Multi-Criteria Analysis Techniques to Assist Decision-Making in Renewable Energy Supply Chains: A Review

Energy consumption has witnessed a substantial rise globally, especially in industrialized nations, primarily attributed to population growth, technological advancements, and improved living standards in recent times [1]. The significant surge in energy requirements, heightened worries regarding energy security, and the environmental repercussions associated with fossil fuel consumption have collectively driven a shift toward renewable energy sources [2]. The concept of renewable energy has found applications in various sectors. Renewable energy is derived from sustainable sources, including the sun, wind, waves, tides, marine, and geothermal energy, is becoming increasingly prevalent. Numerous countries worldwide have already integrated renewable energy into their energy supplies, and it is anticipated that national renewable energy markets will continue to grow significantly [3].

Supply chain management (SCM) is a component that orchestrates the movement of information, products, and finances from their source to various consumption areas [4]. The performance of the SC is undeniably a paramount factor for companies aiming not only to survive but also to thrive in today's fiercely competitive [5]. SCM in the context of renewable energy necessitates consideration of unpredictable factors and the performance of renewable energy systems, encompassing aspects like storage, distribution, conversion efficiency, and secondary applications. Renewable energy supply chains (RESCs) encompass familiar components of conventional SCM, involving the flow of physical, information, and financial resources. Through these practices, they play a vital role in promoting eco-friendly production processes, logistics, and products to enhance overall efficiency [6].

RESCs are now a prominent subject in academic discussions within the research domain of SCs. Factors such as economic shifts, political changes, and heightened consumer consciousness have led to the rapid expansion of RESCs in recent years [7]. This increase is evident in the growing number of publications in the field of RESCs in recent years (refer to more information in Figure 1). Table 1 shows the most recent reviews in RESCs.

The structure of this paper is as follows: Section 2 outlines the research methodologies employed, Section 3 presents the key findings derived from the literature analysis, and finally, in Section 4, the principal conclusions are summarized.

The data utilized in this paper was sourced from the Scopus database covering the timeframe from 2018 to 2023. The analysis delves into a set of publications that applied quantitative multi-criteria analysis (MCA) techniques.

In the realm of literature, there exist various methods for categorizing MCA approaches. One widely recognized form of MCA is multi-criteria decision analysis (MCDA), which is sometimes referred to as multi-attribute decision analysis (MADA). It's important to note that the terms 'MCDA' and 'MCDM' (multi-criteria decision making) are frequently used interchangeably to refer to MCA. Since achieving optimal values for all objectives simultaneously is unattainable, MCDM entails addressing a selection problem [14].

Furthermore, these approaches are categorized into two main groups: multi-objective decision making (MODM) and multi-attribute decision making (MADM). The principal distinction between these two sets of approaches lies in the number of possible alternatives under evaluation. In MODM, an infinite number of alternatives are defined by constraints on a vector of decision variables, resulting in an optimized alternative. Conversely, MADM approaches, commonly applied in environmental decision-making scenarios, involve a finite number of explicitly defined alternatives, which are then assessed against a set of principles or criteria. These assessments often involve qualitative and/or quantitative attributes [15], [16].

In this study, the title, abstract, and keyword search was conducted in the Scopus database. The choice of Scopus was made due to its status as the largest abstract and citation database for peer-reviewed scientific literature. Notably, it encompasses journals that are also part of the Web of Science and provides the capability for citation analysis. Similar to other databases, Scopus offers various features and functions that simplify bibliometric analysis.

The major keywords employed in this study, namely 'supply chain,' 'renewable energy,' and 'MCA,' were thoughtfully selected to align with the main research goal: to investigate the connection between SC decision-making techniques and the adoption of renewable energies. Keywords related to various forms of renewable energy were chosen after a thorough literature analysis. These included 'renewable energy,' its synonyms, and alternative designations found in the literature. These keywords were used inclusively (linked by 'OR') to ensure comprehensive retrieval of articles encompassing all types of renewable energies. At the same time, we considered the keywords 'supply chain' and those associated with 'MCA,' stipulating their presence in the search fields (title, abstract, keywords) of each publication. This approach aimed to identify publications that address both SC-supported MCA techniques and specific types of renewable energy for analysis. Table 2 illustrates the use of various keywords for data collection.

Additionally, we limited our search to articles written in English, the predominant language of scientific communication, and to articles published in journals. Other forms of publication, such as books, book chapters, and conference proceedings, often do not adhere to the same standardized peer-review process and are therefore considered less credible sources for academic dissemination. Consequently, these sources were not included in our search criteria.

The correlation between SCM and the adoption of renewable energies has garnered growing attention in scholarly literature, as depicted in Figure 1. There is a discernible upward trajectory in interest, particularly noticeable since 2015. Consequently, this not only signifies a rising area of interest but also a notably recent one. In this section, we outline the findings of our content analysis regarding the utilization of MADM techniques within the context of RESCs. According to the keywords mentioned in Table 2, the search results yield a collection of 153 studies on Scopus. Figure 2 illustrates the trends of these articles, and in terms of the number of publications, it is evident that a significant growth has been observed since 2018.

Table 3 presents the most pertinent publications, which encompass a type of renewable energy production and employ MCA-supported SCM, all of which were published after 2018. Sustainability Management refers to strategic business endeavors aimed at concurrently mitigating sustainability risks in the environmental, economic, and social dimensions [17]. In the realm of the renewable energy SC, scholars, managers, and vested parties can effectively employ insights derived from reviewed literature to align with their respective decision-making domains.

This paper utilizes keyword analysis as a bibliometric indicator. It employs the science mapping approach of bibliometric analysis, utilizing the visualization software VOSviewer to illustrate the subject structure [54]. VOSviewer Software was selected due to its robust graphical user interface and mapping visualization capabilities [55]. The data underwent standardization and formatting to make it compatible with VOSviewer. This process involved converting the database into plain text and thoroughly assessing all articles to ensure their relevance and accuracy. Ultimately, 36 records met the filtering criteria outlined above (refer to Table 3).

Keywords serve as fundamental components within a specific field of study, offering insights into the knowledge structure and research trends. Through keyword co-occurrence analysis, a network can be constructed where each node represents a keyword, and the links between nodes signify their co-occurrence. In this analysis, keywords initially included in the search were excluded to ensure a comprehensive examination of literature features, with a minimum threshold of 2 keywords required for inclusion in Figure 3. The construction of the keyword co-occurrence network involves assigning each keyword as a node and establishing edges based on keyword co-occurrences. When two keywords have a co-occurrence relationship, it indicates a connection between the respective nodes.

The keywords depicted in Figure 3 were generated from a prior analysis conducted using VOSviewer, which considered density, degree, centralization, and inclusiveness as representative indicators of keyword connections. Density reflects the ratio of formed relationships to total possible relationships, while centralization measures the concentration of relationships on a few individuals. Inclusiveness indicates the number of interconnected nodes in the network. Centrality analysis identifies key nodes and determines which keywords hold a central position within the network. Keywords positioned closely to one another are considered influential, with closed centrality being a crucial concept in co-occurrence word analysis.

The methodology applied in this study revealed the presence of five primary clusters. The blue sub-network addresses supply chain issues, encompassing environmental concerns, sustainable development, environmental impact assessment, supplier selection and life cycle considerations. Notably, the blue cluster exhibits strong connectivity with the green cluster, indicating numerous links between them. The green cluster predominantly revolves around renewable energy resources, biomass and biofuels production issues, and energy supply chain management. Keywords related to programming, such as mixed integer linear programming, integer programming, and multi-objective optimization, form the purple cluster. The yellow cluster encompasses multi-criteria decision-making topics, including considerations for renewable energy and climate change. Lastly, the red cluster focuses on decision-making techniques, encompassing energy policy, alternative energy sources, and prioritization, with the Analytic Hierarchy Process (AHP) and Data Envelopment Analysis (DEA) methods emerging as pivotal within this cluster. Overall, the key themes in the analyzed articles center on sustainability, management, and optimization, particularly in the context of energy policy decision-making and the performance of renewable energy supply chains, with a notable emphasis on biomass-based energy.

The most prominent keywords identified in the literature include supply chains, sustainable development, multi-objective optimization, integer programming, biomass, biofuels, renewable energy, decision-making, DEA, and energy policy. This analysis underscores that the literature concerning renewable energies and supply chain performance is heavily influenced by the sustainability of biomass as a renewable energy source, with a strong emphasis on decision-making techniques, particularly DEA and AHP, alongside planning processes for supply chains, such as integer programming and multi-objective optimization. For further investigation, Figure 4 can be utilized to illustrate developments over time, while Figure 5 offers a quick overview of the main areas in a bibliometric network.

This overview presents and discusses several pertinent issues related to the application of MCA in RESCs. Bioenergy stands out as the most commonly utilized renewable energy source in these SCs. In terms of SC analysis, some studies have focused on the entire SC, while others have narrowed their focus to specific aspects of the bioenergy SC, such as distribution logistics.

Various literature reviews have concentrated on modelling techniques within sustainable SCM, highlighting how operational research methods and MCA can effectively reduce process costs. Within the realm of renewable energy analysis, the AHP emerges as the most frequently employed MCA methods. This study aims to identify the primary contributions and findings in current and future research using the MCA approach. Predominantly, the challenges in this field revolve around issues related to site selection and supplier selection. The central discovery of this study underscores the need for further research into the application of MCA approaches in the RESC, with a particular focus on thoroughly investigating the complete and comprehensive RESC. Furthermore, to align with the goals of sustainable development, social aspects should be integrated into SC analysis when sustainability is a primary target.