on the priorities of DMs in ISY.
Technology and innovation
“Inappropriate technology”, “technical risk” and “immaturity of the technology” were
the main barriers in the “Technology and Innovation” discipline for the yard in question.
Some cost-effective energy-efficient technologies may be unsuitable for certain industries
or regions [76]. This type of barrier is more common for industries and companies that
have a specific or single type of production, or for technologies that have limitations due
to temperature or pollution [38,77]. As an example for the shipyard in question, the new
J. Mar. Sci. Eng. 2021, 9, 1113 16 of 25

modern winches were not suitable for the shipyard due to the extremely high temperature
in the region, which caused them to stop working during operation.
The technical risk (the second main barrier), was directly related to the third main
barrier, which was the “immaturity of the technology”. Investment risk can lead to a
widening of the energy efficiency gap [13,44]. The “immaturity of technology” barrier is
due to the innovative nature of the technologies and measures [24]. A new technology
is subject to future performance and evaluation to confirm its maturity. The company
DMs claimed that some technologies on the market are not proven or do not have the
effectiveness stated in their information, and company managers do not want to invest in
high risk technologies whose effectiveness are subject to further evaluation.
Economics
“Lack of funding”, “limited access to capital costs” and the “cost of identifying op-
portunities” were the main barriers to the yard’s financial discipline. Energy efficient
technologies, especially in shipbuilding, are very expensive and limited access to capital
was the main challenge for ISY [33], which may improve during the economic reces-
sion [111]. According to the respondents, these barriers have increased mainly due to the
international sanctions against Iran. The limited access to capital leads to poor budgeting,
which makes energy efficiency less important for the yard [36].
“Costs of production interruptions” are an important barrier in energy-intensive in-
dustries and are categorized as hidden costs in the classical categorization [44]. The barrier
is placed as the highest priority in the case yard. In addition, the “increase in perceived
costs of energy saving measures” (fifth highest priority barrier) and “cost-effectiveness
analysis and tendering” (sixth highest priority barrier) were categorized as hidden costs.
Hidden costs are categorized as barriers that are not market-based and may hinder energy
efficiency investments [44]. The hidden cost may be due to the cost of collecting and
analyzing information [66], and it is more common in small industries [129].
“Split incentives” were ranked as the sixth highest priority. The common example
of the shipyard was a typical case in industry, where shipyard managers, who have a
short tenure, were reluctant and hesitant to invest in projects, measures and technologies
with a long payback period [59]. The last four barriers were “capital market barriers”,
“lack of investment capacity”, “lack of information on the profitability of energy saving
measures” and “competition from other projects”. The “capital market barriers” and “lack
of investment capacity” were related to problems in accessing capital and were classified
as behavioral barriers [33]. According to the interviewees, this problem is becoming
more serious, especially after the international sanctions against Iran. Finally, the “lack
of information on the profitability of energy saving measures” and “competition from
other projects” are considered as insufficient information. Lack of information can lead to a
widening of the energy efficiency gap [35,46].
Operation
“Disruption” was the main barrier in the discipline. The cost of any disruption is
categorized as a hidden cost [67]. Shipyard managers were concerned that ship production
would be disrupted due to energy efficiency measures. This concern led to an increased
lack of energy efficiency within the shipyard. “Not using information”, “wrong form of in-
formation”, “wrong information” and “not maintaining information” were ranked second,
third, fourth and fifth, respectively. “Form of information”, “incorrect form of information”
and “incorrect information” are crucial to improving energy efficiency. The information
provided to users must be specific, vivid, simple and personalized (SVSP) [13,36]. As an
example, the form of the information provided on winches (see the section on technology
and innovation) at the shipyard was not specific, leading to an increased investment risk.
The relevant staff in the shipyards must be well trained to use information on technology
and maintain it in order to remove the barriers around information [111].
Human element
An “inappropriate organizational structure” was the main barrier in the discipline.
The barrier is categorized under the main theory of organizational barriers [36]. The
J. Mar. Sci. Eng. 2021, 9, 1113 17 of 25

shipyard did not have an independent and classical energy department, and only one junior
engineer was responsible for managing the energy sector within the shipyard. The “low
position of the energy manager” and the “lack of influence of the energy manager” (fourth
highest barrier) in the shipyard made energy efficiency a low priority for managers (sixth
highest priority) [13], and other issues a higher priority for capital investment (seventh
highest priority) [52]. In addition, an “inappropriate organizational structure” (first barrier)
affects the organizational culture. By developing a culture of environmental values, the
yard can promote energy efficiency [82]. Improving the organization’s culture towards
environmental protection can overcome the eleventh most important barrier, namely, the
“lack of management support” for energy efficiency investments.
“Lack of technical skills” and “lack of trained staff” were considered to be the second
and fifth biggest barriers in the shipyard respectively. Managers considered the “lack of
technical skills of staff” as a barrier to energy efficiency [106]. This may increase the energy
gap if it exists among top managers, as they may not realize the importance of investing in
energy efficiency measures. The interviewees highlighted that staff need further training
on energy efficiency meters and the possibilities to improve energy efficiency in the yard.
“Limited rationality” (the third largest barrier) is a type of behavioral barrier [78].
“Limited rationality” can increase the gap in energy efficiency. Instead of making the
optimal decision, DMs, due to a lack of information tend to make decisions based on
the rule of thumb and strive to make a satisfactory decision instead of optimizing their
decision [13]. The DMs in the shipyard highlighted many constraints, such as “lack of
time” (eighth of the main barriers), that cause them to replace the imprecise decisions and
rules of thumb with the optimized decisions [89]. Shipyard managers also highlighted that,
due to a lack of access to the required budget, there are conflicting interests and goals of
individuals and departments within the shipyard. This conflict presents a complicated
situation for the shipyard managers in terms of optimizing their decisions to obtain the
lowest price for the energy efficiency measures, but at the same time achieving the most
efficient effects [90,91].
“Inertia” was the seventh biggest barrier for the shipyard. Interviewees highlighted
that there is resistance among individuals to invest in energy efficiency measures. This is
related to the habits and routines formed by individuals and organizational behavior [85,86]
and it is usually difficult to change behavior in the short term because individuals strive to
prevent changes in the environment [87].
Policy and regulation
ISY meets the requirements of ISO 14,001 (Environmental Management System). How-
ever, there were questions about how it would be implemented and how environmental
issues would be prioritized in relation to business objectives. The “lack of an environmental
policy from the government” was considered by the yard manager as the main barrier.
They considered that there are not sufficient and appropriate policies for environmental
protection, and if there are any, they are weak and not sufficiently enforced (fourth highest
priority) and monitored by the government [111]. They also complained that they do not
receive any support in the form of incentives or subsidies from the government to improve
energy efficiency within their portfolio. Shipyards need more of this kind of support during
the economic recession and international sanctions [129].

4.1.5. Strategy Building

“Strategy formation” is the final step in the framework. Shipbuilding authorities
can identify and rank barriers according to their priorities. This helps them to have a
better understanding of the concept of barriers, allowing them to make rational decisions
on investments in energy efficiency measures [132]. In this section, the weights for the
disciplines calculated in Section 4.1.2 were applied and, using Equations (5)–(10) of the
FMAGDM, the ranking of the barriers was then completed. Table 4 shows the final ranking
of the barriers. Based on the final ranking, ISY DMs can eliminate the bounded rationality
J. Mar. Sci. Eng. 2021, 9, 1113 18 of 25

barrier in the shipyard, and DMs can make rational decisions and also optimize their
decisions in improving the energy efficiency within their portfolio, overcoming the barriers.

Measures to Overcome the Barriers

Disciplines
The ranking of barriers in Table 4 shows that the discipline weights have a significant
impact on prioritizing the barriers. The analysis show that there is a severe imbalance
among the disciplines for prioritizing among the barriers. The top discipline was economic,
and its weight was 18 times that of the least important discipline (human element). The
significant imbalance among the importance of the disciplines might lead to bias in priori-
tizing the barriers and the potential in improving the energy efficiency may be lost. There is
a recommendation that the DMs of the case shipyard normalize their perspectives among
the various disciplines. This way of approach will help them to identify the potential
for improving energy efficiency, even with lower cost [106]. For example, training of the
personnel can increase awareness of (both low and high ranked) personnel. It will affect the
behavioral culture of the organization and can eliminate barriers, such as the low position
of the energy manager, the lack of influence of the energy manager, and a low priority of
energy efficiency measures for managers [13].
Technology
Inappropriate technology, technical risks and the immaturity of the technology were
placed as the second, sixth and eleventh main barriers for the yard, respectively. Shipyard
managers should be well educated and aware of the concepts of technology, and they
should be able to access sufficient information in order to make rational decisions. The
impacts of inappropriate and immature technologies can be mitigated through an appropri-
ate feasibility study and by examining the current state of the art and following the good
examples of other yards.
The use of immature technologies increases the technical risk [26] and depends on
the risk appetite of the yard management. There are managers who want to be pioneers
and will accept the risk. However, the technical risk can be reduced if the manufacturer
takes into account the smart contract. Cooperation with technology manufacturers and
acceptance to introducing the technology in the yard can be a win-win deal [38]. The
manufacturer can evaluate the product and the yard can benefit from any energy savings
without having to pay the huge capital cost and with a lower risk.
Economic
The lack of funding was the first priority for the yard. The interviewees claim that
this has become serious after the implementation of international sanctions. Although
removing the international sanctions is at the government level, by introducing appropriate
budgeting, DMs can solve the cost of identifying opportunities. Applying appropriate
budgeting and considering a bottom-up approach to budget allocation in the organization
can help to overcome the limited availability of capital for those departments that need
more budget to improve energy efficiency [111]. There are extensive opportunities for im-
proving energy efficiency by implementing low-cost products, such as system optimization
measures to achieve an average energy efficiency of 20–30% [38].
Operation
Interviewees highlighted that disruption due to the implementation of energy effi-
ciency measures is the biggest problem in the discipline, but it was ranked 14th in the final
ranking. Appropriate planning, well trained and skilled staff (especially senior managers),
and use of the right (credibility and trust) form of information (specific, vivid, simple and
personalized (SVSP)) [13] can prevent disruption of the yard’s operations and prevent the
wasting of time and costs. In order to overcome the barrier, it is important that an appropri-
ate feasibility study and an up-to-date technical survey are conducted with respect to the
shipyard’s current and future projects. This is also linked to the appropriate organization
of the yard [36]. By predicting the appropriate time for the installation of new technologies
and considering related tasks, disruptions to shipyard operations can be minimized and
J. Mar. Sci. Eng. 2021, 9, 1113 19 of 25

resolved. In addition, if there is a well-trained and strong energy manager (team) with
sufficient power and independence, the process can be accelerated and the risks can be
minimized.
Human element
The human element was the least important discipline among the others. Since the
importance of the different disciplines has a significant impact on the final ranking of the
barriers, the human factor barriers were ranked 22–33. By overcoming barriers, such as
inappropriate organizational structure (23rd biggest barrier), other barriers from other
disciplines, such as disruption (14th biggest barrier), can also be overcome. To resolve
the barrier of organizational structure, the value of the organization should be changed to
environmental protection [44]. To achieve this goal, individuals—especially top managers—
must have the ambition to increase energy efficiency within the shipyard [36]. This can be
accelerated by increasing staff awareness through training [106] and by conducting various
workshops. It is suggested that the yard’s top managers pay more attention to human
discipline and the related barriers. In addition, the problem of time constraints can be
overcome by recruiting well-trained and skilled staff [111].
Policy and regulations
Energy efficiency plays an important role in promoting the energy security of states
and in tackling climate change [2]. It is a top priority in the policy decisions of nations
responding to the Paris Agreement [3]. As a major stakeholder, the government can jus-
tify energy efficiency measures in industry. In the case of the shipyard, the international
sanctions against Iran present a major problem for the shipyard in accessing capital. The
sanctions affect the ability of Iranian industries to invest in energy efficiency measures.
Due to the lack of access to capital, the yard’s priorities for energy efficiency measures
have changed and they have been placed at a lower level of priorities. However, by imple-
menting projects, such as Voluntary Agreements (VAs) and Long Term Agreements (LTAs),
the government—like some other countries, such as the Netherlands and Japan—can help
industries to improve energy efficiency in their context [97]. Training of energy managers
in the shipbuilding industry, raising awareness, funding research and development (R&D)
and establishing codes and protocols are other measures that can help the government to
improve energy efficiency in the shipbuilding industry.

5. Conclusions
It is important to consider a holistic, systematic and transdisciplinary approach to
overcoming barriers to improving energy efficiency. In this study a framework for over-
coming the barriers to energy efficiency within shipyards is proposed. The framework
is holistic, systematic and transdisciplinary, and takes into account the interrelationship
and interaction between different barriers. It consists of five main disciplines, namely,
the human element, technology and innovation, operation, policy and regulations, and
economics. The focused group of the study includes the top managers of shipyards, who
have the highest weight in the final decision for investment in energy efficiency measures.
Implementing the framework overcomes the bounded rationality barrier and helps DMs to
make rational and optimal decisions in a vague environment regarding improving energy
efficiency. Additionally, following the PDCA cycle and considering the changeable DMs
priorities contributes a dynamic feature to the framework. Due to the framework’s generic
characteristics, it may be applied to other industries, and policy makers and researchers in
those industries can also employ the framework to overcome barriers to improving energy
efficiency.
The following steps was taken to design and develop the framework:
• Systematic literature review on barriers of energy efficiency from various sectors and
industries;
• Analysis and categorizing different barriers in the shipbuilding industry;
• Design and develop the framework by categorizing the identified barriers in to five
main disciplines;
J. Mar. Sci. Eng. 2021, 9, 1113 20 of 25

• Using two MCDM methods (FAHP and FMAGDM) to understand the interaction
between the barriers.
To validate the proposed framework, a case study was conducted for an Iranian
shipyard. Interviews were conducted with five senior managers of the shipyard, and their
priorities of the categorized barriers in each discipline were determined. Four criteria
were presented to them for prioritizing the barriers, namely, how severe the barriers are,
how easy it is to remove the barriers, how much impact the removal of the barriers can
have on energy efficiency, and how much impact the removal of the barriers can have on
economic performance. By applying FAHP and FMAGDM, the priorities of the barriers
were identified in each discipline and in general aspects. The results show that:
• There is a significant imbalance between the importance of the different disciplines.
Economic barriers are the most important and are considered to be 18 times more
important than the least important discipline, i.e., barriers due to the human factor.
• In the criteria side, the most important thing for DMs is the impact of “barriers removal
to economic performance”. “Improvement of energy efficiency after barriers removal”,
“simplicity of barriers removal”, and “severity of barriers” are ranked in order of
importance”.
• The weights assigned to the disciplines have significant impact on the final ranking of
the barriers.
• Due to international sanctions, the yard has problems with financing and limited
access to capital. In addition, costs caused by disruptions of production is another
major barrier to implementing energy efficiency measures at the yard.
• In terms of technological discipline, the main barriers for the yard are “lack of appro-
priate technology at the yard”, “technical investment risk” and “immaturity of the
technology”.
• “Lack of environmental policy from the government”, “lack of government incentives”
and “lack of environmental policy from the company” are considered to be the highest
priority for the yard in terms of policy and regulations.
• The least important barriers are related to the human factor. The main barriers in this
discipline are “inappropriate organizational structure”, “lack of technical competence”,
and “bounded rationality”.
In order to overcome the barriers, the integration of the barriers must be taken into
account. Imbalances between the importance of the disciplines can lead to a loss of
potential for improving energy efficiency. Appropriate budgeting and consideration of
the bottom-up approach can help the yard to overcome the lack of budgeting and ensure
that different departments have access to the necessary budget for investments in energy
efficiency measures. There is significant potential for improving energy efficiency (20–30%)
by implementing (low hanging fruit) measures that are cheap and affordable for the yard.
In addition, addressing the organizational structure of the yard can play a major role in
overcoming other barriers. It is important to shift the value of the yard to environmental
protection. The role of top management and its commitment in this respect is extremely
important. This approach can be accelerated by raising staff awareness through training
and conducting various workshops. Governments as key stakeholders can motivate
energy efficiency measures and regulations in the shipbuilding industry. By providing
incentives, subsidies and implementing projects, such as voluntary agreements and long-
term contracts, training shipbuilding energy managers, raising awareness, funding research
and development and establishing codes and protocols, the shipbuilding industry can
overcome the barriers to improving energy efficiency.

Author Contributions: Conceptualization: S.V.V. and A.I.Ö.; methodology: S.V.V., A.I.Ö. and A.S.;
validation: S.V.V., A.I.Ö. and A.S.; formal analysis: S.V.V.; investigation: S.V.V.; resources: S.V.V.;
data curation: S.V.V.; writing—original draft preparation: S.V.V.; writing—review and editing: S.V.V.,
A.I.Ö. and A.S.; visualization: S.V.V.; supervision: A.I.Ö. and A.S. All authors have read and agreed
to the published version of the manuscript.
J. Mar. Sci. Eng. 2021, 9, 1113 21 of 25

Funding: This research received no external funding.

Informed Consent Statement: Not applicable.
Acknowledgments: The authors would like to thank the reviewers and the journal’s editor for their
valuable comments, which have greatly improved the study.
Conflicts of Interest: The authors declare no conflict of interest.

Notes
1 SEEMP is mandatory for all vessels of 400 GT or more and must be carried on board. However, it does not contain any explicit
and mandatory content and implementation requirements. On the other hand, the ESEEMP is an improved SEEMP including
mandatory content, such as an implementation plan on how to achieve the Carbon Intensity Indicator (CII) targets, and should
be subject to approval.
2 Depending on the DMs’ preference, different criteria may be considered.
3 Depending on the type of data (explicit or linguistic) and the number of DMs, exceptional MCDM techniques can be used. The
number of respondents may also vary depending on many factors, such as ownership and the organisational form of the yard.
There are usually no more than three decision makers when it comes to making strategic choices regarding the strengths of a
yard.
4 In an unclear environment of lack of information and inconsistency, managers generally fail to make rational decisions and set
their priorities. MCDM methods can help managers to identify problems and make the best decision among different options in
an ambiguous environment.
5 The Fuzzy TOPSIS (FTOPSIS) method was used to handle FMAGDM.
6 Equations (5)–(10) were used to rank the alternatives, please see Section 3.3.2.

References
1. International Energy Agency (IEA). 2020. Available online: https://www.iea.org/reports/tracking-industry-2020 (accessed on
17 March 2021).
2. Hess, D.J.; Renner, M. Conservative political parties and energy transitions in Europe: Opposition to climate mitigation policies.
Renew. Sustain. Energy Rev. 2019, 104, 419–428. [CrossRef]
3. Sun, H.; Ikram, M.; Mohsin, M.; Abbas, Q. Energy security and environmental efficiency: Evidence from oecd countries. Singap.
Econ. Rev. 2021, 66, 489–506. [CrossRef]
4. Rammer, C.; Gottschalk, S.; Peneder, M.; Wörter, M.; Stucki, T.; Arvanitis, S. Does energy policy hurt international competitiveness
of firms? A comparative study for Germany, Switzerland and Austria. Energy Policy 2017, 109, 154–180. [CrossRef]
5. Chatzistamoulou, N.; Kounetas, K.; Tsekouras, K. Energy efficiency, productive performance and heterogeneous competitiveness
regimes. Does the dichotomy matter? Energy Econ. 2019, 81, 687–697. [CrossRef]
6. International Energy Agency (IEA). 2019. Available online: https://www.iea.org/reports/world-energy-outlook-2019 (accessed
on 9 May 2021).
7. United Nation Conference on Trade and Development (UNCTAD), 2017. Review of Maritime Transport. 2017. Available online:
https://unctad.org/en/pages/publicationwebflyer.aspx?publicationid=1890 (accessed on 14 February 2021).
8. International Maritime Organization (IMO). MEPC\75\MEPC 75-7-15. REDUCTION of GHG EMISSIONS FROM SHIPS. Fourth
IMO GHG Study 2020- Final Report. 2020. Available online: http://www.imo.org/en/About/Pages/Default.aspx (accessed on 7
January 2021).
9. Chatzinikolaou, S.D.; Ventikos, N.P. Applications of Life Cycle Assessment in Shipping; INT-NAM 2014At: Istanbul, Turkey, 2014.
10. Merk, O. Shipping Emissions in Ports; OECD: Paris, France, 2014.
11. Nordtveit, E. Life Cycle Assessment of a Battery Passenger Ferry. Master’s Thesis, University of Agder, Kristiansand and
Grimstad, Norway, 2017. Available online: http://hdl.handle.net/11250/2493457 (accessed on 12 May 2021).
12. Sorrell, S.; Schleich, J.; Scott, S.; O’Malley, E.; Trace, F.; Boede, U.; Radgen, P. Reducing Barriers to Energy Efficiency in Public and
Private Organizations; Science and Policy Technology Research (SPRU); University of Sussex: Sussex, UK, 2000.
13. Gillingham, K.; Newell, R.; Palmer, K. Energy Efficiency Economics and Policy. Annu. Rev. Resour. Econ. 2009, 1, 597–620.
[CrossRef]
14. Gillingham, K.; Palmer, K. Bridging the Energy Efficiency Gap: Policy Insights from Economic Theory and Empirical Evidence.
Rev. Environ. Econ. Policy 2014, 8, 18–38. [CrossRef]
15. Rehmatulla, N.; Calleya, J.; Smith, T. The implementation of technical energy efficiency and CO2 emission reduction measures in
shipping. Ocean Eng. 2017, 139, 184–197. [CrossRef]
16. Dewan, M.H.; Yaakob, O.; Suzana, A. Barriers for adoption of energy efficiency operational measures in shipping industry. WMU
J. Marit. Aff. 2018, 17, 169–193. [CrossRef]
17. Acciaro, M.; Hoffmann, P.N.; Eide, M.S. The energy efficiency gap in maritime transport. J. Shipp. Ocean. Eng. 2013, 3, 1.
J. Mar. Sci. Eng. 2021, 9, 1113 22 of 25

18. Killip, G.; Fawcett, T.; Cooremans, C.; Wijns-Craus, W.; Subramani, K.; Voswinkel, F. Multiple Benefits of Energy Efficiency
at the Firm Level: A Literature Review; European Council for an Energy Efficient Economy: 2019. Available online: https:
//ora.ox.ac.uk/objects/uuid:b4a51feb-071e-474e-a284-e95285145f29 (accessed on 10 May 2021).
19. Pastor, A.; Vieira, D.; Soudijn, F.; Edelenbosch, O. How uncertainties are tackled in multi-disciplinary science? A review of
integrated assessments under global change. Catena 2020, 186, 104305. [CrossRef]
20. van Smirren, J. Influences and impact of the energy industry on the new blue economy and its workforce development. In
Preparing a Workforce for the New Blue Economy; Elsevier: Amsterdam, The Netherlands, 2021; pp. 201–209.
21. Gemak Shipyard. OFFSHORE & OIL GAS PROJECTS. 2021. Available online: https://www.gemak.com/BusinessActivities.
aspx?q=12 (accessed on 17 March 2021).
22. Thollander, P.; Danestig, M.; Rohdin, P. Energy policies for increased industrial energy efficiency: Evaluation of a local energy
programme for manufacturing SMEs. Energy Policy 2007, 35, 5774–5783. [CrossRef]
23. Voinov, A.; Filatova, T. Pricing strategies in inelastic energy markets: Can we use less if we can’t extract more? Front. Earth Sci.
2014, 8, 3–17. [CrossRef]
24. Jaffe, A.B.; Stavins, R.N. The energy-efficiency gap What does it mean? Energy Policy 1994, 22, 804–810. [CrossRef]
25. Brent, D.A.; Ward, M.B. Energy efficiency and financial literacy. J. Environ. Econ. Manag. 2018, 90, 181–216. [CrossRef]
26. Rohdin, P.; Thollander, P.; Solding, P. Barriers to and drivers for energy efficiency in the Swedish foundry industry. Energy Policy
2007, 35, 672–677. [CrossRef]
27. Allcott, H.; Greenstone, M. Is There an Energy Efficiency Gap? J. Econ. Perspect. 2012, 26, 3–28. [CrossRef]
28. Camarasa, C.; Nägeli, C.; Ostermeyer, Y.; Klippel, M.; Botzler, S. Diffusion of energy efficiency technologies in European residential
buildings: A bibliometric analysis. Energy Build. 2019, 202, 109339. [CrossRef]
29. Gissey, G.C.; Dodds, P.; Radcliffe, J. Market and regulatory barriers to electrical energy storage innovation. Renew. Sustain. Energy
Rev. 2018, 82, 781–790. [CrossRef]
30. Thollander, P.; Ottosson, M. An energy efficient Swedish pulp and paper industry—Exploring barriers to and driving forces for
cost-effective energy efficiency investments. Energy Effic. 2008, 1, 21–34. [CrossRef]
31. Pavlyk, V. Assessment of green investment impact on the energy efficiency gap of the national economy. Financ. Mark. Inst. Risks
2020, 4, 117–123. [CrossRef]
32. Blumstein, C.; Krieg, B.; Schipper, L.; York, C. Overcoming social and institutional barriers to energy conservation. Energy 1980, 5,
355–371. [CrossRef]
33. Hirst, E.; Brown, M. Closing the efficiency gap: Barriers to the efficient use of energy. Resour. Conserv. Recycl. 1990, 3, 267–281.
[CrossRef]
34. Howarth, R.; Andersson, B. Market barriers to energy efficiency. Energy Econ. 1993, 15, 262–272. [CrossRef]
35. Eyre, N. Barriers to Energy Efficiency: More Than Just Market Failure. Energy Environ. 1997, 8, 25–43. [CrossRef]
36. Weber, L. Some reflections on barriers to the efficient use of energy. Energy Policy 1997, 25, 833–835. [CrossRef]
37. Fleiter, T.; Hirzel, S.; Worrell, E. The characteristics of energy-efficiency measures–a neglected dimension. Energy Policy 2012, 51,
502–513. [CrossRef]
38. Thollander, P.; Palm, J. Improving Energy Efficiency in Industrial Energy Systems: An Interdisciplinary Perspective on Barriers, Energy
Audits, Energy Management, Policies, and Programs; Springer Science & Business Media: Berlin/Heidelberg, Germany, 2012.
39. TYap, J.; Escresa-Guillermo, L.; Paras, Y.G.M. Access to Sustainable Energy in the Philippines: Market Failure and Political
Economy Considerations (April 2020). ASOG WORKING PAPER 20-006. Available online: https://ssrn.com/abstract=3578493
(accessed on 12 August 2021). [CrossRef]
40. Munoz, L. Is Environmental Externality Management a Correction of Adam Smith’s Model to Make it Environmentally Friendly
and Shift it Towards Green Markets or is it a Distortion on Top of Another Distortion? Int. J. Econ. Bus. Manag. Stud. 2017, 4, 1–16.
[CrossRef]
41. Gerarden, T.D.; Newell, R.; Stavins, R.N. Assessing the Energy-Efficiency Gap. J. Econ. Lit. 2017, 55, 1486–1525. [CrossRef]
42. Apeaning, R.W.; Thollander, P. Barriers to and driving forces for industrial energy efficiency improvements in African industries—
A case study of Ghana’s largest industrial area. J. Clean. Prod. 2013, 53, 204–213. [CrossRef]
43. Fawcett, T.; Hampton, S. Why & how energy efficiency policy should address SMEs. Energy Policy 2020, 140, 111337. [CrossRef]
44. Sorrell, S.; O’Malley, E. The Economics of Energy Efficiency; Edward Elgar Publishing: Cheltenham, UK, 2004; p. 2607.
45. Backlund, S.; Thollander, P.; Palm, J.; Ottosson, M. Extending the energy efficiency gap. Energy Policy 2012, 51, 392–396. [CrossRef]
46. Chai, K.-H.; Yeo, C. Overcoming energy efficiency barriers through systems approach—A conceptual framework. Energy Policy
2012, 46, 460–472. [CrossRef]
47. Cagno, E.; Worrell, E.; Trianni, A.; Pugliese, G. A novel approach for barriers to industrial energy efficiency. Renew. Sustain.
Energy Rev. 2013, 19, 290–308. [CrossRef]
48. Dunlop, T. Mind the gap: A social sciences review of energy efficiency. Energy Res. Soc. Sci. 2019, 56, 101216. [CrossRef]
49. Coccia, M. The theory of technological parasitism for the measurement of the evolution of technology and technological
forecasting. Technol. Forecast. Soc. Chang. 2019, 141, 289–304. [CrossRef]
50. Hötte, K. How to accelerate green technology diffusion? Directed technological change in the presence of coevolving absorptive
capacity. Energy Econ. 2020, 85, 104565. [CrossRef]
J. Mar. Sci. Eng. 2021, 9, 1113 23 of 25

51. Batlle, E.A.O.; Palacio, J.C.E.; Lora, E.E.S.; Reyes, A.M.M.; Moreno, M.M.; Morejón, M.B. A methodology to estimate baseline
energy use and quantify savings in electrical energy consumption in higher education institution buildings: Case study, Federal
University of Itajubá (UNIFEI). J. Clean. Prod. 2020, 244, 118551. [CrossRef]
52. Soepardi, A.; Thollander, P. Analysis of relationships among organizational barriers to energy efficiency improvement: A case
study in Indonesia’s steel industry. Sustainability 2018, 10, 216. [CrossRef]
53. Dean, E.; Elardo, J.; Green, M.; Wilson, B.; Berger, S. The Problem of Imperfect Information and Asymmetric Information. In
Principles of Economics: Scarcity and Social Provisioning, 2nd ed.; BCcampus Open Publishing: Victoria, BC, Canada, 2020.
54. Myers, E. Asymmetric information in residential rental markets: Implications for the energy efficiency gap. J. Public Econ. 2020,
190, 104251. [CrossRef]
55. Adetutu, M.O.; Stathopoulou, E. Information Asymmetry in Voluntary Environmental Agreements (VEAs): Theory and Evidence
from UK Climate Change Agreements (CCAs). Oxf. Econ. Pap. 2021, 73, 644–670. [CrossRef]
56. Wrigley, K.; Crawford, R.H. Identifying policy solutions for improving the energy efficiency of rental properties. Energy Policy
2017, 108, 369–378. [CrossRef]
57. Kangas, H.-L.; Lazarevic, D.; Kivimaa, P. Technical skills, disinterest and non-functional regulation: Barriers to building energy
efficiency in Finland viewed by energy service companies. Energy Policy 2018, 114, 63–76. [CrossRef]
58. Reina, V.J.; Kontokosta, C. Low hanging fruit? Regulations and energy efficiency in subsidized multifamily housing. Energy
Policy 2017, 106, 505–513. [CrossRef]
59. Shaikh, I.; O’Connor, G.C. Understanding the motivations of technology managers in radical innovation decisions in the mature
R&D firm context: An Agency theory perspective. J. Eng. Technol. Manag. 2020, 55, 101553. [CrossRef]
60. Qiu, Y.; Colson, G.; Wetzstein, M.E. Risk preference and adverse selection for participation in time-of-use electricity pricing
programs. Resour. Energy Econ. 2017, 47, 126–142. [CrossRef]
61. Palm, J.; Reindl, K. Understanding barriers to energy-efficiency renovations of multifamily dwellings. Energy Effic. 2017, 11,
53–65. [CrossRef]
62. Kahn, B.E. Using Visual Design to Improve Customer Perceptions of Online Assortments 1. Routledge Companion Consum. Behav.
2017, 93, 304–327. [CrossRef]
63. Rehmatulla, N.; Smith, T. Barriers to energy efficiency in shipping: A triangulated approach to investigate the principal agent
problem. Energy Policy 2015, 84, 44–57. [CrossRef]
64. Graus, W.; Worrell, E. The principal–agent problem and transport energy use: Case study of company lease cars in the Netherlands.
Energy Policy 2008, 36, 3745–3753. [CrossRef]
65. Menanteau, P.; Lefebvre, H. Competing technologies and the diffusion of innovations: The emergence of energy-efficient lamps
in the residential sector. Res. Policy 2000, 29, 375–389. [CrossRef]
66. Cardoso, C.A.; Torriti, J.; Lorincz, M. Making demand side response happen: A review of barriers in commercial and public
organisations. Energy Res. Soc. Sci. 2020, 64, 101443. [CrossRef]
67. Good, N.; Ellis, K.A.; Mancarella, P. Review and classification of barriers and enablers of demand response in the smart grid.
Renew. Sustain. Energy Rev. 2017, 72, 57–72. [CrossRef]
68. Golove, W.; Eto, J. Market Barriers to Energy Efficiency: A Critical Reappraisal of the Rationale for Public Policies to Promote
Energy Efficiency. In Market Barriers to Energy Efficiency: A Critical Reappraisal of the Rationale for Public Policies to Promote Energy
Efficiency, OSTI.GO, U.S.; 1996. Available online: https://www.osti.gov/biblio/270751-RXrf5G/webviewable/ (accessed on 16
May 2021).
69. Trianni, A.; Cagno, E.; Thollander, P.; Backlund, S. Barriers to industrial energy efficiency in foundries: A European comparison. J.
Clean. Prod. 2013, 40, 161–176. [CrossRef]
70. Trianni, A.; Cagno, E. Dealing with barriers to energy efficiency and SMEs: Some empirical evidences. Energy 2012, 37, 494–504.
[CrossRef]
71. O’Keeffe, J.M.; Gilmour, D.; Simpson, E. A network approach to overcoming barriers to market engagement for SMEs in energy
efficiency initiatives such as the Green Deal. Energy Policy 2016, 97, 582–590. [CrossRef]
72. Sorrell, S.; Mallett, A.; Nye, S. Barriers to Industrial Energy Efficiency: A Literature Review, Background Study for the UNIDO Industrial
Development Report (IDR)‘Industrial Energy Efficiency Pays, Why Is It Not Happening?’; SPRU, University of Sussex: Brighton, UK,
2010.
73. Hilorme, T.; Zamazii, O.; Judina, O.; Korolenko, R.; Melnikova, Y. Formation of risk mitigating strategies for the implementation
of projects of energy saving technologies. Acad. Strateg. Manag. J. 2019, 18, 1–6.
74. Sutherland, R.J. Market Barriers to Energy-Efficiency Investments. Energy J. 1991, 12, 15–34. [CrossRef]
75. Leskinen, N.; Vimpari, J.; Junnila, S. Using real estate market fundamentals to determine the correct discount rate for decentralised
energy investments. Sustain. Cities Soc. 2020, 53, 101953. [CrossRef]
76. Sun, J.; Wang, Z.; Li, G. Measuring emission-reduction and energy-conservation efficiency of Chinese cities considering manage-
ment and technology heterogeneity. J. Clean. Prod. 2018, 175, 561–571. [CrossRef]
77. Akram, R.; Chen, F.; Khalid, F.; Ye, Z.; Majeed, M.T. Heterogeneous effects of energy efficiency and renewable energy on carbon
emissions: Evidence from developing countries. J. Clean. Prod. 2020, 247, 119122. [CrossRef]
78. Nehler, T.; Parra, R.; Thollander, P. Implementation of energy efficiency measures in compressed air systems: Barriers, drivers
and non-energy benefits. Energy Effic. 2018, 11, 1281–1302. [CrossRef]
J. Mar. Sci. Eng. 2021, 9, 1113 24 of 25

79. Sussman, R.; Chikumbo, M.; Miller, N. After the Audit: Improving Residential Energy Efficiency Assessment Reports; American Council
for an Energy-Efficient Economy: Washington, DC, USA, 2019.
80. Fresner, J.; Morea, F.; Krenn, C.; Uson, J.A.; Tomasi, F. Energy efficiency in small and medium enterprises: Lessons learned from
280 energy audits across Europe. J. Clean. Prod. 2017, 142, 1650–1660. [CrossRef]
81. Raymond, C.M.; Kenter, J.O.; van Riper, C.J.; Rawluk, A.; Kendal, D. Editorial overview: Theoretical traditions in social values for
sustainability. Sustain. Sci. 2019, 14, 1173–1185. [CrossRef]
82. Lopes, J.R.; de Araújo Kalid, R.; Rodríguez, J.L.; Ávila Filho, S. A new model for assessing industrial worker behavior regarding
energy saving considering the theory of planned behavior, norm activation model and human reliability. Resour. Conserv. Recycl.
2019, 145, 268–278. [CrossRef]
83. Hewett, M.J. Achieving energy efficiency in a restructured electric utility industry. In Report Prepared for Minnesotans for an Energy
Efficient Economy; Centre for Energy & Environment: Minneapolis, MN, USA, 1998.
84. Li, F.G.; Strachan, N. Modelling energy transitions for climate targets under landscape and actor inertia. Environ. Innov. Soc.
Transit. 2017, 24, 106–129. [CrossRef]
85. Verplanken, B.; Wood, W. Interventions to Break and Create Consumer Habits. J. Public Policy Mark. 2006, 25, 90–103. [CrossRef]
86. Sardianos, C.; Varlamis, I.; Chronis, C.; Dimitrakopoulos, G.; Alsalemi, A.; Himeur, Y.; Bensaali, F.; Amira, A. Reshaping
consumption habits by exploiting energy-related micro-moment recommendations: A case study. In Smart Cities, Green Technologies
and Intelligent Transport Systems; Springer: Cham, Switzerland, 2019; pp. 65–84.
87. Bindl, U.K.; Unsworth, K.L.; Gibson, C.B.; Stride, C.B. Job crafting revisited: Implications of an extended framework for active
changes at work. J. Appl. Psychol. 2019, 104, 605–628. [CrossRef]
88. Nguyen, N.; Greenland, S.; Lobo, A.; Nguyen, H.V. Demographics of sustainable technology consumption in an emerging market:
The significance of education to energy efficient appliance adoption. Soc. Responsib. J. 2019, 15, 803–818. [CrossRef]
89. Koops, L.G. Optimized Maintenance Decision-Making—A Simulation-Supported Prescriptive Analytics Approach Based on
Probabilistic Cost-Benefit Analysis. PHM Soc. Eur. Conf. 2020, 5, 14.
90. Sanstad, A.H.; Howarth, R. ‘Normal’ markets, market imperfections and energy efficiency. Energy Policy 1994, 22, 811–818.
[CrossRef]
91. Bonilla-Campos, I.; Nieto, N.; del Portillo-Valdes, L.; Manzanedo, J.; Gaztañaga, H. Energy efficiency optimisation in industrial
processes: Integral decision support tool. Energy 2020, 191, 116480. [CrossRef]
92. Goren, P.; Motta, M.; Smith, B. The Ideational Foundations of Symbolic Ideology. Politi. Psychol. 2020, 41, 75–94. [CrossRef]
93. Chang, W.J.; Liao, S.H.; Wu, T.T. Relationships among organizational culture, knowledge sharing, and innovation capability: A
case of the automobile industry in Taiwan. Knowl. Manag. Res. Pract. 2017, 15, 471–490. [CrossRef]
94. Westervelt, E.T.; Loechl, P.M.; Clark, S.A.; DuPont, C.E. Enhancing Army Energy Culture with Behavioral Approaches; Engineer
Research and Development Center (US) Champaign United States: Champaign, IL, USA, 2020.
95. Ruth, M.; Goessling-Reisemann, S. Handbook on Resilience of Socio-Technical Systems. In Handbook on Resilience of Socio-Technical
Systems; Edward Elgar Publishing: Cheltenham, UK, 2019.
96. Hong, J.; Gu, J.; Liang, X.; Liu, G.; Shen, G.Q. Characterizing embodied energy accounting with a multi-dimensional framework:
A study of China’s building sector. J. Clean. Prod. 2019, 215, 154–164. [CrossRef]
97. European Commission (EU). Communication from the Commission. Action Plan for Energy Efficiency: Realizing the Potential. 2006,
p. 545. Available online: https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=celex%3A52006DC0545 (accessed on 13 June
2021).
98. Kondo, R.; Kinoshita, Y.; Yamada, T. Green Procurement Decisions with Carbon Leakage by Global Suppliers and Order Quantities
under Different Carbon Tax. Sustainability 2019, 11, 3710. [CrossRef]
99. Coady, D.; Parry, I.W.H.; Shang, B. Energy Price Reform: Lessons for Policymakers. Rev. Environ. Econ. Policy 2018, 12, 197–219.
[CrossRef]
100. Filippidou, F.; Nieboer, N.; Visscher, H. Are we moving fast enough? The energy renovation rate of the Dutch non-profit housing
using the national energy labelling database. Energy Policy 2017, 109, 488–498. [CrossRef]
101. Bertoldi, P. European Union Energy Efficiency Policies for Buildings’. In Handbook of Energy Efficiency in Buildings’; Elsevier:
Amsterdam, The Netherlands, 2018.
102. Palm, J.; Thollander, P. Reframing energy efficiency in industry: A discussion of definitions, rationales, and management practices.
Energy Behav. 2019, 1, 153–175. [CrossRef]
103. Jekabsone, A.; Kamenders, A.; Rosa, M. Implementation of Certified Energy ManagementSystem in Municipality. Case Study.
Environ. Clim. Technol. 2020, 24, 41–56. [CrossRef]
104. Pregger, T.; Naegler, T.; Weimer-Jehle, W.; Prehofer, S.; Hauser, W. Moving towards socio-technical scenarios of the German
energy transition—lessons learned from integrated energy scenario building. Clim. Chang. 2020, 162, 1743–1762. [CrossRef]
105. Bingham, N.; Manohar, R. A Systematic Approach for Arbitration Expressions. IEEE Trans. Circuits Syst. I Regul. Pap. 2020, 67,
4960–4969. [CrossRef]
106. Thollander, P.; Karlsson, M.; Rohdin, P.; Johan, W.; Rosenqvist, J. Introduction to Industrial Energy Efficiency: Energy Auditing, Energy
Management, and Policy Issues; Academic Press: Cambridge, MA, USA, 2020.
107. Fernando, Y.; Hor, W.L. Impacts of energy management practices on energy efficiency and carbon emissions reduction: A survey
of malaysian manufacturing firms. Resour. Conserv. Recycl. 2017, 126, 62–73. [CrossRef]
J. Mar. Sci. Eng. 2021, 9, 1113 25 of 25

108. Trianni, A.; Cagno, E.; Marchesani, F.; Spallina, G. Classification of drivers for industrial energy efficiency and their effect on the
barriers affecting the investment decision-making process. Energy Effic. 2017, 10, 199–215. [CrossRef]
109. Nagesha, N.; Balachandra, P. Barriers to energy efficiency in small industry clusters: Multi-criteria-based prioritization using the
analytic hierarchy process. Energy 2006, 31, 1969–1983. [CrossRef]
110. Wang, G.; Wang, Y.; Zhao, T. Analysis of interactions among the barriers to energy saving in China. Energy Policy 2008, 36,
1879–1889. [CrossRef]
111. Jafarzadeh, S.; Utne, I.B. A framework to bridge the energy efficiency gap in shipping. Energy 2014, 69, 603–612. [CrossRef]
112. Hasanbeigi, A.; Menke, C.; Du Pont, P. Barriers to energy efficiency improvement and decision-making behavior in Thai industry.
Energy Effic. 2010, 3, 33–52. [CrossRef]
113. Prashar, A. Adopting PDCA (Plan-Do-Check-Act) cycle for energy optimization in energy-intensive SMEs. J. Clean. Prod. 2017,
145, 277–293. [CrossRef]
114. Adler, P. When knowledge is the critical resource, knowledge management is the critical task. IEEE Trans. Eng. Manag. 1989, 36,
87–94. [CrossRef]
115. Sovacool, B.K.; Axsen, J.; Sorrell, S. Promoting novelty, rigor, and style in energy social science: Towards codes of practice for
appropriate methods and research design. Energy Res. Soc. Sci. 2018, 45, 12–42. [CrossRef]
116. Ashouri, M.; Fung, B.C.; Haghighat, F.; Yoshino, H. Systematic approach to provide building occupants with feedback to reduce
energy consumption. Energy 2020, 194, 116813. [CrossRef]
117. Williams, B.; Hummelbrunner, R. Soft systems methodology. In Systems Concepts in Action; Stanford University Press: Redwood
City, CA, USA, 2020; pp. 241–261.
118. Mingers, J.; White, L. A review of the recent contribution of systems thinking to operational research and management science.
Eur. J. Oper. Res. 2010, 207, 1147–1161. [CrossRef]
119. Rossi, P.H.; Lipsey, M.W.; Henry, G.T. Evaluation: A Systematic Approach; Sage Publications: Thousand Oaks, CA, USA, 2018.
120. Klein, J.T. Evaluation of Interdisciplinary and Transdisciplinary Research: A Literature Review. Am. J. Prev. Med. 2008, 35,
S116–S123. [CrossRef]
121. Park, J.-Y.; Son, J.-B. Transitioning toward Transdisciplinary Learning in a Multidisciplinary Environment. Int. J. Pedagog. Learn.
2010, 6, 82–93. [CrossRef]
122. Batra, S. Interdisciplinary Research in Social Sciences: Approaches and Rationale. J. Res. 2017, 8, 37. [CrossRef]
123. Klein, J.T. A taxonomy of interdisciplinarity. In The Oxford Handbook of Interdisciplinarity; Oxford University: Oxford, UK, 2010;
Volume 15, pp. 15–30.
124. Penprase, B.E. Interdisciplinary science. In STEM Education for the 21st Century; Springer: Cham, Switzerland, 2020; pp. 93–119.
125. Schary, D.P. Breaking Silos: Interdisciplinary Research in Kinesiology. 2014. Available online: https://ir.library.oregonstate.edu/
concern/graduate_thesis_or_dissertations/jq085n95f (accessed on 23 July 2021).
126. Szostak, R.; Gnoli, C.; López-Huertas, M. Interdisciplinary Knowledge Organization; Gabler, JASIST; Springer: Berlin/Heidelberg,
Germany, 2016. [CrossRef]
127. A Brown, M. Market failures and barriers as a basis for clean energy policies. Energy Policy 2001, 29, 1197–1207. [CrossRef]
128. Mulder, P.; de Groot, H.L.; Hofkes, M.W. Explaining slow diffusion of energy-saving technologies; a vintage model with returns
to diversity and learning-by-using. Resour. Energy Econ. 2003, 25, 105–126. [CrossRef]
129. Rohdin, P.; Thollander, P. Barriers to and driving forces for energy efficiency in the non-energy intensive manufacturing industry
in Sweden. Energy 2006, 31, 1836–1844. [CrossRef]
130. Faber, J.; Markowska, A.; Nelissen, D.; Davidson, M.; Eyring, V.; Cionni, I.; Selstad, E.; Kågeson, P.; Lee, D.; Buhaug, Ø.; et al.
Technical Support for European Action to Reducing Greenhouse Gas Emissions from International Maritime Transport. Available
online: https://ec.europa.eu/clima/system/files/2016-11/ghg_ships_report_en.pdf (accessed on 23 July 2021).
131. Faber, J.; Behrends, B.; Nelissen, D. Analysis of GHG Marginal Abatement Cost Curves. Available online: https://cedelft.eu/wp-
content/uploads/sites/2/2021/04/7410_defreportJFSD.pdf (accessed on 10 July 2021).
132. Fleiter, T.; Worrell, E.; Eichhammer, W. Barriers to energy efficiency in industrial bottom-up energy demand models—A review.
Renew. Sustain. Energy Rev. 2011, 15, 3099–3111. [CrossRef]
133. Broto, V.C. Urban Energy Landscapes; Cambridge University Press: Cambridge, UK, 2019. [CrossRef]
134. Wang, J.-W.; Cheng, C.-H.; Huang, K.-C. Fuzzy hierarchical TOPSIS for supplier selection. Appl. Soft Comput. 2009, 9, 377–386.
[CrossRef]
135. Coffey, L.; Claudio, D. In defense of group fuzzy AHP: A comparison of group fuzzy AHP and group AHP with confidence
intervals. Expert Syst. Appl. 2021, 178, 114970. [CrossRef]
136. Mohammadi, L.; Meech, J.A. AFRA—Heuristic expert system to assess the atmospheric risk of sulphide waste dumps. J. Loss
Prev. Process. Ind. 2013, 26, 261–271. [CrossRef]
137. Xu, Z.; Chen, J. An interactive method for fuzzy multiple attribute group decision making. Inf. Sci. 2007, 177, 248–263. [CrossRef]
138. Mu, Z.; Zeng, S.; Wang, P. Novel approach to multi-attribute group decision-making based on interval-valued Pythagorean fuzzy
power Maclaurin symmetric mean operator. Comput. Ind. Eng. 2021, 155, 107049. [CrossRef]