Sustainable Supply Chain Management and Multi-Criteria Decision-Making Methods: A Systematic Review

Multi-criteria decision-making (MCDM) methods are smart tools to deal with numerous criteria in decision-making. These methods have been widely applied in the area of sustainable supply chain management (SSCM) because of their computational capabilities. This paper conducts a systematic literature review on MCDM methods applied in different areas of SSCM. From the literature search, a total of 106 published journal articles have been selected and analyzed. Both individual and integrated MCDM methods applied in SSCM are reviewed and summarized. In addition, contributions, methodological focuses, and findings of the reviewed articles are discussed. It is observed that MCDM methods are widely used for analyzing barriers, challenges, drivers, enablers, criteria, performances, and practices of SSCM. In recent years, studies have focused on integrating more than one MCDM method to highlight methodological contributions in SSCM; however, in the literature, limited research papers integrate multiple MCDM methods in the area of SSCM. Most of the published articles integrate only two MCDM methods, and integration with other methods, such as optimization and simulation techniques, is missing in the literature. This review paper contributes to the literature by analyzing existing research, identifying research gaps, and proposing new future research opportunities in the area of sustainable supply chain management applying MCDM methods.


Introduction
In this competitive era, every business is part of a supply chain which involves efficient and effective movement of products or services from suppliers through to customers via manufacturers, distributors, and retailers. A typical supply chain involves multiple businesses, resources, people, technologies, and information for buying, manufacturing, distributing, storing, and selling products [1]. Several activities within a supply chain present direct social, environmental, and economic impacts [2]. These impacts are referred to as the triple bottom line (TBL) in sustainable supply chain literature. Social impact includes modern slavery, gender discrimination, unfair wages, child labor, and so on [3,4]. Environmental impact includes emission of carbon dioxide, polluting water and the environment, global warming, and so on [5,6]. Economic impact includes the return of investment, impact on profit, and productivity [2]. Considering their significant impact on society, the environment, and the economy, every supply chain is now taking steps to ensure sustainability.
Sustainable supply chain management (SSCM) integrates the economic, social, and environmental goals of the supply chain to improve long-term performance [7], evaluating and monitoring business performance against social, environmental, and economic dimensions [2]. Any good social and environmental performance with economic performance ensures better sustainability; however, ensuring all three performances are good creates the [19] √ Examples of economic sustainability include cost reduction, on-time delivery, reliability, and quality [11]. Sustainable supply chains simultaneously assess supply chain performance in terms of social, environmental, and economic aspects.
In the last few years, a good number of studies have been conducted on different dimensions of SSCM, including a number of review papers, such as a: • review of green supply chain management [35][36][37]; • review of different theories in sustainable supply chains [38]; • review of the evolution of and future challenges in sustainable supply chain management [8,[39][40][41]; • review of trends and future directions in social aspects of sustainable supply chains [42]; • review of SSCM in global supply chain context [43]; • review of drivers in SSCM [44]; and • review of MCDM methods in green supply chains which focuses only on the environmental dimension of supply chain management [22,45].
In order to become more sustainable, supply chains should implement sustainable practices, with a certain impact on various TBL areas; however, decision makers need to consider multiple criteria to evaluate suppliers, practices, success factors, drivers, and challenges in SSCM in smart ways. For this purpose, MCDM methods have been widely applied in the area of SSCM [46]. In spite of having a reasonable number of contributing articles which applied different MCDM methods in SSCM, earlier literature is lacking a review on different MCDM methods applied to SSCM areas considering the social, environmental, and economic dimensions. In this paper, we aim to fill this gap by conducting a literature review on different MCDM methods applied in SSCM, contributing to the literature by analyzing existing studies systemically and proposing a future research framework in the area of MCDM methods in SSCM.
The paper is organized as follows. In Section 2, the scope of the literature review is described. The review of both individual and integrated MCDM methods is conducted in Section 3. Section 4 explains the bibliometric analysis for published articles. Section 5 summarizes the review and research gaps. Finally, conclusions and future research directions are presented in Section 6.

Scope of the Literature Review
The Scopus database was used to collect the relevant articles with the following phrases in the article's title, abstract, and keywords: "sustainable supply chain" and "multicriteria decision making" or "multi-criteria decision analysis" or "MCDM". From the preliminary search of the literature, most of the studies in the area of SSCM modelling were found to be published since 2010, which is shown in Figure 1. Based on this observation, in this paper, the literature on MCDM methods applied in the area of SSCM is reviewed from 2010 to 2020.
After the preliminary search in Scopus, the search database was refined using the following criteria: • Other databases, such as the Web of Science and Google Scholar, were used to enhance the search. After a first screening of the articles (by title and abstract), the final subset of 106 relevant manuscripts for review was created. The inclusion criteria were articles focused on any dimension of supply chain sustainability and the search phrases appeared in the body text. The exclusion criterion was one or more keywords presented in the text or reference list without discussing supply chain sustainability using MCDM methods. After finalizing the list of articles, a deep review was conducted of the applications of different Other databases, such as the Web of Science and Google Scholar, were used to enhance the search. After a first screening of the articles (by title and abstract), the final subset of 106 relevant manuscripts for review was created. The inclusion criteria were articles focused on any dimension of supply chain sustainability and the search phrases appeared in the body text. The exclusion criterion was one or more keywords presented in the text or reference list without discussing supply chain sustainability using MCDM methods. After finalizing the list of articles, a deep review was conducted of the applications of different MCDM methods in SSCM, and a bibliometric analysis was carried out within the set of finalized articles.

Review of Applications of MCDM Methods in SSCM
This section reviews the articles on MCDM methods applied in SSCM areas. The following sub-sections review the applications of both individual and integrated MCDM methods in detail.

Applications of Individual MCDM Methods
From the literature search, 59 articles applied individual MCDM methods in SSCM. The names of the methods and their abbreviated terms are as follows: i.
Decision-making trial and evaluation laboratory (DEMATEL) and Fuzzy/Grey DE-

Review of Applications of MCDM Methods in SSCM
This section reviews the articles on MCDM methods applied in SSCM areas. The following sub-sections review the applications of both individual and integrated MCDM methods in detail.

Applications of Individual MCDM Methods
From the literature search, 59 articles applied individual MCDM methods in SSCM. The names of the methods and their abbreviated terms are as follows: i.
Decision-making trial and evaluation laboratory (DEMATEL) and Fuzzy/Grey DEMATEL ii. Analytical hierarchy process (AHP) and Fuzzy AHP iii. The technique for order of preference by similarity to ideal solution (TOPSIS) and Fuzzy TOPSIS iv. Best-worst method (BWM) v.
VIseKriterijumska In particular, applications of DEMATEL include the identification and analysis of success factors for sustainability initiatives (grey [47]), sustainable food supply chain management [48], green supply chain practices (fuzzy [49]), SSCM for Industry 4.0 [50], and implementing green supply chain management [51]. In addition, DEMATEL was used in a number of studies to analyze and evaluate barriers or challenges for sustainable development [52], remanufacturing (grey [53]), and green supply chain [54]. A few studies also analyzed drivers for sustainable consumption and production adoption applying grey DEMATEL [55] and drivers to ICT for sustainability initiatives in supply chains using a fuzzy one [56]. DEMATEL was used in other applications including the analysis of criteria and alternatives in sustainable supply chains (grey [57]), evaluation of influential indicators for adopting sustainable supply chains [58], analysis of causal relationships between practices and performance in green supply chains (fuzzy [59]), assessing performance in green supply chains considering economic, logistics, operational, organizational, and marketing aspects (fuzzy [23]), and selection of suppliers based on multiple criteria (fuzzy [60]).

AHP and Fuzzy AHP
AHP is one of the most widely applied MCDM methods in SSCM. Eleven articles applied AHP or Fuzzy AHP. Among these, six articles applied AHP, and the remaining five articles applied the fuzzy AHP method.
Six studies applied AHP to evaluate barriers to adopting sustainable consumption and production initiatives [61], to analyze criteria for improving effectiveness in green supply chain management implementation [62], to analyze challenges for industry 4.0 initiatives toward SSCM [63], to evaluate pressures to implement GSCM [64], to evaluate manufacturing practices for sustainability [65], and to analyze drivers for sustainable manufacturing processes [66]. Fuzzy AHP was also used to identify and analyze risks in green supply chains [67], analyze success factors for sustainable food supply chain management [68], evaluate indicators of SSCM [69], assess the supply chain performance based on sustainability criteria [70], and evaluate European countries for renewable energy sectors [71].

TOPSIS and Fuzzy TOPSIS
Earlier studies applied TOPSIS (two articles) or fuzzy TOPSIS (six articles) in the context of SSCM. The applications encompass the suppliers' evaluation and selection in sustainable and green supply chains based on multiple criteria. These criteria include applications of TOPSIS in selecting sustainable suppliers [72,73] and applications of fuzzy TOPSIS in evaluating green supplier performance [74,75], evaluating sustainable and green suppliers [24,76,77], and assessing areas for improvement in implementing green supply chain initiatives [78].
Researchers applied TOPSIS and Fuzzy TOPSIS to select suppliers and performance in sustainable or green supply chains based on identified multi-criteria.

BWM
The eight articles which have applied BWM in SSCM include an assessment of sustainability in green supply chains in an emerging economy [79], assessment of social sustainability in supply chains [33], evaluation of external forces for sustainable supply chains in the context of the oil and gas industries [80], analysis of enablers for social sustainability in an emerging economy [4], evaluation and prioritization of criteria for sustainable innovation [13], analysis of product-package alternatives in food supply chains [81], ranking sustainable suppliers [82], and analyzing barriers for sustainable supply chain innovation [83].

VIKOR and Fuzzy VIKOR
Five articles applied VIKOR or fuzzy VIKOR in SSCM. These articles include evaluating green supply chain management practices using fuzzy VIKOR [26], selecting devel-Sustainability 2021, 13, 7104 8 of 28 opment programs for green suppliers using fuzzy VIKOR theory [84], evaluating green environmental factors in reverse logistics using fuzzy VIKOR [85,86], and assessing green supply chain initiatives using a probabilistic linguistic VIKOR method [87].

Rough Set
The Rough set method has been applied in SSCM to select suppliers with sustainability [88], analyzing relationships between organizational attributes, supplier development programs, and performance in green supply chains [89], evaluating a selection, performance measurement, and program development tool in green supply chains [90], and measuring SSCM performances [91].

ELECTRE and Fuzzy ELECTRE
ELECTRE and fuzzy ELECTRE have been applied in SSCM to classify suppliers in the manufacturing industry using the ELETCRE TRI-nC method [92] and to evaluate supplier performance in green supply chains using the fuzzy ELECTRE method [93].

ANP
Two studies applied the ANP method in SSCM. The applications include selecting suppliers for managing sustainability [94] and selecting suppliers integrating the triplebottom-line aspect [95].

Rough Strength-Relation Analysis Method, RSAW, ISM, and PROMETHEE
One article applied the Rough strength-relation analysis method for analyzing risk factors in SSCM [96], the RSAW for sustainable supplier selection [97], the ISM for ranking of barriers in SSCM [98], and the PROMETHEE for analyzing alternatives of biomass [99].

Summary of Applications of Individual Methods
Researchers applied DEMATEL and Fuzzy/Grey DEMATEL, AHP, and BWM mostly for analyzing success factors, barriers and challenges, drivers, and enablers for different aspects of SSCM. Success factors are the important factors decision makers should consider to ensure success in different dimensions of SSCM. Barriers and challenges are the causes preventing the success of any dimension of SSCM. Drivers and enablers are the aspects driving toward the achievement of sustainable performance within any dimension of supply chain sustainability. The different MCDM methods applied to analyze and prioritize success factors, barriers and challenges, and drivers and enablers in SSCM are summarized in Tables 3-5, respectively.  Table 3. Cont.

Analyzed Success Factors in SSCM Reference Method
Technology development and process innovation, training, reverse logistics and waste minimization, ecological considerations in organizations' policies and missions, green design and purchasing, societal considerations, ethical and safe practices, and community welfare and development [47] Grey DEMATEL Climatic change, implementing green practice, governance and cooperation, technological innovation, and government regulation Proper use of irrigation, demographic and environmental conditions, risk analysis, government policies, and food packaging [68] Fuzzy AHP Table 4. Applications of MCDM methods to analyze barriers and challenges.

Analyzed Barriers and Challenges in SSCM Reference Method
Lack of sufficient governmental policies, poor infrastructure, low level of integration, skill shortage, and poor quality of raw materials Lack of channels to collect used products, imperfect legal system, consumption attitude, customer willingness to return the products, uncertainty in demand of remanufactured product, uncertainty in quality, and quantity and timing of returned products [53] Grey DEMATEL Lack of environmental regulation, lack of potential liability, high cost of disposal of hazardous materials, poor environmental performance, lack of information, lack of governmental support, high cost for renewable energy, lack of new technology, insufficient societal pressure, poor legislation, lack of adoption of green practices, health and safety issues, employment stability, less profit in remanufacturing, lack of adequate training, and lack of management support [54]. DEMATEL Lack of support from management, lack of innovative methods, lack of technology developments, communication gap, lack of rewards and encouragement programs, lack of governmental regulations, lack of promotion of ethical and safe practices, reluctance of consumers toward sustainable development practices, lack of promotion of sustainable products, and lack of knowledge among stakeholders [61] AHP Low understanding of industry 4.0 implications, poor research and development of industry 4.0 adoption, legal issues, low management support and dedication, lack of global standards and data-sharing protocols, security issues, lack of governmental support and policies, and financial constraints [63] AHP Technological, regulatory, social, cultural, organizational, market, and networking barriers [83] BWM Table 5. Applications of MCDM methods for analyzing drivers and enablers.

Analyzed Drivers and Enablers in SSCM Reference Method
Top management role and support, government support systems and subsidies, information systems network design, socio-environmental impacts of the products, culture related factors, approach to ICT to adopt sustainability, understanding the nature of sustainability, security and support services, and human expertise [56] Fuzzy DEMATEL Management support, dedication and involvement, educating suppliers and vendors, understanding the customer requirements about sustainability, governmental policies and regulations, information flow and sharing among supply chain members, competency and skill of workforce, integration of social, environmental, and economic advantages, and understanding the importance of sustainability [55] Grey DEMATEL Researchers applied TOPSIS, Fuzzy TOPSIS, VIKOR, Rough Set, and ANP to analyze and evaluate suppliers and practices in sustainable or green supply chains based on sustainable criteria. These studies are summarized in Table 6.

Applications of Integrated MCDM Methods
A total of 47 articles applied integrated MCDM methods in SSCM. Among these, AHP or Fuzzy AHP were most widely integrated with other methods such as DEMATEL, ELECTRE, ISM, TOPSIS, VIKOR, and SOWIA, followed by TOPSIS or Fuzzy TOPSIS with FPP, Rough set, CRITIC, and VIKOR. Researchers have applied more integrated MCDM methods in recent years, making a significant methodological contribution; this section summarizes such studies.
ANP is mostly integrated with quality function deployment (QFD) to analyze supplier selection and environmental sustainability, and for designing sustainable supply chains [113][114][115][116]. Other integrations of ANP with VIKOR [117] and grey rational analysis (GRA) [118] were applied in green/sustainable supplier evaluation.
BWM or fuzzy BWM is mostly integrated with VIKOR or fuzzy VIKOR for evaluating transportation service providers and outsourcing partners based on sustainable criteria [119,120]. Other applications of integrated BWM or fuzzy BWM include evaluating dimensions of human resources in green supply chains using BWM-DEMATEL [121], selecting sustainable suppliers in manufacturing supply chains by integrating BWM and an alternative queuing method (AQM) [122] and selecting sustainable suppliers using integrated BWM and combined compromise solution [123].
TOPSIS or Fuzzy TOPSIS is mostly integrated with VIKOR or fuzzy VIKOR, fuzzy preference programming (FPP), Rough set, and criteria importance through intercriteria correlation (CRITIC). TOPSIS-VIKOR (and their fuzziness) integrated methods [124,125] were applied in selecting third-party reverse logistics service providers and classifying rural areas based on social sustainability criteria. TOPSIS-VIKOR-GRA (integrating three methods) was applied in analyzing locations for remanufacturing plants based on multiple criteria [126]. Other applications of integrated TOPSIS or fuzzy TOPSIS include evaluating supply chain practices by integrating TOPSIS and Rough set [127], analyzing risk factors in SSCM using TOPSIS-CRITIC [128], and selecting sustainable suppliers using TOPSIS-FPP [129].
Other integrated methods, such as ELECTRE with VIKOR, were applied in environmental performance evaluation [130]; DEMATEL with MABAC was applied in sustainable freight transport systems [131]; RSAW with MABAC applied in sustainable supplier selection [132]; factor relationship (FARE) with MABAC for selecting third-party logistics provider [133]; step-wise weight assessment ratio analysis (SWARA) and fuzzy complex proportional assessment of alternatives (COPRAS) were used for analyzing risks and solutions in sustainable manufacturing supply chains [134]; and fuzzy entropy and fuzzy multi-attribute utility were applied for sustainable performance measure in supply chain [135].
In summary, most of the integrated MCDM methods in SSCM were used for evaluating or analyzing suppliers, service providers, barriers, enablers, success factors, and evaluating performance. A summary of different integrated MCDM methods applied in SSCM is presented in Table 7.

Bibliometric Analysis on MCDM Methods Applied to SSCM
This section presents a bibliometric analysis of MCDM methods applied to SSCM. From the finalized literature search (see Section 2), we can note a lack of reviews, with only five studies (about 4.5%) reviewing particular topics such as: (i) green supplier evaluation and selection [22]; (ii) modelling approaches in SSCM [146]; (iii) MCDM approaches in green supply chains [45]; (iv) hybrid MCDM for general sustainability [147]; and (v) sustainable supplier selection [148]. In total, 106 contributing articles (about 95.5%) applied MCDM methods to better understand SSCM issues; this means, Figure 2 presents the keyword network obtained from the keywords used in each of the contributing articles. It is evident that supply chain management, decision-making, sustainable development, sustainability, and green supply chains, environmental management, and sustainable supply chains are the top keywords. Figures 3 and 4 present the citation networks of selected contributing papers based on source journals and authors, respectively. The Journal of Cleaner Production and International Journal of Production Economics are two leading cited journals. Govindan, K., and Mangla, S.K. are two leading cited authors.

Bibliometric Analysis on MCDM Methods Applied to SSCM
This section presents a bibliometric analysis of MCDM methods applied to S From the finalized literature search (see Section 2), we can note a lack of reviews only five studies (about 4.5%) reviewing particular topics such as: (i) green supplier uation and selection [22]; (ii) modelling approaches in SSCM [146]; (iii) MCDM proaches in green supply chains [45]; (iv) hybrid MCDM for general sustainability and (v) sustainable supplier selection [148]. In total, 106 contributing articles (about 9 applied MCDM methods to better understand SSCM issues; this means, Figure 2 pr the keyword network obtained from the keywords used in each of the contributin cles. It is evident that supply chain management, decision-making, sustainable dev ment, sustainability, and green supply chains, environmental management, and su able supply chains are the top keywords. Figures 3 and 4 present the citation netwo selected contributing papers based on source journals and authors, respectively. The nal of Cleaner Production and International Journal of Production Economics ar leading cited journals. Govindan, K., and Mangla, S.K. are two leading cited author        The number of contributing articles published from 2010 to 2020 is shown in Figure 5.  Figure 6. Basel, Switzerland), and the International Journal of Production Research (Publisher: Taylor and Francis, Oxfordshire, UK). The number of articles published in each of these journals is presented in Figure 6.    Basel, Switzerland), and the International Journal of Production Research (Publisher: Taylor and Francis, Oxfordshire, UK). The number of articles published in each of these journals is presented in Figure 6.     Figure 7 presents the authors who published the most articles in the area of MCDM for SSCM. Both Govindan, K. and Mangla, S.K. are at the top of the list, with 10 published articles, followed by Sarkis, J., Bai, C., and Luthra, S. with 9, 7, and 6 articles, respectively.
The affiliation by authors' institutions is also presented in Figure 8. Syddansk Universitet (University of Southern Denmark) is at the top of the list with 12 articles, followed by the Indian Institute of Technology Roorkee (India) and Dalian University of Technology (China). The affiliated countries which published most articles are presented in Figure 9. India is at the top of the list with 27, followed by China, the United States, and Denmark with 19, 15, and 12 articles, respectively. The affiliation by authors' institutions is also presented in Figure 8. Syddansk Universitet (University of Southern Denmark) is at the top of the list with 12 articles, followed by the Indian Institute of Technology Roorkee (India) and Dalian University of Technology (China). The affiliated countries which published most articles are presented in Figure  9. India is at the top of the list with 27, followed by China, the United States, and Denmark with 19, 15, and 12 articles, respectively. The affiliation by authors' institutions is also presented in Figure 8. Syddansk Universitet (University of Southern Denmark) is at the top of the list with 12 articles, followed by the Indian Institute of Technology Roorkee (India) and Dalian University of Technology (China). The affiliated countries which published most articles are presented in Figure  9. India is at the top of the list with 27, followed by China, the United States, and Denmark with 19, 15, and 12 articles, respectively.   Fifty-nine articles (about 56%) used individual MCDM methods for problem analysis, while the remaining 47 articles (about 44%) used integrated (two or more methods) methods. Figure 10 shows that integrated or hybrid MCDM methods were mostly applied in SSCM. Within individual MCDM methods: DEMATEL, AHP, and TOPSIS are the top three methods applied in SSCM literature.

Summary of the Review and Research Gaps
The following key points present a summary of the extensive literature review and its findings. Firstly, Section 3 described the application of different MCDM methods in SSCM, Fifty-nine articles (about 56%) used individual MCDM methods for problem analysis, while the remaining 47 articles (about 44%) used integrated (two or more methods) methods. Figure 10 shows that integrated or hybrid MCDM methods were mostly applied in SSCM. Within individual MCDM methods: DEMATEL, AHP, and TOPSIS are the top three methods applied in SSCM literature. Fifty-nine articles (about 56%) used individual MCDM methods for problem analysis, while the remaining 47 articles (about 44%) used integrated (two or more methods) methods. Figure 10 shows that integrated or hybrid MCDM methods were mostly applied in SSCM. Within individual MCDM methods: DEMATEL, AHP, and TOPSIS are the top three methods applied in SSCM literature.

Summary of the Review and Research Gaps
The following key points present a summary of the extensive literature review and its findings. Firstly, Section 3 described the application of different MCDM methods in SSCM, showing most studies applied an individual or two integrated MCDM methods, with a

Summary of the Review and Research Gaps
The following key points present a summary of the extensive literature review and its findings. Firstly, Section 3 described the application of different MCDM methods in SSCM, showing most studies applied an individual or two integrated MCDM methods, with a lack of studies integrating three or more MCDM methods for analyzing different dimensions of sustainability in supply chain management. Only a few studies integrated three methods (see Table 7); for example, AHP integrated with SOWIA and TOPSIS [112], TOPSIS integrated with GRA and VIKOR [126], and ANP integrated with DEMATEL and TOPSIS [140].
Secondly, in SSCM, most of the studies contributed to advancing the methodology by integrating different MCDM methods. MCDM methods were integrated within AHP, ANP, BWM, TOPSIS, ELECTRE, DEMATEL, ISM, VIKOR, SOWIA, GRA, QFD, Rough set, CRTIC, FPP, MABAAC, and AQM along with their fuzzy sets (see Table 7). It was observed that both single and integrated MCDM methods were applied, but there is a lack of studies which integrate MCDM methods with other operations research methods to improve the accuracy in decision making. For example, the integration of MCDM methods with mathematical modelling and optimization could help to obtain optimal decisions in SSCM.
Thirdly, most of the research considered social sustainability, environmental sustainability, and/or triple bottom line in the scope of their studies. Tables 6 and 7 presented the areas of application of individual and integrated MCDM methods. The application areas of MCDM methods were limited to different aspects of sustainability. Most of the studies integrated MCDM methods in models for green supply chains (see Table 7). Most of the studies on individual MCDM methods applied them in evaluating factors, barriers, challenges, drivers, enablers, and suppliers in different SSCM areas (see Tables 3-6).
Fourthly, most SSCM researchers limit their studies to analyzing factors, barriers, challenges, drivers, enablers, and suppliers (see Section 3). There are fewer studies that integrate sustainable strategies and performance with barriers/challenges and drivers/enablers using integrated MCDM methods.
Fifthly, from Section 3, most of the studies explained their results using individual or integrated MCDM methods. There are limited studies which compared the results between different MCDM methods. For example, there is only one study comparing fuzzy DEA and fuzzy TOPIS in sustainable supplier selection [76].
Finally, earlier literature largely overlooked small and medium enterprises (SMEs) and the emerging economy as study contexts for MCDM applied in SSCM. A small number of articles considered SMEs as their context, analyzing drivers for integrated lean-green manufacturing [149], innovation ability for supplier selection [150], determinants for cloud computing adoption [151], and entrepreneurship policies [152].

Conclusions and Future Research Directions
This paper presented a literature review on MCDM methods applied in different dimensions of SSCM. Although there have been many papers published in the area of SSCM, there are very few review papers published. A systematic literature analysis and review to identify different domains of supply chain sustainability and applications of different MCDM methods was lacking. Contributing to the academic discussion on this topic, this is the first effort to analyze both individual and integrated MCDM methods applied in different areas of supply chain sustainability.
Findings revealed the following important aspects of previous studies.
i. Most of the research applied either individual or integrated methods of two MCDM techniques. All of the integrated MCDM methods applied were carried out in recent years, i.e., after 2015. ii. Since 2015, environmental and social sustainability have been garnering the attention of researchers (see Tables 1 and 2). In recent years, several MCDM methods, such as ISM-TOPSIS [31], grey BWM-grey TODIM [18], and BWM [4,33] were applied in social sustainability while DEMATEL [25], fuzzy DEMATEL [23], fuzzy VIKOR [26], and TOPSIS [24] were applied in environmental sustainability areas. iii. Most of the individual methods analyzed barriers, drivers, enablers, challenges, success factors, forces, and criteria in SSCM and green supply chains (see Tables 3-6). iv. Integrated MCDM methods were applied in evaluating and analyzing sustainable suppliers and different alternatives in SSCM (see Table 7). v.
There are very few studies which integrated three or more MCDM methods and applied them in SSCM [112,126].
In conclusion, we would like to acknowledge some study limitations and propose some future research directions. At present, the study has been limited to three databases: Scopus (primary), Web of Science, and Google Scholar. Therefore, potential additional sources of information and knowledge such as conference proceedings or books not indexed in the three databases have not been included. Further, the research was limited to documents published in English exclusively, therefore potentially excluding local contributions redacted in other languages. Additionally, as there is limited research on the integration of three or more MCDM methods, it would be interesting to develop new integrated methodologies with three or more MCDM methods for decision-making in SSCM. Moreover, the fuzzy BWM method is a newly developed tool that can optimize a decision based on multiple criteria [153]. Currently, few studies apply BWM and fuzzy BWM in SSCM [4,13,79,[119][120][121][122]. In the future, BWM and Fuzzy BWM could be applied widely in SSCM. In addition, the literature is lacking contributions able to integrate barriers and challenges with strategies and analyzes using an integrated MCDM method. In this regard, further research could explore the following research question: 'How can sustainable strategies be analyzed along with barriers and challenges to extend the current studies using MCDM methods?' In the current literature, there is no study that correlated success factors, enablers, and drivers with sustainability performance analyzed using integrated MCDM methods. Hence, future research could address how barriers, challenges, drivers, enablers, and/or success factors can be linked with strategies and performance in one study, so as to develop a unique research framework using MCDM methods. In addition, it would be significantly new to the literature to integrate different aspects of SSCM with other supply chain areas such as risk management, lean supply chains, quality management, and supply chain network design. For example, the recent coronavirus (COVID-19) pandemic has impacted supply chains significantly [154]. It would be worth investigating the impacts of large-scale disruptions such as the COVID-19 pandemic in supply chain sustainability. Different MCDM methods can be applied to investigate different aspects of impacts on sustainable practices.
Our review pointed out that though service supply is an important area, research on applications of MCDM methods in sustainable service supply chains is very limited [130]. Different dimensions of sustainability in service supply chain management could be explored using MCDM methods.
In the future, another research direction could be integrating MCDM methods with operations research techniques, such as mathematical modelling and optimization, to make decisions more accurate. For example, integration of MCDM methods with mathematical modelling and optimization techniques could help to take optimal and accurate decision making in SSCM. Moreover, as there is only one study comparing the results of MCDM methods [76], it is necessary to develop benchmark problems and compare the results obtained from different MCDM methods used in SSCM. It would also be interesting to compare the sustainability results between supply chains from developed and emerging economies.
Finally, as SMEs and emerging economies play significant roles in local and global supply chains [155,156], it is important to analyze the sustainability of supply chains of SMEs and emerging economies. Usually, SMEs and emerging economies have fewer resources to deal with supply chain sustainability challenges. Hence, it is essential to analyze their supply chain sustainability and develop policies for improving sustainability practices. However, in the literature of MCDM methods and SSCM, SMEs and emerging economies were widely ignored. It would be interesting to develop decision-making tools using MCDM methods for analyzing different aspects of supply chain sustainability in the context of SMEs and emerging economies.