Supply chains enable and represent the flow of goods, services, and information without which it is impossible to imagine the normal functioning of the market. They include and are part of the primary, secondary, and tertiary sectors and cover the entire process and flow of production [8]. Supply chains create added value in the business process, which is why it is necessary to manage the supply chain in order to create added value constantly. The supply chain management process includes an “integrated process that includes planning and managing all stakeholder selection activities, procurement of materials, transformation of materials into the final product, as well as related logistics activities within the entire chain” [9] (p. 187). In order to examine the efficiency of the supply chain, the “Game-theoretical Design Technique” approach was tested. The design approach allows easier access to business decisions within the supply chain when the supply chain is exposed to some uncertain parameters and ensures agile, cooperative, and resource-efficient design of multi-stage supply chains [10].

The supply chain is evolving rapidly, and while this development follows the technology development, it also intends to ensure sustainable development. Many authors equate sustainable development with economic and social growth. Sustainable development is the aspiration of society to achieve sustainable economic growth to the extent that it will meet the needs of present and future generations. It is necessary to preserve production capacity for the long term while achieving social goals such as increasing real income per capita, improving hygiene and nutrition, educational achievement, access to resources, equitable distribution of wealth, and increasing freedom [11,12,13,14]. The key goal of any organization is to ensure sustainability and create sustainable business models. Such an approach and business model rely primarily on the selection of sustainable and green suppliers that will ensure the path to sustainable development. Supply chains and their management and governance are a complex system that strives to minimize costs and maximize the level of service, and its focus today should primarily be on the selection of sustainable (green) suppliers.

Sustainable (green) suppliers improve and assist sustainable business models in the field of supply chain management. With an emphasis on sustainability and environmental care, the selection of green suppliers should be a central component and goal in supply chain management [15]. Suppliers and supplier relationship management create several benefits for the company—from creating long-term and loyal partners to creating greater visibility through better communication with them. Choosing a reliable business partner is the task of the management of companies that want to achieve long-term competitiveness and sustainability in the market. Sustainable (green) supplier selection and assessment are the most significant and complex challenges for supply chain managers [16,17]. Supplier selection is “intrinsically related to the Multi-Criteria Decision Making (MCDM) problem” [18]. Moreover, the process of selecting a supplier represents a “key competence in the sourcing function” [19]. Today, everything is intensified through environmental (green) factors that have, i.e., should have an impact on the selection of reliable and green suppliers. The question of how to identify and select sustainable suppliers can be answered through a system of weights that determine environmental factors as important decision factors when selecting suppliers by changing the DEA methodology [20]. One of the problems that arises in the business environment today is the selection of suppliers, primarily the selection of sustainable suppliers. Quality and proper selection of suppliers enable and facilitate the cross-efficiency fuzzy DEA technique. This method evaluates the efficiency of the supply chain and all its members (primarily in this case—suppliers) with the correct ranking which then allows the correct selection of suppliers and later reflects on the overall efficiency, costs, and delivery time within the supply chain [21].

Choosing sustainable suppliers means choosing business partners with the most beneficial monetary value on the one hand and the least harmful impact on society and the environment on the other [22]. Assessing the sustainability and performance of the supplier and the supply chain itself requires a selection of optimal tools and methods for the selection process, where DEA is an important tool for measuring the performance of sustainable suppliers and sustainable supply chains [23]. That the DEA methodology and the evaluation index method are an objective and quantitative tool for evaluating and selecting sustainable (green) suppliers is a fact confirmed by numerous researchers through their empirical research [23,24,25,26,27,28,29,30].

However, considering the importance of suppliers in the strategy framework of supply chains, it is rather surprising that “the sourcing function has not been subject to more focused research on the development of adequate decision support tools” [19].

The DEA is a linear mathematical programming technique that is used for the evaluation of the performance (i.e., the relative efficiencies) of a group of complex entities referred to as Decision-Making Units (DMUs) [31]. DEA is one of the most widely applied non-parametric methodologies ever since its introduction in 1978 by Charnes, Cooper, and Rhodes (1978) [5,6] that has grown into a powerful mathematical and linear programming technique. Namely, according to [7], there are five areas in which DEA has been most applied, and these are Agriculture, Banking, Supply Chain, Transportation, and Public Policy. Research conducted in the field of supply chain related to planning and resource management in intermodal terminals proves that the application of the DEA model allows planning decisions under conflicting requirements (the DEA method provides data on the number, capacity, and allocation of resources to address increasing flows) [32]. Moreover, the positive implication and real case study regarding the usage of the DEA method in the supply chain is used “to determine the efficiency of the rescheduled timetable (in terms of reduction of delays and maximization of robustness) and to rank alternatives according to their efficiency values [33] (p. 256).

DEA is a non-statistical and non-parametric approach that “makes no assumptions regarding the distribution of inefficiencies or the functional form of the production function (although it does impose some technical restrictions such as monotonicity and convexity)” [34], which represents one of its main advantages over parametric methodologies. It is a “data-oriented“ method that converts multiple inputs to multiple outputs when evaluating peer units—DMUs [35]. DEA focuses on the extreme observations, which is its main distinction from the parametric methodologies that “focus on average tendencies and deviations from it”. Moreover, DEA can employ multiple inputs and outputs, while parametric methodologies can only employ one output, which represents one of their biggest limitations [36]. This makes DEA “an excellent data-oriented efficiency analysis method” when using multiple inputs and outputs and “a useful performance evaluation and decision-making tool” [37].

DEA identifies the relative efficient DMUs in the observed sample that shape the efficiency frontier that measures the inefficiency of inputs or outputs of the other DMUs in the sample by comparing them and benchmarking with the relative efficient DMUs. Thus, DEA is also an econometric frontier method. The DMUs that lie on the frontier have the best relative efficiency, whereas the ones that are inefficient lie below the efficiency frontier [38]. Moreover, the results from the DEA vary from 0 to 1 (or 0 to 100%), with 1 being relatively efficient and a result below 1 being relatively inefficient. Thus, this approach enables a simple comparison of the DMUs in the sample [39,40].

Two basic DEA models are named in honor of their founders (CCR—Charnes, Cooper, and Rhodes and BCC—Banker, Charnes, and Cooper [5,6]). The CCR model employs a constant return to scale (CRS) assumption, i.e., “the output variables increase proportionally with input variables” [41], whereas the BCC DEA model employs a variable returns-to-scale (VRS) assumption, assuming that the proportional change in inputs does not necessarily lead to a proportional change in the outputs. The CCR model is graphically represented as a straight line, whereas the BCC model is represented by a convex hull [42].

However, it should be kept in mind that DEA is a methodology that explores the relative (and not absolute) efficiency within the analyzed sample of peer decision-making units [43]. The limitations and downsides of DEA are often scholarly researched, but most of the authors agree that “its advantages outweigh its limitations” and DEA should be considered “a significant diagnostic tool” [31].