An Analysis of Barriers to Sustainable Supply Chain Management Implementation: The Fuzzy DEMATEL Approach

: This study aims to identify and analyze the sustainable supply chain management (SSCM) implementation barriers in the context of Iran to determine the barriers and establish a cause-and-effect relationship between them. The Fuzzy Delphi Method (FDM) is used to assess the barriers to SSCM implementation in Iran, and a fuzzy decision-making trial and evaluation laboratory (DEMATEL) is used to obtain qualitative assessments from multiple experts regarding how to mitigate the impact of the barriers. To validate the proposed methodology, out of 20 barriers identiﬁed through the literature review and 9 more barriers suggested by experts, 26 barriers are identiﬁed to be relevant to the context of Iran using FDM. Six SSCM implementation barriers (i.e., economic sanctions; banking problems; economic instability; high inﬂation rate; lack of regulation and guidance from the government and lack of a globally competitive environment) were determined as important barriers in the cause group. All of them are related to Iran’s economic situation, the economic sanctions against Iran, and the government’s neglect of the importance of sustainability issues, all of which are obviously out of the companies’ control; moreover, the impact-relation map for each has been developed to facilitate the visualization of its interactions (i.e., inﬂuencing and inﬂuenced) with other barriers. The outcomes achieved are unique to Iranian companies and cannot be generalized to every company in other countries; however, the proposed approach might serve as a foundation for analyzing and comprehending the interactions between barriers. The suggested technique would identify the most effective areas for directing resources and efforts to reduce the effects of barriers to successful SSCM implementation.


Introduction
Sustainability has drawn business and academia's attention in recent years as a result of the world's present industrial growth, which has created several social and environmental concerns [1]. Since sustainability has become a global concern, many forward-thinking organizations are revisiting their supply chain operations to account for their supply chains' environmental and social consequences [2][3][4]. Traditional supply chain management (SCM) accomplishes SSCM by addressing economic, environmental, and social issues through the "triple bottom line" concept [5,6]. According to Seuring and Muller [7], SSCM refers to the management of material, information, and capital flows, as well as cooperation among companies along with the supply chain, while taking into account the economic, environmental, and social dimensions based on customer and stakeholder requirements.
Sustainability practices enable organizations to improve the environmental and social performance of their supply chains and to build competencies that can be used to gain a competitive edge through sustainability initiatives [8] as cited in Ref. [9]; however, sustainability-related efforts and sustainability practices' implementation are frequently affected by a variety of internal and external factors [10], which are known as "drivers and barriers". In developing countries, barriers are more prominent than drivers. Additionally, emerging and developing economies' SSCMs face greater implementation challenges than those of developed ones [11]. Many organizations in developing countries are moving toward SSCM, but they are encountering challenges in implementing it [1]. These challenges are influenced by a variety of cultural, political, and economic factors.
The identification of the barriers to the adoption and implementation of sustainable practices in various industrial sectors is a critical stage in the implementation of sustainable practices [12,13]. Many researchers have worked on identifying barriers to SSCM adoption and implementation in various countries and industries; however, limited research has been conducted in the context of developing countries [14], particularly Middle Eastern countries [15,16]. In the context of Iran, Heidary et al. [17] identified the seven most important barriers to SSCM implementation in the Iranian Iron and Steel Development industry; however, they did not consider Iran's economic situation and the challenges it entails. Furthermore, they only focused on the Iron and Steel Development business; moreover, the interrelationship of barriers was not investigated. Narimissa et al. [18] employed the Delphi three-round technique to identify and prioritize the drivers and barriers to SSCM implementation in the Iranian oil sector; however, this study just focused on Iran's oil sector. Movahedipour et al. [19] used ISM to identify barriers to SSCM implementation in Iran's electrical power supply businesses; this study, however, just relied on Iran's electrical power supply industries. Mohammadjafari et al. [20] categorized 20 barriers to green SCM implementation in Iranian businesses using AHP.
After reviewing the existing literature on the barriers to SSCM implementation in the context of Iran, it is found that there are not many studies on this subject; moreover, the existing literature has just focused on single industries rather than comprehensive and general research covering these barriers in the context of Iran: Iran's economic and social situation has experienced a tremendous impact in recent years, which could impact all aspects of Iranian companies' plans and strategies without considering their activities. In addition, since domestic demands have increased in recent years due to economic sanctions on Iran, Iranian enterprises have expanded their activities in recent years to meet these demands, posing additional environmental and social challenges. Therefore, it seems necessary that greater attention be paid to comprehensive research on SSCM implementation in the context of Iran. On the other hand, the existing studies on this subject did not measure the interrelationship between barriers to SSCM implementation, which is vital for decision-makers to prioritize barriers to overcome.
The aforementioned research gaps have contributed to the selection of this topic. The present work addresses these gaps with a two-phased research method that comprises: Phase (1) A fuzzy Delphi study to identify the most related barriers in the context of Iran; Phase (2) Identifying the interrelationship between the barriers to SSCM implementation using the fuzzy DEMATEL method to provide a comprehensive understanding of barriers and their priority to be overcome. The following are the research aims:

•
Identifying the main barriers encountered while implementing SSCM; • To model and comprehend the complex relationships among SSCM barriers.
The barriers to the implementation of SSCM are identified and classified using an extensive literature review. The FDM is applied to reidentify SSCM barriers based on Iran's situation; then the causal relationships among these barriers are analyzed by the fuzzy DEMATEL method.
The remainder of this paper is organized as follows: In Section 2, the barriers to implementing SSCM are summarized by referring to the literature. Section 3 presents the basic theories of related methods, and then the research framework with integrated techniques is elaborated. In Section 4, detailed calculations and analyses are carried out to recognize the barriers to SSCM implementation, and then, relative suggestions based on the discussions are given. Finally, conclusions are drawn in Section 5.

SSCM in Developing Countries
Research on SSCM in developing countries is in its early stages [26], as cited in Ref. [27]. Developing countries are distinguished by lower per capita income, commercial or industrial activity, and inadequate infrastructure [28]. In these countries, the dynamic and uncertain character of the business environment, combined with the lack of institutions, impedes supply chains from learning and developing, thus limiting their potential to achieve sustainability; moreover, companies have adopted a basic stance on their contributions to sustainability efforts and practices [29].
Since 1992, there have been over 900 publications in the management literature dealing with sustainable supply chains, with the most prominent studies coming from a small group of experts, mostly from Europe, the United States, and certain Asian countries [14]. Based on the findings of [26], which were the result of an extensive bibliometric and network analysis, SSCM-related research in the rest of the world, especially in developing countries, is limited. Furthermore, this issue is more prominent in Middle Eastern countries and it has been highlighted by a large body of studies in the SSCM literature: Flores et al. [15], conducted a systematic literature review of 56 articles published between 2010 and 2020 on SSCM in emerging economies, and discovered that only two research works on SSCM have been conducted in the context of Middle Eastern countries; additionally, Jia et al. [16] adopted the same method as Flores et al. [15] to examine 85 academic articles on SSCM Sustainability 2022, 14, 13622 4 of 30 in developing countries published between 2000 and 2016 and found out that just four of them were conducted in Middle Eastern countries.
The following are some SSCM studies conducted in the context of developing countries: Khalid and Seuring [30] presented an understanding of the importance of SSCM activities in emerging economies; Zailani et al. [31] analyzed the relationship between SSCM practices and outcomes in economic, environmental, social, and operational performance in Malaysia; their findings have empirically demonstrated that SSCM practices have a strong effect on long-term supply chain performance, particularly from an economic and social standpoint. Furthermore, to study the social dimension; Lund-Thomsen et al. [32] reviewed the literature on social responsibility in developing countries' industrial clusters and discovered a lack of participation in issues of social development and poverty alleviation; Mani et al. [33] studied supply chain social sustainability in developing countries by identifying significant characteristics of social sustainability at the focal company level as well as first-tier suppliers and customer level; moreover, Ali and Kaur [34] conducted a study on the social dimension of sustainability to investigate the efforts and techniques followed by logistics enterprises, notably warehouse operation companies. In this research, social sustainability practices have been discovered and finalized using the Best Worst Method (BWM) and are graded according to their impact on enhancing firms' social sustainability footprint. Additionally, numerous empirical studies on SSCM have been conducted in developing countries, examining the extent to which SCM activities are adopted by businesses in developing countries and underlining the crucial role of sustainability in their performance [35,36]. Additionally, Esfahbodi et al. [37] conducted an empirical research study of 128 manufacturing enterprises, 72 of which were in China and 56 of which were in Iran; their findings indicate that using SSCM procedures helps Chinese and Iranian enterprises enhance their environmental performance. Another empirical study on this issue was undertaken by Younis et al. [38], who examined the relationship between sustainability approaches and three intangible resources: innovation, human capital, and ethical culture in the United Arab Emirates; moreover, some empirical studies were conducted on the role of blockchain and big data analytics in SSCM in developing countries by Kshetri [39] and Mageto [40], respectively. Kshetri has proposed seven propositions to investigate how blockchain might assist diverse stakeholders to overcome a variety of challenges in boosting SSC in developing countries [39]. Mageto has also contributed to the SCM literature conceptually and methodologically by connecting big data analytics and SSCM in industrial SCs by utilizing Toulmin's argumentation model in management studies [40]. Furthermore, Galal et al. [27] also came up with a methodology for evaluating supply chains in developing countries that integrates the three dimensions of sustainability.
Furthermore, extensive literature reviews have been published in this area. For instance, Jia et al. [16] conducted a comprehensive review of published articles between 2000 and 2016. Four common themes emerged from the research on SSCM adoption: motivations, barriers, mechanisms, and outcomes. To explain the adoption of sustainable supply chain practices in developing countries, a conceptual model was developed that integrates these aspects and is based on institutional theory; moreover, Flores et al. [15] conducted a systematic review of the literature, analysing 56 articles published between 2010 and 2020. The findings indicate that context is critical when doing empirical or case study research in developing countries. Furthermore, from the perspective of a growing economy, it is vital to study the integration of the three aspects of sustainability and their impact on supply chain performance; moreover, Sahu et al. [41] conducted a descriptive content analysis of 39 publications published between 2007 and 2020. The findings illustrate the significance of context while doing empirical or case study research in developing countries. Furthermore, the findings strongly suggest that more research should be conducted into the integration of the three sustainability pillars and their impact on supply chain performance in the context of a developing economy. It is worth noting that Flores et al. [15] and Sahu et al. [41] both employed the systematic literature review method and had extremely similar results. Another significant challenge of research in this area is identifying SSCM implementation and adoption barriers, which are crucial to the effective and successful implementation and adoption of SSCM. Several scholars have attempted to reveal SSCM implementation and adoption barriers in developing countries [1,15,19]. Due to the critical nature of this subject for our study, a comprehensive literature review of SSCM implementation barriers will be provided in the following section.

Identification of Key Barriers to SSCM Implementation
In its simplest form, a barrier is a concept that limits a person's and/or a business's capacity to participate in an activity. Companies that implement SSCM practices may encounter barriers that prevent or significantly reduce the success of SSCM practices. Generally, barriers in the literature have been identified through industry or countryspecific studies [6,42,43].
The first step in making research items is to review relevant literature that fits the study's scope [44]. For this purpose, a comprehensive review of the literature on barriers to SSCM implementation was conducted, most of which were conducted in the developing countries' contexts. The procedure started with a comprehensive search in the databases, such as SCOPUS, Emerald Insight, MDPI, Elsevier, Science Direct, Springer, and Taylor & Francis from 2008 to 2020. The keywords used to search for articles are selected from four categories as follows: • Barrier and its synonyms, such as impediment and hurdle; • Sustainability and its related words, such as Corporate Social Responsibility (CSR), environment, triple bottom line, and sustainable; • Supply chain management and its related words, such as supply chain and value chain; • Developing countries and their related words, such as developing nations and emerging economies.
The identified barriers to SSCM implementation will be illustrated in Table 1, which will be used to develop the research questionnaire for the FDM; moreover, to provide an in-depth overview of the barriers, they were classified according to their areas into financial, technology, policy, and human resource categories to cover the main areas of an SC system. The features of each category will be presented in the following to provide an understanding of each of them.

•
Financial barriers are related to an organization's financial condition, the country's economic situation, financial resources, and costs; • Technology barriers include those barriers that are related to R&D, monitoring and information systems, and equipment and machinery; • Policy barriers can be defined as barriers that are related to planning and strategy of the whole chain, countries' rules and regulations, and capabilities; and • Human resource barriers include human skills, abilities, experiences, and behaviors at all levels of the supply chain.

Methodology and Framework
The Delphi method has been used for over a half-century to acquire group knowledge, and to do so, expert views are used to calculate the Delphi procedure [61]; however, human judgment assessment is affected by many dimensions of an individual's life experience, emotions, perceptions, subjectivity, and personal issues [62], which increase the ambiguity and uncertainty in experts' opinions. To compensate for the deficit of the traditional Delphi method, Ishikawa [63] developed a modified and improved version of the traditional Delphi approach with triangular fuzzy numbers, which is called the Fuzzy Delphi Method (FDM). FDM has many pros, such as being better for real-world decision-making, having a lower level of errors, and being applicable in only one round [61]; moreover, the capacity to get an expert opinion, reach consensus, foresee trends, and connect with research participants without being constrained by time or geography can be considered as the other advantages of FDM [64]. Based on these points and a review of the methodology of previous research conducted to forecast and screen items [63,65,66], FDM is chosen to investigate the barriers to SSCM implementation in Iran.
On the other hand, the fuzzy Decision Making Trial and Evaluation Laboratory (DE-MATEL) approach is a well-known and extensive method for generating structural models to define the casual links between complicated real-world elements [67]. In practice, expert opinions are subjective, with considerable inaccuracy and ambiguity [68], and fuzzy sets can be used to resolve the inaccuracy and ambiguity in decision-making processes and gain expert viewpoints. The fuzzy DEMATEL technique can model the relationships between the criteria and compute the significance of each criterion by taking this relationship network into account [17]. Reviewing the methodology of the literature [67][68][69] and considering the main aims of this study, which are to identify the most significant barriers to SSCM implementation in Iran and design a framework for prioritizing barriers to overcome, the fuzzy DEMATEL seems a suitable method to apply in this research.

Fuzzy Set Theory
L.A. Zadeh [70] pioneered the Fuzzy Set Theory, a mathematical theory that employs exact numerical intervals to solve complex real-world issues [71]. Numerous fuzzy membership functions exist to describe various fuzzy concepts: the triangle fuzzy numbers (TFNs); trapezoidal; and their extended forms are among the most frequently used concepts for quantifying linguistic variables [72]. Due to TFNs' practical applications [73], we employ it in fuzzy logic to identify the key barriers to implementing SSCM.
A Fuzzy Number: A fuzzy number is a generalization of a regular, real number; it refers to a connected set of possible values where each possible value has its own weight between 0 and 1. A fuzzy number is thus a special case of a convex, normalized fuzzy set of the real line [74]. Triangular Fuzzy Number: A fuzzy number A = (a, b, c), is called a triangular fuzzy number, if its membership function is given by: Let A 1 = (a L 1 , a M 1 , a R 1 ) and A 2 = (a L 2 , a M 2 , a R 2 ) be TFN S . Some basic operational principles in this paper are as follows: A graded mean integration representation (GMIR) technique is used to remove the ambiguity of a TFN A = a L , a M , a R in order to directly reflect the criteria.

Fuzzy Delphi Method
Ishikawa et al. introduced the Fuzzy Delphi Method (FDM), which was derived from the traditional Delphi approach and fuzzy set theory [63]. The traditional Delphi technique has been widely used in decision-making, policy formulation, long-term forecasting, etc.; however, it has some drawbacks when it comes to resolving intangible decision information, such as a high workload and executory costs, an excessively long feedback time, a low convergence rate, and statement misrepresentation [71]. Therefore, the Delphi method was modified to address the shortcomings of the traditional method using fuzzy logic theory. The following are the advantages of this technique: (1) It can reduce the length and expense of the study by limiting the number of Delphi rounds to one; (2) The questionnaire is more structured; (3) There is a high response rate; (4) It is easier to achieve consistent and stable outcomes [75]. FDM can also screen many variables and create the foundation for DEAMTEL fuzzy analysis.
The following procedures must be followed when using the FDM technique in a study: Step 1: Creating a relevant questionnaire to evaluate the importance of potential variables: In this study, 11-point Likert scales are employed to improve data processing and data reliability caused by lower measurement error [76]. The numbers 0 and 10 represent "extremely unimportant" and "extremely important", respectively.
Step 2: Establishing a team of experts from academia and industry familiar with the SSCM as well as Iran's economic and political structures who will participate in the survey.
Step 3: Emailing the questionnaire to the respondents and gathering responses, indicating (measuring) the degree of agreement with each item; this stage involves inviting experts to answer a web-based questionnaire.
Step 4: After receiving expert views and corresponding comments, the next stage is to collect statistics from questionnaires, sort the maximum and minimum values of all score intervals, and calculate the geometric mean values for each factor. The maximum score interval is used to define optimistic cognition, whereas the minimum score interval is used to describe pessimistic cognition. An optimistic TFN O j = L o j , M o j , U o j and a pessimistic TFN P j = L P j , M P j , U P j for criterion j should be gathered, where j = 1, 2, . . . , N.
Step 5: Calculating the consistency of the questionnaire results: To measure the consistent class of decision makers' remarks on each factor, the consensus significance value C j for factor i is computed. The larger the C j number, the higher the consistency. The processing can be carried out by: 1. If L o j ≥ U p j , and factor j satisfies a consensus, the C j value is: (2) and (3). a. If T j is smaller than the interval H j = U o j − M p j , the comments form a consensus.
b. If T j is greater than H j , the comments on criterion j do not reach consistency. A new check for factor j should be conducted, followed by the completion of a report informing all experts of the other experts' views; this new questionnaire and report will be submitted to the experts, and Step 3 will begin over.
Step 6: Setting a threshold, θ, compared by the C j value to distinguish the significant criteria: The θ value might be determined by the experts as a minimum degree of acceptable consistency. If C j is less than θ, criteria j is eliminated from the evaluation index system, and the remaining factors can form a final index system.

The Fuzzy Decision-Making Trial and Evaluation Laboratory (DEMATEL)
The DEMATEL approach is a well-known and comprehensive technique for constructing a structural model that provides the casual relationships between complicated real-world variables [67]. DEMATEL is a group decision-making process that entails eliciting ideas and then analyzing the cause-effect connection of complicated issues [77]. An IRM can be used to organize the causal links between variables [57]. Assuming that the final index system created using the FDM comprises a set of barriers: B = B1, B2, . . . , Bn. Using pair-wise comparison, the criteria are to be ranked for influence against one another. According to Dalalah et al. [77], the pair-wise rating scale in this study is split into five levels, as indicated in Table 2. The DEMATEL model is built around direct-relation fuzzy matricesZ (k) ∀k = 1, . . . , p, whereZ (k) is (n × n) matrix, n denotes the number of criteria, and p is the number of experts that provide comments along with linguistic evaluations on pairwise criteria comparisons based on Table 2.
Assumez ij = z ij,l ,z ij,m ,z ij,u represents a TFN of expert k's linguistic rating. Equation (7) illustrates a fuzzy matrixz (k) . To determine the relationships between the influencing components, the following mathematical steps are used: Step 1: Initializing direct-relation fuzzy matrix tildeX of the average of matricesx 1 , x 2 . . . , x p , which displays the experts' views on how each criterion affects the others, can be determined using the following formula: Step 2:X should be normalized using a linear scale variation. Assume thatS is a normalized direct relation matrix and that TFNS ij = L s ij , M s ij , U s ij is a component ofS. A normalized matrix is obtained by referring to Equations (9) and (10).
Step 3: Splitting the fuzzy numbers inside the initial direct-relation fuzzy matrix (S) into independent sub-matrices, X l , X m , and X u , as follows: Step 4: Achieving the total-relation fuzzy matrixT, by calculating the following term: As a result,T will be as follows: Step 5: Calculating the sum of rows and columns ofT: The sum of the rows and the sum of the columns of the sub-matrices T l , T m , T u , indicated by the fuzzy number D i and R j , respectively. where:t ij=(t ij,l ,t ij,m ,t ij,u ) Step 6: To confirm the performance of influencing factors, D i − R i and D i + R i should be computed. In general, D i − R i indicates whether the factor is the cause or the effect. D i + R i displays the effect intension of the factor among the criteria. The higher the value, the stronger the effect. where: The following equations are used to obtain a precise value for D i − R i and D i + R i , which are denoted as W * i and V * i , respectively.
Step 7: Creating a structural model to represent causal links: By graphing a sequence of data pairs (W * i , V * i ), a cause-and-effect relationship diagram can be created.

The Proposed Research Framework
A framework is proposed for identifying the key barriers to the implementation of SSCM in Iran. According to FDM, important barriers should be selected first from the primary index system. Then, using the fuzzy DEMATEL approach, the causal relationships between these important SSCM barriers are investigated. The integrated MCDM approach is comprised of the two-phase shown in Figure 1.
Phase 1: Creating expert groups and collecting potential barriers to building the index system. Professors, researchers, and company managers experienced with SCM, sustainability concepts, and familiar with Iran's business culture and environment, as well as its economic circumstances, are chosen to form relevant expert groups. All potential barriers are gathered and classified by following an extensive literature review, and an initial criterion system is then established. Afterward, significant SSCM barriers using the FDM are recognized. To begin with, a questionnaire was distributed to expert groups about the major barriers to implementing SSCM in the Iranian context. Then, experts' score intervals are collected to demonstrate the relevance of the barriers. Following that, the consensus significance values for each component are determined using optimistic and pessimistic TFNs. Finally, key barriers are identified using a criterion set by expert groups.
Phase 2: Processing a structural model for causal links and identifying critical barriers to SSCM implementation using the fuzzy DEMATEL approach. A questionnaire on the relationships between key barriers is issued to obtain the fuzzy scores of comparisons. Then, using the computation and normalization of the direct-relation fuzzy matrix, the total-relation fuzzy matrix is created. Row values D, column values R, D i − R i and D i + R i are calculated and de-fuzzified using the total-relation fuzzy matrix. Finally, a structural model of SSCM implementation barriers is developed using two-dimensional criteria, with important barriers identified by in-depth analysis.

Expert Panel and Data Collection
The establishment of an expert panel is one of the first steps in any FDM research and is critical to its success due to the panel's composition having a significant impact on the quality of the results. Knowledgeable and experienced experts can ensure the integrity and correctness of related data and research findings [71]. The expert selection technique advocated by Williams and Webb, who propose the formulation of inclusion criteria to which potential participants are then matched, is used for this study [78]. Based on their expertise in the major issues of this research, namely SCM, sustainability, and Iran's business culture and economic condition, experts were matched to the following inclusion criteria: Have experience in sustainability and/or SCM, as evidenced by: • At least two years of academic experience in the fields related to SCM and/or sustainability at an Iranian institution or at a foreign institution, in which the expert is familiar with Iran's business and social culture and economic condition, or; • At least two years of employment as an SCM and/or sustainability practitioner in Iranian enterprises, or at a foreign enterprise, in which the expert is familiar with Iran's business and social culture and economic condition, or; • Having publication in reputable journals in the study fields in the context of Iran, or; • Employment in Iran's governmental or non-governmental organizations dealing with sustainability concepts.
The minimum sample size for FDM studies is 10 experts to get a high degree of consistency across experts [66]. In this study, a total of 28 identified experts from two groups of academicians and practitioners were invited to fill out 2 questionnaires for FDM and Fuzzy DEMATEL, resulting in 15 responses to the FDM questionnaires over the 10week period, and 11 responses to the Fuzzy DEMATEL questionnaire in 8 weeks. An overview of the expert panel and their response rates are provided in Table 3. The significance of barriers to SSCM implementation in Iran was initially expressed by experts from the two groups. The threshold value for the FDM is set by the majority because "the minority is subordinate to the dominant" [79]. Then the experts' opinions on the relationships between each pair of barriers were collected.

Creating the Primary Index System
To comprehensively reflect the barriers to SSCM implementation in Iran, four categories of barriers are integrated into the evaluation system, including Financial (B1), Technological (B2), Policy (B3), and Human Resource (B4). Barriers were determined by an extensive review of the literature, and a primary evaluation system including potential barriers was developed and listed in Table A1 (Appendix A). As a result of the FDM process, experts determine the barriers to SSCM implementation based on Iran's business, social culture, and economic situation.

Identifying the Significant Barriers to SSCM Implementation in Iran
The FDM was initially used to identify significant barriers. All opinions made by expert groups regarding the relative importance of barriers, and were reported as interval numbers. Following the survey, the data were collected and collated, and the findings are presented in Table 4. Conservation and optimistic remarks in terms of the aforementioned score intervals were integrated into pessimistic and optimistic TFNs, respectively. Then, using step 5 of the FDM introduction, the consistency of each expert group's viewpoints was determined. Finally, major influencing factors are found using the consensus value C j obtained from Equations (5) and (6). Additionally, the expert panels concluded that the threshold value should be set at 7.00. As a result, the significant barriers denoted by "Accepted" in Table 5 were chosen using the FDM. The results of the calculation are shown in Table 5. Lack of a global competitive environment;

3.
Lack of gender balance in the board of directors; 4.
Low cost of energy, e.g., electricity, natural gas, and oil; 7.
Inadequate monitoring and control of importing used equipment and machineries; 8.
Lack of sustainability and CSR committees in enterprises; 9.
Lack of rules and regulations for sustainability reports; Next, the suggested barriers in the open-ended question are distributed to expert groups in the form of a questionnaire to determine their relevance, and the data was then collected, analyzed, and summarized in Table 6; moreover, the final index system is established and illustrated in Figure 2.

Creating the Structural Model for Causal Relationships
The inner causal relationships of 26 barriers, which were determined by FDM in the previous phase, were explained based on the Fuzzy DEMATEL method. To begin, experts provided their opinions on barriers using fuzzy ratings in accordance with Table 2. Second, by gathering linguistic ratings, the initial integrated direct causal relationships fuzzy matrix was created based on Equation (8). Then, the fuzzy matrix was normalized using Equations (9) and (10). Third, the fuzzy total-relation matrix is constructed by applying Equations (11)- (13). Unfortunately, due to the large size of the initial and normalized directrelation matrices and the fuzzy total-relation matrix of SSCM implementation barriers, it is not possible to present them here.
Based on the former steps, Equations (14) and (15) were used to calculate the sum of rows and columns values, denoted as D and R, respectively. Then, these values were transformed to v * i and w * i values using Equations (18) and (19). The ability to influence the other barriers is denoted by w * i ; positive values indicate that the barrier is the cause of the other barriers and if the value is negative, it indicates that the barrier is being influenced by other barriers. v * i represents the correlation intensity of the effects of criteria on one another. The higher v * i , the more significant the barrier. The above process was used to prioritize the barriers to SSCM implementation and also rank and classify them into the cause-and-effect groups, as shown in Tables 7-9.
To determine the significance of SSCM implementation barriers, the v * i was used to rank them (Table 8). For the Iranian organizations, the most significant barrier to SSCM implementation was inadequate sustainability research and development (B2.3), with a v * i value of 1.606; the least important barrier was the lack of suppliers' capability and facility to perform sustainability (B4.4), with a v * i value of 0.511. Table 8 summarizes the degree of importance of each barrier. A Pareto chart (Figure 3) 3) were identified as key SSCM implementation barriers using the Pareto chart. These barriers have a significant negative impact on the implementation of the SSCM. To begin, Iranian businesses must focus on overcoming these identified key barriers, and once the level of SSCM implementation reaches a targeted level, the SSCM implementation process will be recalibrated.   The barriers to SSCM implementation were classified into cause-and-effect groups using the w * i ( cility to perform sustainability (B4.4), with a * value of 0.511. Table 8 summarizes the degree of importance of each barrier. A Pareto chart (Figure 3) was created to highlight the group of significant barriers to SSCM implementation. Eighteen SSCM implementation barriers (i.e., B2.3, B3.1, B4.1, B4.3, B1.4, B1.5, B3.7, B4.2, B1.6, B1.1, B1.7, B3.3, B3.2, B2.1, B3.9, B3.5, B2.2, B1.3) from four categories (financial category: 6, technological category: 3, policy category: 6, and human resource category: 3) were identified as key SSCM implementation barriers using the Pareto chart. These barriers have a significant negative impact on the implementation of the SSCM. To begin, Iranian businesses must focus on overcoming these identified key barriers, and once the level of SSCM implementation reaches a targeted level, the SSCM implementation process will be recalibrated.   The barriers to SSCM implementation were classified into cause-and-effect group using the * (Table 9). Twelve barriers were identified in the cause group (i.e., B1.5, B1.      Moreover, as illustrated in Figure 6, a casual diagram was created to provide a clear understanding and examine the significance and impact of the SSCM implementation barriers.

R PEER REVIEW 22 of 36
Moreover, as illustrated in Figure 6, a casual diagram was created to provide a clear understanding and examine the significance and impact of the SSCM implementation barriers. To organize the causal relationships between SSCM implementation barriers, an IRM was developed for each barrier. Since this study includes 26 barriers to SSCM implementation, it is difficult to represent the interactions of all barriers in a single IRM. As a result, an IRM has been developed for each barrier based on the reduced total-relation matrix (see Table 10) to depict its interactions (i.e., influencing and affected) with other barriers. To obtain the reduced total-relation matrix, the fuzzy total-relation matrices should first be turned into defuzzied total-relation matrices using Equation (4). Then take the mean of To organize the causal relationships between SSCM implementation barriers, an IRM was developed for each barrier. Since this study includes 26 barriers to SSCM implementation, it is difficult to represent the interactions of all barriers in a single IRM. As a result, an IRM has been developed for each barrier based on the reduced total-relation matrix (see Table 10) to depict its interactions (i.e., influencing and affected) with other barriers. To obtain the reduced total-relation matrix, the fuzzy total-relation matrices should first be turned into defuzzied total-relation matrices using Equation (4). Then take the mean of all the numbers in the matrices as the threshold to identify the significant interrelationships. Finally, the values that are less than the threshold value (0.020) should be eliminated.
IRMs for all the barriers to SSCM implementation have been developed, but for ease of presentation, only the IRM for B1.1 is shown in Figure 7; moreover, the strength of influencing on and influenced by other barriers for each barrier can be retrieved from the reduced total-relation matrix. The strength of influencing on and influenced by the other barriers for B1.1 is shown in Table 11.

PEER REVIEW 24 o
IRMs for all the barriers to SSCM implementation have been developed, but for e of presentation, only the IRM for B1.1 is shown in Figure 7; moreover, the strength influencing on and influenced by other barriers for each barrier can be retrieved from reduced total-relation matrix. The strength of influencing on and influenced by the ot barriers for B1.1 is shown in Table 11.      Furthermore, each barrier influences and is influenced by the number of barriers, and the importance of each barrier is determined by its own Pareto chart. The details are shown in Tables 12 and 13.

Discussion
This study analyzed the significant barriers to SSCM implementation and their inner causal relationships in the context of Iran. 26 SSCM barriers were identified, prioritized, ranked, and classified into the cause-and-effect groups. In general, "inadequate sustainability research and development (B2.3)" from the category of "technology" has been recognized as the most significant barrier to SSCM implementation in Iran; this is also confirmed in other studies [1,80], in which presenting managerial solutions to sustainabilityrelated issues can be addressed by R&D; moreover, inadequate R&D is found in [53] as the top barrier to implementing and adopting SSCM in Indian companies. On the other hand, based on our results, the least important barrier to SSCM implementation in Iran is the "lack of suppliers' capability and facility to perform sustainability (B4.4)", while Sajjad et al. [45] concluded that the lack of supplier ability is of high importance in SSCM implementation.
A The results show that 85% of the "financial category", 100% of the "technology category", 60% of the "policy category", and 50% of the "human resource category" from the final index system are of the most significant barriers. To ease of presentation, only the first three barriers in the list will be discussed as follows: 1. "Inadequate sustainability research and development (B2.3)" is the most important barrier to SSCM implementation in Iran, as mentioned above. 2. "Lack of commitment and support by the top management level (B3.1)" is one of the most repeatable barriers to SSCM implementation and adoption in the literature; this finding is consistent with the previous studies [1,18,24,43,45], which they investigated the importance of the support and commitment of the managerial board to obtain sustainability values. 3. "Inadequate sustainability training and education (B4.1)" was found as the third important barrier; this is also confirmed by Ref. [18], in which this barrier is found as the top barrier to implementing SSCM practices in Iranian oil companies. In addition, some studies confirm our findings in a way, in which they have investigated sustainability training and education as a driver: Moktadir et al. [43] have found a high impact for training and education in the context of the leather industry as a driver to sustainable manufacturing practices and circular economy; Moreover, on a similar basis, Zaabi et al. [6] found training and education as an important driver for Indian manufacturing companies toward sustainability. On the contrary, some other studies [81,82] found training and education without a considerable impact on SSCM practices.
To provide a clear understanding of the relationship between barriers, the identified barriers have been clustered into cause-and-effect groups.
The cause group contains 12 barriers from three categories: financial, policy, and human resources. It should be noted that half of the barriers come from the financial category. By using a Pareto chart, the most important barriers of the cause group (B1.5, B1.6, B1.4, B1.7, B3.9, and B3.5) are defined and as can be seen, two-thirds of the barriers are from the financial category and the rest are from the policy category. It shows that the financial barriers have the most significant impact on SSCM implementation in Iran. Unfortunately, all the financial barriers in the cause group are related to Iran's economic situation and the economic sanctions against Iran, all of which are out of the companies' control; these results could be a gentle reminder to the Iranian government of how economic sanctions and the problems they entail could affect all aspects of Iranian life. In the following, these barriers will be discussed: 1. "Economic sanctions (B1.5)" are shown to be the most influencing barrier to SSCM implementation in the cause group. Since economic sanctions are not a usual barrier and it has just come true in the context of a few countries, there is not any other research to define it as an SSCM implementation barrier. There are, however, studies covering environmental aspects, e.g., the overview by Madani [83], in which the environmental impacts arising from international economic sanctions on Iran are highlighted as one of the most possible reasons for unintended environmental issues; moreover, Le and Hoang [84] also investigated the negative influences of international economic sanctions on environmental performance, and ranked it as the most significant barrier to environmental performance in developing countries. 2. "Banking problem (B1.6)" is recognized as the second prominent barrier in the cause group, which is directly caused by economic sanctions. 3. "Economic instability (B1.4)" is the third highlighted barrier in the cause group. Movahedipour et al. [19] have also defined economic instability as a significant dependent barrier to SSCM implementation [19]. 4. "High inflation rate (B1.7)" is the fourth important barrier in the cause group, and it is from the financial category as well, like the first three barriers. While the sanctions are certainly not the main cause of Iran's high inflation rate, they have had an incontestable impact on Iran's inflation rate. It is worth mentioning that a high inflation rate leads to economic instability, and as far as we know from the literature review, a high inflation rate is a new barrier to SSCM implementation. *After the deliberation of the first four barriers of the cause group, it is recognized that economic sanctions are the cause of the other highlighted barriers from the financial category. 5. "Lack of regulation and guidance from the government (B3.9)" is the fifth most important barrier out of six barriers in the cause group. Several studies in the literature support the importance of this barrier: Movahedipour et al. [19] highlighted that the lack of regulation and governmental policies leads to the irresponsibility of the Iranian industry toward social responsibility; moreover, Faisal et al. [36] have confirmed that the absence of related rules may decrease the speed of SSCM adoption. Sajjad et al. [45] have also concluded that these rules and regulations are promising drivers for the SSCM implementation; others have also defined the lack of regulation and guidance from the government as a top-significant barrier [24,43]. 6. The last significant barrier in the cause group is the "lack of global competitive environment (B3.5)"; this barrier has also been extremely influenced by economic sanctions.
Since the lack of a globally competitive environment is a specific barrier to a few countries, which are marked as high-risk and dangerous for financial transactions, this barrier was not mentioned earlier in the literature. 5. "Lack of new technologies, materials, and processes (B2.2)" comes after B3.3 in the list of important barriers in the effect group. In a study by Govindan et al. [12], the lack of new technologies, materials, and processes is ranked as the most important barrier in the technology category [12]; however, in another study, Menon et al. analyzed SSCM implementation barriers in the Indian electronics industry, and the lack of new technologies, materials, and processes is ranked as an effective but not important barrier [1]. 6. "Old equipment and machinery (B2.1)" is the sixth barrier in the list of highlighted barriers in the effect group. Narimissa et al. [18] have also defined the old equipment and machinery as a barrier to implementing SSCM in the Iranian oil industry. 7. "Behavioral and psychological barrier (B4.3)" is the last significant barrier in the effect group; this barrier is found to be the most influenced barrier to the implementation of SSCM.
In this study, the reduced total relation matrix (Table 10) is presented to assist managers of industries, policymakers, and decision-makers in easily understanding the casual interrelationships between barriers to SSCM implementation in Iran; moreover, the influencing strength of each barrier on the others and the influenced strength of each barrier from the others are investigated. These results help decision-makers decide the priority of barriers to overcome. For instance, about B1.1 (Table 11), B1.5 is the most strongly influencing barrier on B1.1, and B4.2 is the most strongly influenced by barrier B1.1.
Furthermore, in this research, a list of SSCM implementation barriers in Iran, which are Influencing/influenced by each other, was presented (Tables 12 and 13). In Table 12, the impact of each barrier on the other barriers is identified, and the most significant ones are listed, taking advantage of a Pareto chart; it helps managers, policymakers, and decisionmakers to prioritize the most influencing barriers to overcome. Table 13 shows each of the barriers influenced by the others. Decision-makers can easily overcome a complicated barrier by focusing on the conquering challenges that have the most significant impact on the barrier.

Result
This study conducted an in-depth analysis of the key barriers to implementing SSCM. To begin with, the literature on SSCM implementation barriers was reviewed to highlight the important ideas under study. Twenty potential barriers to SSCM implementation were identified from the existing literature, and nine more barriers to SSCM implementation in the context of Iran were suggested by experts to develop an initial index system. Second, using FDM, 26 significant barriers were chosen from the pool of potential SSCM barriers. Thirdly, fuzzy DEMATEL was used to construct the causal relationships and structural models.
The main findings of this study are as follows: 26 barriers to SSCM implementation in Iran were identified using the FDM, and most of them are meaningful in other developing countries as well. While it may be difficult to manage or influence certain barriers, companies should prioritize the most significant ones. The barriers were prioritized using DEMATEL, and the most significant SSCM implementation barriers in Iran were determined to be B2. 3 as effective barriers; thus, as most of the reported significant barriers were found to be from the effect group, it is obvious that just finding the significant barriers does not indicate which barrier should be emphasized. As a result, it is critically needed to analyze barriers by classifying them into cause-and-effect groups based on their w * i value. Classification of barriers into cause-and-effect groups generally gives the relative ease of controlling or being influenced among the barriers. Unfortunately, this does not hold for Iran because the most important SSCM implementation barriers in the cause group (B1.5, B1.6, B1.4, B1.7, B3.9, B3.5) are not at the company level. All of them are related to Iran's economic situation and the economic sanctions against Iran that are not controllable by the companies; however, Iranian companies can focus on overcoming the rest of the barriers in the cause group, which are defined at the company level, by expanding their capabilities.
The interactions displayed by IRM for SSCM implementation barriers assist the organization in visualizing and focusing on the most significant barriers; moreover, companies can easily overcome the organization-related barriers by focusing on prominent inter-relationships between barriers and considering the strength of their influencing/ influenced factors.

Limitation
FDM is quite successful in identifying barriers, and the Fuzzy DEMATEL method is extremely effective for evaluating interdependent relationships among barriers and finding the critical ones through a visual structural model in an imprecise environment; however, these methods largely depend on expert feedback, and convincing experts to spend more than 30 min responding to the two-phase questionnaire within the specified time frame is difficult. Additionally, we evaluated just 26 barriers to developing the SSCM implementation framework.

Future Research
We expect that this research will help evaluate the interdependent relationships among barriers to SSCM implementation in other developing countries. Other developing countries may need to consider more or less relevant barriers in their analyses. In the future, other multi-criteria decision-making tools, such as Fuzzy-AHP, Fuzzy VIKOR, ISM, and TISM, could be used to evaluate the most influential barriers to implementing SSCM practices. In addition, Total Interpretive Structural Modelling (TISM) and Interpretive Ranking Process (IRP) methodologies can be used to determine the relationships between barriers and performance outcomes.

Conflicts of Interest:
The authors declare no conflict of interest.
Appendix A Table A1. Potential barriers to SSCM implementation.

Financial Constraints
Finance is important to the successful implementation of SSCM. Financial constraints are the most significant impediment barrier to the implementation of SSCM practices in any sector pressure for lower prices In a competitive market, companies need to provide a price that is equal to or less than the market level. Because green/sustainable products are more expensive, businesses aren't willing to use sustainable supply chain management practices in order to keep their market share High initial investment Cost is a big impediment to any firm implementing sustainable practices. Sustainable design, manufacturing, packing, shipping, and return product management all require a lot of money at the start

Economic instability
According to the literature, economic instability might be a serious barrier to SSCM implementation. Economic instability refers to an organization or a country's having financial difficulties as a result of inflation, consumer confidence concerns, growing unemployment rates, and rising prices Technology

Inadequate assessment system for suppliers
The characteristics of the raw materials might have the greatest influence on the end product's attributes. Thus, choosing the right raw materials and the right suppliers is very important for making perfect goods that please all the people who are affected by or affected by the business's work

Old equipment and machinery
According to the literature, old machinery appears to be a substantial barrier that has a high effect on SSCM practices due to other causal barriers. It is critical to manage outdated equipment and machinery in order to decrease energy consumption and pollution, as well as to encourage interior items made by local manufacturers for economic and social reasons Lack of new technologies, materials, and processes Proper infrastructural facilities, such as cutting-edge equipment and information technology, are necessary for a successful supply chain management implementation. Some materials and procedures are detrimental to the environment and hence act as a barrier. New technology, materials, and processes are necessary to replace pollution-producing goods and activities that endanger workers Inadequate sustainability research and development In SSCM, there has been insufficient emphasis dedicated to the development of theories and grounded research. R & D assists organizations in understanding the value and impact of SSCM implementation on their business, as well as in defining more effective SSCM procedures

Policy
Lack of commitment and support by the top management level Top management's unwillingness to modify current unsustainable procedures and investments is a significant impediment to successfully implementing SSCM principles. Lack of top management engagement results in unsustainable company policies. Without top management commitment, there would be no sustainability priority or proper resource allocation. There is no direction for policy development or the achievement of environmental goals Lack of performance measuring, monitoring tools and evaluation standards There is a scarcity of effective evaluation tools and standards. It is difficult to measure sustainable accomplishments, and accounting techniques limit reporting. Companies are hesitant to switch to the new supply chain approach. Hence, as a result, government monitoring and supervision is very important to make sure that SSCM procedures are being used in any business Lack of strategic planning Sustainable practices must be integrated into every organization's business decision-making process and be included in the organization's mission and vision statements. The lack of a strategic plan is a big problem when it comes to using SSCM practices.

Lack of suppliers' commitment
Suppliers' reluctance to abandon old ways influences the transition to sustainable practices. It is only possible for the supply chain network to use environmentally friendly practices if its suppliers are willing to deliver raw materials that are good for the environment Lack of benchmarking standards in developing countries Benchmarking is a critical step in determining a system's strengths and weaknesses. The lack of such a framework is a limitation to implementing the SSCM practice Lack of regulation and guidance from government In the sustainable supply chain, there is a lack of regulation and enforcement of sustainable norms. The implementation of sustainable programs is not supported by strong legislation and government commitment Lack of proper rewards and motivational programs for SSCM implementation Even if rules and regulations are in place, their effective implementation is a key issue. In order to carry out the practices, the government must adopt restrictive action. Similarly, those who effectively deploy must be appropriately rewarded. The lack of such acceptability is a significant impediment Table A1. Cont.

Human Resource
Inadequate sustainability training and education Lack of information about appropriate SSCM practices for the company is a significant barrier, and hence correct training must be provided to staff to ensure successful implementation of SSCM procedures Lack of human skills, experience, and necessary tools for implementing SSCM practices.
The human aspects of SCM, such as labour, skill, experience, and their connection, must be viewed as the primary component.
There is a dearth of expertise and experience in the subject of sustainability that members of the organization should possess. While some businesses have recognized the value of SSCM and attempted to implement it, most of these businesses lack expertise, as well as the requisite tools and managerial abilities Behavioral and psychological barriers Behavioral barriers and behavior change are major concerns. People are unwilling to change. For instance, a company has a certain procedure that has been in place for the previous 20 years, and to effectively do something different, they must manage behavioral issues. There is also a psychological barrier for many individuals who regard sustainability as a pleasant, fluffy thing we can do, but it costs more money and provides no actual value lack of suppliers' capability and facility to perform sustainably The skills and training of suppliers' human resources generate fresh ideas for businesses and allow them to more readily embrace new technology, and a lack of these matters might be considered a barrier. Companies must also spend money on environmental and social norms, which suppliers see as extra costs and an invasion of their business processes

Lack of customer awareness
According to the literature, most customers in many regions of the world, particularly in developing nations, still lack "sustainable" understanding. Consumption patterns are influenced by major prior behaviors. Increasing consumer awareness has not resulted in more sustainable consumption behavior. Moving to SSCM requires major changes in human behavior, both on an individual and organizational level