A sustainability and resilience-oriented multi-criteria assessment of agri-food systems across representative European countries

Abstract

Agri-food systems face increasing pressures from climate change, resource scarcity, geopolitical disruptions, and socio-economic inequalities, making the integrated assessment of sustainability and resilience essential. However, these dimensions are often examined separately in the literature. This study proposes a multi-criteria decision-making (MCDM) framework to comparatively evaluate the sustainability and resilience performance of agri-food systems in six representative European countries: Türkiye, Greece, Italy, France, Germany, and the Netherlands. Seven criteria aligned with the Sustainable Development Goals—import dependency, household food waste, food-system greenhouse gas emissions, agricultural water withdrawal, food insecurity, climate resilience, and logistics capacity—are analyzed using data from internationally recognized sources. Criterion weights are determined using the Entropy and CRITIC methods, while country rankings are obtained through TOPSIS, VIKOR, and COPRAS. Spearman rank correlation is employed to test inter-method consistency. The results show strong agreement among the ranking methods, confirming the robustness of the framework. France consistently ranks highest due to a balanced performance across cost and benefit criteria, whereas Türkiye ranks last because of cumulative disadvantages in food insecurity, emissions, water use, and import dependency. Variations in the rankings of Germany and the Netherlands reveal structural trade-offs between environmental pressures and resilience-related capacities. Overall, the study demonstrates that integrated, multi-method MCDM approaches provide more nuanced and policy-relevant insights than single-dimensional assessments.

1 Introduction

Agri-food supply chains are highly complex, multi-actor systems located at the intersection of agriculture and industry, involving a wide range of stakeholders including farmers, processors, distributors, retailers, and consumers (Tomasiello and Alijani, 2021). These supply chains encompass a wide range of activities, including pre-production, production, postharvest handling, storage, processing, packaging, marketing, and distribution, forming a single interconnected network (Verma et al., 2021). The heterogeneous structure, ranging from large-scale agricultural enterprises to small family farms, increases operational variability across a wide spectrum, from crop patterns and technology use to labor conditions and supply relationships, making standardization and consistent performance management challenging (Haileslassie et al., 2016). Furthermore, economic efficiency, environmental impact, and social well-being must be considered simultaneously in agri-food systems (Wollni et al., 2025). Therefore, the factors determining sustainability and resilience are numerous and have strong interrelationships and causal links (Lamine, 2015; Popescu et al., 2023; Muyulema-Allaica et al., 2025). In this respect, strengthening agri-food systems is directly related to the United Nations Sustainable Development Goals (SDGs), primarily SDG 2, SDG 6, SDG 12, and SDG 13 (Farooq, 2023). For this very reason, one-dimensional or linear analysis approaches are insufficient to explain this complexity, thereby necessitating the simultaneous, consistent, and comparable consideration of different dimensions.

In recent years, environmental pressures such as climate change, global warming, widespread droughts, and biodiversity loss, combined with demand pressures triggered by geopolitical uncertainties, conflicts, and increasing population, have made agri-food supply chains more fragile. The COVID-19 pandemic, along with crises such as Russia-Ukraine and Israel-Palestine, has clearly demonstrated that shocks in food systems can be transmitted from local to global levels, simultaneously affecting not only production processes but also supply, distribution, and consumption patterns (Callens et al., 2022; Popescu et al., 2023). Under these conditions, evaluating the performance of agri-food systems cannot be reduced solely to “sustainability” indicators; “resilience,” which expresses the system’s capacity to absorb, withstand, and recover from shocks, must also be placed at the center of the analysis.

Structural features that increase the vulnerability of agri-food supply chains, especially during times of crisis, are noteworthy: high perishability of products, demand uncertainty, distribution constraints, logistical bottlenecks, and labor-related disruptions weaken the continuity of the system. During the pandemic, movement restrictions, border closures, and sudden changes in consumption behavior made agri-food supply chains among the most affected systems, and the effects became more visible, especially in vulnerable population groups (Zavala-Alcívar et al., 2020). This vulnerability is further highlighted by the ongoing problems in accessing healthy and regular nutrition worldwide; when considered together with the dual health risks such as malnutrition, obesity, and the spread of diet-related noncommunicable diseases, the need for transformation of agri-food systems becomes even more apparent (Callens et al., 2022).

A critical point here is that shocks often occur simultaneously or sequentially, rather than individually. Under “multiple shocks,” compound effects occur in supply chains, increasing food and nutrition insecurity (Liu et al., 2025). However, analyzing sustainability and resilience together under multiple shock conditions is challenging due to the mutually reinforcing effects of shocks, spatial dispersion dynamics, and trade-offs between criteria. This situation renders linear assessment approaches that focus only on individual risks or use limited sets of criteria insufficient; it increases the need for system-based, interdisciplinary frameworks that consider human-nature interactions and supply chain layers together.

Furthermore, agri-food supply chains are not only affected by climate change but also contribute to the climate crisis. Greenhouse gas emissions generated during the production, distribution, and food loss/waste phases, combined with energy-intensive transportation and refrigeration processes, increase environmental pressure and are directly linked to SDG 13. However, many indices and assessment frameworks developed in the current literature mostly focus on specific stages of the chain or limited product groups; they are limited in evaluating sustainability and resilience simultaneously, comparably, and holistically (Chen et al., 2025). Similarly, although the concepts of sustainability and resilience are increasingly mentioned together in the literature, in practice, most studies treat these two dimensions separately; resilience strategies can sometimes have undesirable effects on environmental or social performance. Therefore, integrated approaches that consider resilience as a complementary capacity of sustainability are gaining importance in order to secure long-term sustainable performance (Paredes-Rodríguez et al., 2024).

In this context, addressing sustainability and resilience together in agro-food systems is not only a conceptual necessity but also a policy and governance imperative. Increased global interdependence and the cross-border nature of value chains highlight the need for strategic investments in infrastructure and digital technologies that strengthen transparency and risk management; inclusive and fair trade policies that promote sustainable practices; targeted interventions for bottlenecks such as logistics, food security, and inclusive business models; and the establishment of proactive and multi-actor governance mechanisms among policymakers, industry stakeholders, and academia (Awokuse et al., 2024). Furthermore, while digitalization (blockchain, IoT, big data, artificial intelligence, etc.) offers significant opportunities for increasing traceability from farm to fork and predicting risks, this potential cannot always be systematically measured due to obstacles such as interoperability, cost, digital competence, and regulatory uncertainties (Stanescu et al., 2025). Therefore, there is a need for analytical frameworks that evaluate countries’ agri-food systems in a consistent and comparable way using multidimensional indicators.

At this point, Multi-Criteria Decision Making (MCDM) approaches offer a strong analytical foundation thanks to their holistic approach to interacting and sometimes conflicting criteria. For example, the combined use of AHP and DEMATEL can support more rational and evidence-based decision-making processes by making it possible to determine the relative importance of criteria and analyze the causal relationships between them (Muyulema-Allaica et al., 2025). However, many indices and evaluation frameworks developed in the literature often focus on specific stages of the chain or limited sets of indicators; therefore, they are limited in evaluating sustainability and resilience simultaneously, comparably, and holistically (Chen et al., 2025). Consequently, when comparing interconnected but structurally different agri-food systems such as those of European countries, there is a need for comparative analyses that enhance the robustness of results by combining MCDM methods with different decision logics.

This study aims to comparatively evaluate the agri-food systems of six representative European countries using a multi-criteria framework that considers both sustainability and resilience dimensions. Within this framework, the decision problem is defined as identifying the relative sustainability–resilience performance of the selected countries. The countries are treated as decision alternatives, while the selected indicators represent evaluation criteria. The objective of the multi-criteria decision-making (MCDM) model is therefore to rank the countries according to their overall agri-food system performance under multiple sustainability and resilience indicators. Due to the limitations of representing evaluation criteria with different scales and aspects, such as environmental pressures, resource use, external dependence, food security, climate adaptation, and logistics capacity, with single or fixed-weighted indices, the study adopts a multi-criteria decision-making (MCDM) approach. While several additional criteria such as soil quality, geopolitical disruptions affecting food supply chains, and food losses along the supply chain may also influence agri-food system sustainability, the present study focuses on a set of internationally comparable criteria in order to establish a consistent multi-criteria evaluation framework. Within this framework, the relative importance of the criteria is determined using objective weighting methods such as Entropy and CRITIC, and the integrated sustainability-resilience performance of the countries is analyzed using MCDM methods that reflect different decision logics, such as TOPSIS, VIKOR, and COPRAS. Thus, the goal is both to make the performance differences between countries more visible and to present the balance relationships between sustainability and resilience in a more understandable way for policy development. The MCDM framework is used here as an analytical decision-support tool that enables policymakers and researchers to identify which countries exhibit relatively stronger or weaker sustainability–resilience performance in their agri-food systems.

Although each of the evaluation criteria can provide useful information when examined separately, agri-food systems represent complex socio-economic and environmental structures in which multiple factors interact simultaneously. Evaluating criteria individually may reveal specific strengths or weaknesses, but it does not provide a comprehensive understanding of the overall system performance. For instance, a country may perform well in food security but poorly in water efficiency or greenhouse gas emissions. Therefore, policy-relevant analysis requires an integrated assessment framework capable of simultaneously considering multiple dimensions of the system. Multi-Criteria Decision-Making (MCDM) approaches provide such a framework by enabling the aggregation of diverse criteria into a structured evaluation model while preserving their relative importance. In addition, composite indicator approaches are widely used in sustainability and food system analyses because they allow complex multidimensional phenomena to be summarized into interpretable comparative measures (Reig-Martínez et al., 2011; Talukder et al., 2018; Castillo-Diaz et al., 2023).

This study makes several contributions to the literature on agri-food system assessment. First, it develops an integrated analytical framework that jointly evaluates sustainability and resilience, two dimensions that are still frequently examined separately in the existing literature. Second, it applies this framework at the cross-country level using a representative set of European countries, thereby providing a comparative perspective on structurally different agri-food systems. Third, the study combines objective weighting methods (Entropy and CRITIC) with multiple ranking techniques (TOPSIS, VIKOR, and COPRAS), which allows the robustness of the results to be assessed across different decision logics. Finally, by linking the selected criteria to SDG-related dimensions and internationally comparable datasets, the study provides a transparent and policy-relevant basis for evaluating national agri-food system performance.

The rest of the article is structured as follows: Section 2 summarizes the literature specific to studies addressing the concepts of sustainability and resilience in the context of agri-food systems using MCDM methods. Section 3 details the criteria used in the study, data sources, and the application steps of the Entropy and CRITIC methods for weighting and the TOPSIS, VIKOR, and COPRAS methods for ranking. Section 4 reports the analysis findings, compares country rankings, and discusses inter-method consistency. Section 5 concludes the study by comparing the ranking methods, discussing policy and governance implications, and presenting the study’s limitations and directions for future research.

2 Literature review

In recent years, the concepts of sustainability and resilience have been increasingly addressed in the literature on agri-food supply chains, although these two concepts are often examined as independent approaches. Sustainability and resilience are closely related but conceptually distinct perspectives frequently used in the analysis of socio-ecological and agri-food systems. Sustainability is generally considered a normative and goal-oriented concept that aims to ensure the long-term maintenance of environmental, economic, and social systems while safeguarding the needs of future generations (Derissen et al., 2011; Scown, 2024). In this context, sustainability represents a holistic approach that simultaneously considers economic, environmental, and social dimensions and aims to establish a balance between these dimensions. It emphasizes intergenerational equity, conservation of natural capital, and the maintenance of ecological balance through responsible resource use and long-term development strategies (Derissen et al., 2011; Lew et al., 2016). In contrast, resilience is primarily a descriptive and system-oriented concept referring to a system’s capacity to absorb disturbances, adapt to changing conditions, and recover while maintaining its core functions (Derissen et al., 2011; Tainter and Taylor, 2014). In supply chain contexts, resilience reflects the ability of a system to withstand shocks such as climate-related disruptions, geopolitical crises, or market fluctuations while maintaining operational continuity. While sustainability focuses mainly on long-term structural stability and resource conservation, resilience highlights the dynamic capacity of systems to cope with shocks and uncertainties through adaptation and recovery mechanisms (Scown, 2024; Hyz, 2024). Consequently, the two concepts differ in their analytical orientation and temporal focus: sustainability is primarily associated with long-term development goals and normative policy objectives, whereas resilience emphasizes robustness and adaptive capacity under uncertain and disruptive conditions (Scown, 2024; Hyz, 2024). To clarify these conceptual distinctions, Table 1 summarizes the key differences between sustainability and resilience in the context of agri-food systems.

However, traditional resilience approaches may be insufficient in the context of sustainable supply chains that require multi-dimensional performance. This is because environmental or social performance may be compromised while economic recovery is achieved. This situation creates a dilemma regarding whether resilience should be evaluated in a single dimension or within multiple dimensions. Supply chains that focus solely on sustainability but ignore resilience may fail to maintain their performance under disruptive conditions, while systems that focus solely on resilience may deepen environmental and social problems. Therefore, integrating sustainability and resilience is critically important for securing the long-term performance of agri-food supply chains even under risks and disruptions. In this sense, sustainability and resilience should not be viewed as independent concepts but rather as complementary ones; where resilience should be considered as a fundamental capacity that enables sustainable performance to be maintained even in the face of shocks and systemic disruptions (Paredes-Rodríguez et al., 2024).

This discussion also raises the question of how supply chain length and organizational structure affect the sustainability-resilience relationship. A comprehensive literature review on short food supply chains (SFSCs) shows that features such as locality, producer-consumer proximity, and disintermediation have the potential to reduce environmental impacts, support local economies, and increase adaptability during times of crisis in some contexts; however, these effects can vary depending on the context, scale, and governance conditions. These findings reveal that addressing sustainability and resilience together is necessary not only conceptually but also in terms of policy and governance design (Sciortino et al., 2025).

Various studies in the literature highlight the increasing importance of multi-criteria decision analysis (MCDA/MCDM) methods in agri-food research, noting that these methods possess different strengths and weaknesses depending on the problem type and offer comprehensive decision support to researchers (Saqlain et al., 2025). The review by Gésan-Guiziou et al. (2020) systematically presents the diversity, application areas, and methodological trends of MCDA methods used in the agri-food literature, drawing attention to the importance of transparency and consistency in method selection. This body of literature confirms that MCDA offers a suitable analytical framework for evaluating agri-food systems, which are inherently multidimensional and involve economic, environmental, and social trade-offs.

The conceptual focuses and MCDM techniques used in studies on sustainability and resilience in agri-food systems vary considerably across the literature. This variation makes direct comparison of findings difficult. However, a closer examination of the recent literature reveals three important gaps. First, most previous studies examine either sustainability or resilience separately, while studies that evaluate these two dimensions in an integrated manner remain limited. Second, many existing studies rely on a single decision-making method or a single methodological logic, which restricts the robustness of the results and makes it difficult to assess the stability of rankings across alternative MCDM approaches. Third, a large share of the literature focuses on specific case contexts—such as supplier selection, regional systems, or single-country applications—rather than providing a comparative cross-country assessment of structurally different agri-food systems using internationally comparable indicators.

In order to clarify these methodological differences and highlight the existing research gap addressed by this study, Table 2 presents a comparative overview of representative studies that explicitly apply MCDM approaches in the analysis of agri-food systems. The studies included in the table were identified through a targeted literature search conducted in Google Scholar using keywords “agri-food,” “supply chain,” “sustainability,” “resilience,” and “MCDM.” The search was limited to the 2022–2026 period in order to focus on the most recent developments in the literature, particularly those emerging in the post-COVID-19 context, where the resilience and sustainability of agri-food systems have received increased scholarly attention. Rather than providing an exhaustive review of the broader literature, the purpose of this table is to focus specifically on studies employing multi-criteria decision-making methods; however, several additional studies deemed particularly relevant to the topic are also included.

In Table 2, selected studies are presented comparatively in terms of: (i) sustainability/resilience focus, (ii) key concepts addressed, (iii) methodological approach and MCDA/MCDM tools used, (iv) case/application context (country or supply chain level), and (v) contribution to the literature/findings. This comparison shows that studies addressing sustainability and resilience in an integrated manner are limited. It also reveals an important need to combine methods with different decision logics in order to increase the robustness of ranking results. This situation clearly demonstrates the importance and contribution to the literature of the Entropy–CRITIC weighting and TOPSIS–VIKOR–COPRAS-based comparative framework that we proposed in our study.

Building on these observations, the present study addresses an important gap in the literature by proposing a comparative multi-criteria framework that evaluates sustainability and resilience simultaneously across agri-food systems in representative European countries. Unlike previous studies that often focus on a single dimension, a single method, or a single case context, this study integrates multiple objective weighting and ranking methods within a cross-country analytical design. Accordingly, the novelty of this study lies not in applying a single MCDM technique to a new dataset, but in constructing a comparative and methodologically robust framework for integrated agri-food system assessment. In this way, it contributes not only to the empirical assessment of agri-food systems, but also to the methodological literature on robust multi-criteria evaluation in sustainability and resilience research.

3 Application

In this study, the multi-criteria decision-making (MCDM) problem is formulated as a comparative performance evaluation problem at the country level. The decision alternatives consist of six representative countries (Türkiye, Greece, Italy, France, Germany, and the Netherlands), while the evaluation criteria represent key sustainability and resilience indicators of agri-food systems, including import dependency, household food waste, greenhouse gas emissions, water use, food security, climate resilience, and logistics capacity. The objective of the decision model is to determine the relative performance ranking of these countries based on their overall agri-food system sustainability and resilience. In this context, the study adopts a cross-sectional comparative design aimed at evaluating structural differences among national agri-food systems rather than analyzing temporal trends. Therefore, the analysis is conducted using the same year or the nearest available year for each criterion, ensuring that all indicators are evaluated under the same macroeconomic, environmental, and policy conditions. This approach enhances the comparability of countries within the MCDM framework by minimizing the influence of short-term fluctuations caused by climatic variability, market shocks, or policy changes that may occur across different years.

Following the selection of relevant criteria and alternatives and the construction of the decision matrix, criterion weights are calculated using the Entropy and CRITIC methods, and country rankings are obtained through TOPSIS, VIKOR, and COPRAS. By simultaneously considering multiple environmental, economic, and social indicators, the MCDM framework enables the systematic comparison of countries whose agri-food systems exhibit different structural characteristics. Accordingly, the decision-making context of this study does not involve selecting a single optimal alternative but rather identifying the relative strengths and weaknesses of countries within a multidimensional sustainability–resilience evaluation framework, thereby providing a consistent and transparent basis for cross-country performance assessment.

Detailed descriptions of the employed weighting and ranking methods are provided in this section.

3.1 Data description

Comparing the performance of sustainable and resilient agri-food supply chains at the country level requires the use of a holistic set of criteria that simultaneously consider economic, environmental, and social dimensions. Accordingly, based on widely used criteria in the literature, the most critical ones have been identified, and seven key criteria demonstrating a high level of alignment with sustainability and resilience frameworks have been selected. The selected criteria are: (C1) Import dependence (based on wheat, one of the main cereals), (C2) Household food waste (estimated), (C3) Environmental impact (food-system GHG emissions per capita), (C4) Water use (agricultural / total withdrawals), (C5) Food security status, (C6) Climate resilience, and (C7) Logistics capacity.

(C1) Import dependency (wheat-based) is a key criterion measuring countries’ vulnerability to external shocks, price fluctuations in global food markets, trade restrictions, and geopolitical shocks. High import dependency is considered a structural risk factor threatening the continuity of food supply and national food security. Therefore, import dependency, calculated particularly for strategic staple grains such as wheat, is considered a central variable and widely used in the food security and supply chain resilience literature (Puma et al., 2015; Clapp, 2017; FAO, 2021a). Data for the C1 criterion was obtained from statistics published by FAO (FAOSTAT, 2025).

(C2) Household food waste (estimated) is a key sustainability criterion, representing the efficiency and resource use effectiveness of the food supply chain at the consumption stage. Food waste not only leads to economic losses but also negatively impacts environmental sustainability by causing unnecessary energy, water, and land use. Therefore, the food waste indicator is considered one of the central, fundamental performance metrics in the sustainable consumption and circular economy literature (Gustavsson et al., 2011; Principato et al., 2021). Data for the C2 criterion is taken from the Food Waste Index report prepared by the United Nations Environment Programme (UNEP). These values ​​are estimated values ​​for the food and beverage services and retail sectors (UNEP, 2024).

(C3) The environmental impacts of food supply chains encompass multifaceted elements such as greenhouse gas emissions, soil degradation, land use, and ecosystem disruption. Environmental impact indicators are widely used to quantitatively assess the sustainability of agricultural and food systems and are strongly supported by the life cycle approach. Inclusion of this criterion in the assessment allows for the quantitative analysis of environmental sustainability (Garnett, 2013; Poore and Nemecek, 2018). Data for the C3 criterion was obtained from the EDGAR-FOOD database (EDGAR-FOOD, 2025).

(C4) Water use is one of the most critical indicators of environmental pressure, particularly in food supply chains where agricultural production is intensive. The agricultural sector accounts for a large portion of global freshwater use, and water footprint indicators are accepted as a standard measure in the assessment of sustainable food systems (Hoekstra and Mekonnen, 2012; Mekonnen and Hoekstra, 2014). Data for the C4 criterion was obtained from the AQUASTAT database (FAO, 2025).

(C5) Food security status is a comprehensive social sustainability criterion that assesses countries in terms of their populations’ continuous access to sufficient, safe, and nutritious food. This criterion provides a key indicator in assessing the social sustainability of the food supply chain, encompassing not only the quantity (sufficiency) of supply but also accessibility, stability, and usage dimensions (FAO, 2008; Barrett, 2010). Data for the C5 criterion was obtained from a report prepared in collaboration with FAO, IFAD, UNICEF, WFP, and WHO (2025).

(C6) Climate resilience measures the adaptation and recovery capacity of food supply chains in the face of increasing climate change risks. Since climate uncertainty, extreme weather events, drought, and temperature increases have direct impacts on agricultural production and logistics networks, climate resilience is considered one of the most critical and strategic components of future supply chain performance in the sustainable food systems literature (Wheeler and Von Braun, 2013; Tendall et al., 2015). Data for the C6 criterion was obtained from the ND-GAIN database (ND-GAIN, 2025).

Finally, (C7) logistics capacity represents the infrastructural and managerial adequacy that enables the efficient, rapid, and loss-free transportation, distribution, and storage of food products from harvest to final consumption. Strong logistics systems both reduce food losses and support food supply chain continuity during times of crisis (Beske et al., 2014; Arvis et al., 2018). Data for the C7 criterion was obtained from the World Bank database (World Bank, 2025).

Information regarding all these criteria is summarized in Table 3.

The criteria used in this study were selected to represent key structural and outcome dimensions of agri-food system sustainability and resilience. In particular, import dependency (C1) captures a country’s exposure to external food supply risks, while food security (C5) represents a fundamental outcome reflecting the availability, accessibility, and stability of food supply. Other indicators describe intermediate structural characteristics of food systems, including household food waste (C2), environmental impact (C3), water use (C4), climate resilience (C6), and logistics capacity (C7). Together, these indicators provide a multidimensional perspective on how resource use, environmental pressures, infrastructure capacity, and external dependencies jointly shape food system performance.

From a theoretical perspective, import dependency (C1) can influence food security through multiple transmission channels. Countries that rely heavily on food imports may become more vulnerable to global price shocks, trade restrictions, and supply chain disruptions. These external risks can reduce food availability or increase food prices, thereby affecting household food access. At the same time, the magnitude of this effect depends on other structural characteristics of the food system. For instance, strong logistics capacity (C7) may stabilize supply chains, while higher climate resilience (C6) may reduce production volatility and mitigate external dependency. Consequently, the relationship between import dependency and food security is often context-dependent and mediated by additional system factors, which justifies considering these indicators simultaneously within the analytical framework.

Because agri-food systems are inherently interconnected, the indicators were not assumed to be strictly independent. Instead, they were selected to capture different but potentially interacting dimensions of the system. To examine possible relationships among the indicators, an exploratory correlation analysis was conducted using Pearson and Spearman coefficients.

The Table 4 indicates that several criteria pairs exhibit relatively strong relationships. In particular, high correlation values are observed among the variables C4, C5, C6, and C7. The strongest relationships appear for the pairs C6–C7 (Pearson r = 0.953; Spearman ρ = 0.883), C4–C7 (r = −0.948; ρ = −0.971), and C4–C6 (r = −0.925; ρ = −0.829). In addition, the relationships between C4–C5 (r = 0.882; ρ = 0.926) and C5–C7 (r = −0.754; ρ = −0.953) are also notably strong. In contrast, C1 generally exhibits weak correlations with the other criteria, and no clear relationship is observed. These findings suggest that all criteria in the dataset are not entirely independent of each other, with particularly strong interconnections observed among the C4–C7 group. However, given the very small sample size (n = 6), the results should be interpreted as exploratory in nature.

The presence of a certain level of correlation among some criteria used in this study (particularly water use, climate resilience, and logistics capacity) is an expected situation. This is primarily due to the inherently interconnected and multidimensional nature of agri-food systems. These criteria represent different but interrelated aspects of the system. Therefore, it is not theoretically expected that they would be completely independent. Moreover, the aim of this study is not to estimate causal relationships among criteria, but rather to comparatively evaluate the performance of different countries’ agri-food systems using a multidimensional set of indicators. For this reason, a certain degree of correlation among the criteria does not weaken the methodological validity of the model. On the contrary, it reflects the holistic structure of the food system more realistically. In addition, unlike regression-based models, linear relationships among criteria (multicollinearity) are not considered a critical methodological issue in MCDM approaches. This is because the primary objective of MCDM methods is not to predict a dependent variable, but to evaluate multiple criteria simultaneously in order to generate a relative performance ranking. Therefore, the criteria set was designed to capture complementary dimensions of the agri-food system rather than assuming strict statistical independence among the criteria.

The clear linkage of these criteria to SDG targets allows for a detailed connection between country-level performance and the SDGs. This micro-level alignment strengthens the policy validity and interpretability of the assessment results obtained in the study. It also increases the applicability of the proposed framework for cross-country performance comparisons in sustainable agri-food supply chains. In addition, the diversity of selected criteria allows for the identification of inter-criteria interactions and performance differences between countries in multi-criteria decision-making (MCDM) analyses. In particular, the combined assessment of environmental (water use, climate resilience, and environmental impact), economic (import dependence, food waste, and logistics capacity), and social (food security) dimensions creates a natural performance divergence between countries and increases the analytical significance of the ranking results. Finally, the inclusion of benefit and cost criteria in the study allows for the effective and consistent application of TOPSIS, VIKOR, and COPRAS methods.

On the other hand, this study focuses on six countries —Türkiye, Greece, Italy, France, Germany, and the Netherlands—in order to comparatively analyze how sustainable and resilient agri-food supply chains are shaped in different socio-economic, climatic, and structural contexts. These countries were selected because their agricultural, environmental, and socio-economic indicators are regularly, consistently, and comparably reported in international databases such as FAO, World Bank, OECD, and EUROSTAT, which enables the creation of measurable and verifiable criteria necessary for MCDM analyses.

The selected countries exhibit a high degree of heterogeneity in terms of agricultural production structures, food supply chain organization, sustainability policies, and levels of economic development, allowing them to represent different structural types of sustainable agri-food supply chains in Europe and its surrounding geography. France and Germany represent large-scale agricultural economies within the European Union, characterized by high production volumes, advanced technological adoption, and well-established institutional agricultural policies. In these countries, sustainability strategies are often framed around productivity, environmental balance, carbon footprint management, and circular economy practices.

In contrast, Italy and Greece represent Mediterranean agricultural systems characterized by fragmented farm structures, climate-sensitive production patterns, and relatively high vulnerability to water scarcity and climate change impacts. These countries are also particularly important in terms of biodiversity, local food systems, and the sustainability of traditional agricultural production practices. Türkiye, located at the intersection of Europe and Asia, represents a major agricultural producer with a strategic role in regional food supply and food trade networks, while also sharing several structural characteristics with Mediterranean agricultural systems.

The Netherlands represents a distinct model within the European agri-food system. Despite its limited agricultural land, it is widely recognized for its technology-intensive agricultural production system. Through advanced greenhouse technologies, digital farming applications, and a highly integrated supply chain structure, the Netherlands has developed a sustainability-oriented production model based on innovation and technological efficiency (Van der Ploeg et al., 2019; FAO, 2021b; OECD, 2021; IPCC, 2022). Within this framework, the six selected countries jointly capture both Mediterranean and Northern European agricultural system typologies, encompassing different climatic conditions, production scales, institutional structures and policy approaches. In addition, the sample includes both EU member states and a candidate country with strong economic and regulatory integration with the EU, allowing the analysis to reflect potential patterns of policy convergence and divergence across different governance contexts.

While other European countries such as Spain and Denmark also play important roles within the European agri-food system, they were not explicitly included in the present analysis in order to maintain a balanced and analytically manageable comparison set. Spain shares several structural and climatic characteristics with the Mediterranean agricultural systems already represented by Italy and Greece, while Denmark largely reflects highly industrialized and technologically advanced agricultural structures conceptually comparable to the Dutch model included in this study. Therefore, including these countries would not substantially increase the diversity of agricultural system typologies captured in the analysis but would increase the dimensionality of the decision matrix within the multi-criteria framework.

For this reason, the selected representative country set prioritizes structural representativeness rather than numerical completeness, allowing the study to capture key variations in agri-food systems of representative European countries while preserving methodological clarity and interpretability within the multi-criteria decision-making framework.

3.2 Entropy weighted method

The Entropy Weighted Method (EWM) is a multi-criteria decision analysis approach that assigns weights to criteria based on the degree of value dispersion. The fundamental principle of EWM is that the greater the dispersion and differentiation of values, the more information they contain, and thus, they should be assigned higher weights (Zhu et al., 2020). This method is particularly useful in scenarios where objective weighting is required, as it relies on the inherent information within the data rather than subjective judgments (Zavadskas and Podvezko, 2016). The literature presents two variants of the entropy-based weighting method. The first does not include normalization, while the second usually includes maximum-minimum normalization before calculating the entropy value for each criterion (Chen, 2019). In this study, the following calculation steps were performed, including normalization.

Step 1. Creating the decision matrix: First, the decision matrix consisting of alternatives (countries) and criteria is established. This decision matrix contains the original values of all criteria for the six countries included in the analysis (Table 5).

Step 2. Harmonization of cost/benefit aspects: In MCDM, some criteria are benefit and some are cost criteria. While performance improves as value increases in the benefit criterion, the opposite is true for the cost criterion. In the entropy method, all criteria must be transformed so that “high = good” (benefit). In this step, cost criteria are converted into benefits. The most frequently used transformation for cost criteria is:

The highest value in the columns containing the criterion to be minimized is found, and the relevant cell is subtracted from that value. The purpose of this step is to make all criteria interpretable in one direction. The relevant transformations for C1-C5 criteria have been applied to our dataset and are presented in Table 6.

Step 3. Normalizing the decision matrix (probability matrix): Since the scale of each criterion may be different, the values for each criterion are converted to normalized values that are interpreted as probabilities using Equation 3.

In this normalization, the sum of each criterion column becomes 1. Each is interpreted as the relative share of the alternative. The resulting matrix forms the basis of entropy calculations (Table 7).

Step 4. Calculating Criterion Entropy: Entropy is a concept derived from information theory and measures the level of uncertainty/homogeneity in a distribution. The closer the values of a criterion are to each other, the less “information content” that criterion has.

For each criterion , entropy is calculated as follows:

Here;

→ normalization coefficient (for to remain in the 0–1 range),

: number of alternatives,

In the expression , the term is considered 0.

If the values ​​are close to each other, meaning the variance of the criterion is low. Therefore, the information content is low, and its discriminative power is low (Table 8).

As decreases, the distribution becomes more heterogeneous. That is, some alternatives are very high while others are very low. Therefore, the information content is high, and the criterion is discriminative.

Step 5. Degree of Diversity (Difference): Entropy measures the lack of information. At this stage, the “amount of information it carries” for each criterion is quantified. Therefore, information content is defined in reverse:

If is high, the distribution of the criterion is heterogeneous, and the discrimination is high. If is low, the criterion is almost constant, and the information contribution is weak (Table 9).

Step 6. Calculation of Entropy Weights: In the final step, all values ​​are relativized and an Entropy Weight is obtained for each criterion using Equation 6.

The larger the value of , the higher the information content of criterion in the dataset and it is considered as the more significant in the MCDM model (Table 10).

According to the obtained values, the criteria that most shaped the model’s decision are:

C3 (GHG/person) = 0.2501.

C2 (Household food waste) = 0.2428.

C1 (Import dependency) = 0.2391.

These three together account for approximately 73.2% of the total weight.

Because GHG values vary widely (to very different levels) between countries, entropy is considered the criterion that “carries the most information.” The differences between countries are also significant in C2. Therefore, the entropy weight turned out to be almost as high as in C3. “Household food waste” seriously affects the scores of countries.

There is a significant disparity in the C4 criterion, with Türkiye scoring very high while other countries score quite low. However, this disparity is not as dominant as in C1-C3. Therefore, agricultural water withdrawal as % of total renewable water resources affects the ranking, but it is not the determining factor.

According to the prevalence of moderate or severe food insecurity, defined as C5, Türkiye is at 42.3, while other countries are in the ~5.4–5.5 range. This appears to be quite differentiating. However, the shape of the intra-column distribution in the transformation and normalization structure did not produce a signal as strong as C1–C3. Nevertheless, due to Türkiye’s situation, C5 is a critical “penalty” criterion that will pull Türkiye down in some methodologies.

The ND-GAIN Country Index and LPI criteria have entropy weights calculated as almost zero. Countries are close to each other in these criteria (low variance), and entropy considers low variance as “low information.” Therefore, ND-GAIN and LPI contribute very little to the composite ranking in this dataset. Since these two criteria are important in resilience assessment, the entropy weighting method suppresses resilience. Accordingly, the CRITIC weight was calculated based on variance and correlation between the criteria.

3.3 CRITIC

The CRITIC (Criteria Importance Through Intercriteria Correlation) method is an objective method used to determine the weights of criteria in multi-criteria decision-making (MCDM) processes. This method calculates the importance of criteria by considering the correlation and standard deviation of each criterion with other criteria (Žižović et al., 2020). The CRITIC method aims to obtain more balanced and objective weights, particularly by evaluating conflicting relationships and contrast intensity between criteria (Krishnan, 2025). This method provides decision-makers with a more reliable and consistent framework, offering significant advantages in the analysis of complex data sets and model evaluations.

Step 1. Creating the decision matrix: First, the decision matrix consisting of alternatives (countries) and criteria is established as given in Table 5.

Step 2. Normalizing the decision matrix: CRITIC generally uses min-max normalization, and calculations based on the benefit/cost ratio are as follows (Table 11):

Step 3. Calculating standard deviations: The first “information” component of the CRITIC method is the standard deviation of the criterion (Table 12). For each criterion:

Step 4. Creating the correlation matrix: The second component of the CRITIC method is determining how similar the information carried by the criteria is (Table 13). This is found by calculating the correlation value .

Step 5: CRITIC method rewards “different information” not similarity. Therefore, a matrix is created (Table 14).

Step 6: Calculate the value for each criterion Equation 11 (Table 15).

Step 7: Calculate the CRITIC information metric using Equation 12 (Table 16).

Step 8. Calculation of CRITIC Weights: In the final step, all values ​​are relativized and an Entropy Weight is obtained for each criterion:

The larger the value of , the higher the information content of criterion in the dataset and the more significant it is considered in the MCDM model (Table 17).

The entropy method assigns very low weights to indicators like climate resilience (ND-GAIN) and logistics capacity (LPI), which have high homogeneity and low variance among selected European countries, because it determines criterion weights based solely on the distribution of data among countries. In contrast, the CRITIC method reduces data overlap by considering both the variability of the criteria and their correlations with each other, producing more balanced weights. Therefore, CRITIC ensures the meaningful inclusion of structurally important indicators, such as resilience and logistics, in the evaluation process. For this reason, CRITIC weights will be used in the ranking of countries (Table 18).

3.4 TOPSIS

The Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) is a multi-criteria decision-making (MCDM) method developed by Hwang and Yoon in 1981 (Sharma et al., 2020; Thakkar, 2021a). This method is designed to identify the best alternative from a set of options by comparing their distances to an ideal solution and an anti-ideal solution. The fundamental concept of TOPSIS is that the chosen alternative should have the shortest distance from the ideal solution and the farthest distance from the anti-ideal solution (Mahmudova, 2020; Madanchian and Taherdoost, 2023). This approach is widely recognized for its simplicity and effectiveness in handling decision-making problems involving multiple criteria. The detailed steps of the TOPSIS method are given below.

Step 1. Normalizing the decision matrix: Since the criteria can have different units of measurement (e.g., %, index, tons CO₂/person), the decision matrix is ​​normalized to make it dimensionless. Vector normalization is commonly used in the TOPSIS method. This process makes the values ​​in each criterion column comparable by removing their individual magnitudes (Table 19).

Step 2. Calculating weighted normalized matrix: Each normalized criterion value is multiplied by the weight assigned to that criterion. This reflects the relative importance of the criteria in the decision-making process. In this step, weights obtained by CRITIC method which categorized under objective methods are used. The matrix obtained in this step forms the basis of the TOPSIS analysis (Table 20).

Step 3. Calculating Ideal Solutions: In this step, two reference points are defined. The positive ideal solution ( consists of the best performance values for each criterion (highest values for benefit criteria, lowest values for cost criteria). The negative ideal solution ( consists of the worst performance values for each criterion (lowest values for benefit criteria, highest values for cost criteria). These two points represent the “ideal” and “bad” situations that the alternatives should avoid (Table 21).

Step 4. Calculating the distances of alternatives from ideal solutions: For each alternative, the distance to the positive ideal solution and the distance to the negative ideal solution are calculated separately according to Equations 17, 18. These distances are usually determined using Euclidean distances. The aim is to measure how close the alternatives are to the ideal and how far they are from the worst-case scenario.

Step 5. Calculating the closeness coefficient and ranking the alternatives: A closeness coefficient () is calculated for each alternative. This coefficient is the ratio of the alternative’s distance from the negative ideal solution to the sum of both distances. The closeness coefficient takes values ​​between 0 and 1, and a higher value indicates that the alternative is closer to the ideal solution. Then, the alternatives are ranked from highest to lowest according to their closeness coefficient. The alternative with the highest closeness coefficient is considered the best option according to the TOPSIS method. This ranking constitutes the final result of the multi-criteria decision problem (Table 22).

The analysis results show that France has the highest proximity coefficient () and therefore performs best according to the TOPSIS method. France’s distance from the positive ideal is largely due to the greenhouse gas emissions (C3) criterion, while it is quite close to the ideal in other criteria. Germany ranks second, and despite its strong performance in benefit criteria such as climate resilience (C6) and logistics capacity (C7), its proximity to the positive ideal is limited by greenhouse gas emissions (C3) and import dependence (C1) criteria. Greece is in third place; its distance from the ideal solution is largely due to import dependence (C1), but it has a relative advantage in environmental impact and food insecurity indicators.

The Netherlands and Italy rank in the lower-middle range, and for both countries, the distance from the positive ideal is predominantly determined by the import dependency (C1) criterion. In the Netherlands, in particular, this criterion stands out as a single disadvantage factor explaining almost the entire distance. Türkiye, with the lowest proximity coefficient, ranks last. Türkiye’s distance from the positive ideal solution stems not from a single criterion, but from the combined effect of multiple high-weighted cost criteria, primarily import dependency (C1), greenhouse gas emissions (C3), and food insecurity (C5). This indicates that Türkiye has a structural and multidimensional disadvantage in its overall resilient and sustainable food system performance.

Overall, the TOPSIS results show that the ranking of countries under the CRITIC weights is shaped particularly by the criteria of environmental impact (C3) and import dependence (C1); even extreme weakness in a single criterion can seriously limit proximity to the positive ideal. These findings demonstrate that, thanks to its balance-oriented structure, the TOPSIS method evaluates countries’ sustainability performance from a holistic perspective and clearly reflects multi-criteria trade-offs.

3.5 VIKOR

The VIKOR method, which stands for “VIseKriterijumska Optimizacija I Kompromisno Resenje” in Serbian, meaning Multi-Criteria Optimization and Compromise Solution, is a multi-criteria decision-making (MCDM) tool developed to address problems with conflicting criteria. Initially proposed by S. Opricovic in 1979 and first applied in 1980, VIKOR focuses on ranking and selecting from a set of alternatives by determining a compromise solution that is closest to the ideal solution (Chatterjee and Chakraborty, 2016). This method is particularly valued for its ability to provide a balanced solution that considers the relative importance of different criteria, making it suitable for various strategic decision-making scenarios in social, economic, and environmental contexts (Thakkar, 2021b).

Step 1. Determination of ideal solution values: After creating the decision matrix, the best and the worst values are determined for each criterion (Table 23).

Step 2. Calculating normalized distance matrix: In the VIKOR method, the “distance from the ideal solution” is calculated based on the benefit/cost as follows:

For , 0 is the best, 1 is the worst (Table 24).

Step 3. Calculating weighted distance matrix: Each normalized distance matrix value is multiplied by the weight assigned to that criterion by CRITIC method (Table 25).

Step 4. Calculating the, and The VIKOR analysis, performed using CRITIC weights, ranked countries according to the compromise solution index (Q) as follows: Greece, France, Germany, Netherlands, Italy, and Türkiye (Table 26). Greece’s lowest value is explained by both its relatively low overall performance gap () and its limited level of individual regret () stemming from the weakest criterion. Its superior performance, particularly in food system-related greenhouse gas emissions (C3) and food insecurity (C5), which were assigned high weights by the CRITIC method, brought Greece closer to the ideal solution. France, despite having the best values in terms of overall benefit, ranked second due to its relative weakness, particularly in the environmental impact (C3) criterion. The extremely close values of Germany and the Netherlands indicate a trade-off relationship between the criteria rather than a clear superiority between the two countries. Germany excels in benefit criteria such as climate resilience and logistics capacity, while high greenhouse gas emissions limit its ranking. In contrast, although the Netherlands performs better in environmental and water use indicators, it is at a disadvantage under the CRITIC weights due to its high import dependency. Therefore, it should be checked whether the Germany/Netherlands ranking would change if the parameter were changed from 0.5 to 0.25 or 0.75 using sensitivity analysis. Italy, having moderate values in many criteria, has not shown a significant advantage in either overall performance or individual regrets and ranks low. Türkiye, on the other hand, has been identified as the country furthest from the ideal solution, especially in high-weighted cost criteria such as greenhouse gas emissions, water use, and food insecurity; it ranks last in the VIKOR ranking due to having the highest S and R values. These results show that the VIKOR method under CRITIC weights strongly differentiates the sustainability and resilience performance of countries’ agri-food systems, especially along the axes of environmental impact and external dependency.

Step 5. VIKOR Robustness Assessment: In this step, the sensitivity analysis has been conducted under different values of the compromise parameter . The results were shown in Table 27.

The VIKOR sensitivity analysis, performed using CRITIC weights, shows that country rankings are generally stable under different values of the compromise parameter ( 0.25, 0.50, and 0.75). The most notable change is the shift in leadership from Greece to France as the value increases. This is due to France having a lower overall performance gap (), while Greece has a more balanced structure in terms of individual regret (). The change in ranking between Germany and the Netherlands is a result of these two countries’ superiority in different criteria (e.g., Germany’s resilience and logistics capacity, the Netherlands’ environmental and water use performance) showing different sensitivities to the total benefit and individual risk components. In contrast, Türkiye’s last place ranking in all scenarios reveals that the results obtained are highly robust for this country and that the performance gap persists independently of the methodological parameters. Overall, the sensitivity analysis shows that the basic structure of the VIKOR ranking is preserved, and the results exhibit strong methodological consistency, particularly for countries at both the top and bottom of the rankings.

VIKOR sensitivity analysis shows that the rankings obtained under different values of the consensus parameter are largely preserved, and the results are highly robust against the method parameters, especially for the top and bottom ranked countries (Figure 1).

Figure 1

3.6 COPRAS

Step 0. Positive Adjustment (COPRAS prerequisite): Since the COPRAS method normalizes based on column sums, its classic form assumes all criterion values are positive. Because our dataset contains negative values under the C1 criterion (e.g., France −0.59), a constant is added to the relevant column to move the minimum value to 0 (Table 28).

The decision matrix , which has been made positive, is as follows:

Step 1. Normalizing the decision matrix: This process makes the values in each criterion column comparable by removing their individual magnitudes (Table 29).

Step 2. Calculating weighted normalized matrix:

See Table 30.

Step 3. Calculating the total benefits and costs:

Step 4. Calculating Relative importance and utility level

According to the analysis results, France ranked first in the COPRAS ranking with the highest relative importance and utility level score (Table 31). France’s position stems from having one of the lowest total cost components, particularly due to its low weighted normalized values in cost criteria such as import dependence (C1), food loss (C2), and water use (C4). France’s competitive level in terms of benefit criteria such as climate resilience (C6) and logistics capacity (C7) also brings it closer to the ideal solution within COPRAS’s cost–benefit analysis framework.

Greece ranks second, with its COPRAS performance primarily based on relatively low cost values in the environmental impact (C3) and food insecurity (C5) criteria. Although Greece does not have values as high as Germany and the Netherlands in the benefit criteria, the fact that the cost component is a determining factor in the COPRAS method has allowed it to rank higher. The Netherlands and Germany produced similar results; the Netherlands gained a slight advantage due to its lower total cost compared to Germany, while Germany’s high climate resilience and logistics capacity performance strongly contributed to its benefit score. However, Germany’s relatively high cost values, particularly in greenhouse gas emissions (C3) and import dependence (C1) criteria, increased the total.

Italy ranks in the lower-middle range in the COPRAS results, a situation explained by the country’s inability to demonstrate a significant advantage in benefit criteria and its disadvantageous profile in high-weighted cost criteria such as import dependence and environmental impact. Türkiye, on the other hand, ranks last in the COPRAS ranking. Türkiye’s results are characterized by a significantly higher total cost component, particularly due to high-weighted normalized values in cost criteria such as import dependence, water use, food insecurity, and greenhouse gas emissions. The cost-criteria-sensitive nature of the COPRAS method clearly reveals Türkiye’s multifaceted structural disadvantages and distinguishes it significantly from other countries in terms of relative importance values.

4 Results

The ranking results obtained from the MCDM model provide important insights into the comparative sustainability and resilience performance of the agri-food systems across the six selected countries. While the numerical results presented in the previous section describe the computational procedure of the model, this section focuses on interpreting these findings from a statistical perspective. By examining the relative positions of countries in the final ranking, it becomes possible to identify structural strengths and weaknesses in their agri-food systems. These findings can contribute to a better understanding of how different dimensions—such as resource efficiency, environmental pressure, and food security—collectively shape the overall performance of national agri-food systems.

The results indicate that countries with stronger logistical infrastructure and lower levels of food waste tend to achieve higher overall performance in the integrated assessment. In contrast, countries facing higher resource pressures, particularly in terms of water use and greenhouse gas emissions, tend to receive lower composite scores. This pattern highlights the importance of balancing production efficiency with environmental sustainability in the long-term resilience of agri-food systems.

4.1 Inter-method consistency with spearman rank correlation

In this study, COPRAS, TOPSIS, and VIKOR methods were used for ranking countries from best to worst performance. The Spearman test, used for ranked data, will be used to statistically test whether the methods yield similar rankings for the same countries. For this purpose, the obtained rankings are first arranged below as rank vectors representing the relative positions of the alternatives (Table 32).

Next, the Spearman rank correlation coefficient (ρ) was calculated for the pair of methods. This coefficient measures the degree to which two different methods yield similar rankings for the same countries. Spearman correlation is calculated based on the sum of the squares of the differences between the rankings and takes values between −1 and +1. A coefficient closer to +1 indicates that the rankings of the methods largely overlap; a coefficient closer to 0 indicates no relationship between the rankings; and a negative value indicates an inverse ranking. In this study, three pairwise comparisons were performed as given in Table 33: COPRAS–TOPSIS, COPRAS–VIKOR, and TOPSIS–VIKOR (Table 33).

Spearman rank correlation analysis clearly shows that the country rankings obtained using the COPRAS, TOPSIS, and VIKOR methods are largely consistent with each other. The calculated correlation coefficients ranging from 0.83 to 0.89 indicate that the rankings of the three methods exhibit a high to very high level of positive correlation. In particular, the highest correlation value (ρ = 0.8857) between the COPRAS and VIKOR methods suggests that, despite having different mathematical structures, these two methods evaluate the relative performance of countries similarly. It appears that COPRAS’s linear structure based on cost–benefit divergence and VIKOR’s compromise solution and regret-based approach produce a common evaluation logic, especially for countries at the top and bottom of the rankings. The high and statistically significant correlation coefficients (ρ = 0.8286) between the TOPSIS method and COPRAS and VIKOR demonstrate that TOPSIS’s geometric approach, based on proximity to the ideal solution, largely overlaps with the results of the other two methods. These findings indicate that the small differences observed in the rankings are secondary nuances stemming from the decision logics of the methods, while the overall ranking structure is preserved regardless of the method. Therefore, the Spearman analysis confirms that the three different MCDM methods used in the study produce consistent and mutually reinforcing results, supporting the methodological reliability and robustness of the findings.

5 Discussion and conclusion

This study contributes to the literature by offering a sustainability and resilience-oriented multi-criteria assessment framework for the comparative evaluation of agri-food systems at the country level. Its scientific contribution lies in integrating two closely related but often separately treated dimensions—sustainability and resilience—within a single comparative evaluation model, and in testing the consistency of country rankings through multiple objective weighting and ranking methods. In this respect, the study extends existing agri-food system research by providing a transparent, reproducible, and policy-relevant framework for assessing structural differences across agri-food systems in representative European countries.

The proposed framework allows for the evaluation of countries’ performance not only through relative rankings but also within the context of structural interactions between economic, environmental, and social dimensions. Based on indicators explicitly linked to the SDGs, this approach strengthens the link between academic analysis and policy design, providing a holistic foundation for assessments of sustainable and resilient food systems.

Beyond the specific ranking results for the six countries, the findings also contribute to the broader literature on agri-food system assessment. In particular, they show that sustainability and resilience should be evaluated jointly rather than separately, since countries may display strengths in some dimensions while remaining structurally vulnerable in others. The results also confirm the methodological value of combining different MCDM techniques, as the strong consistency across methods suggests that integrated multi-method frameworks can provide robust and interpretable comparative evidence for agri-food system research.

5.1 Comparative assessment of TOPSIS-COPRAS-VIKOR methods

The COPRAS, TOPSIS, and VIKOR methods, applied using objective weights determined by the CRITIC method, evaluated the performance of countries’ agri-food systems within the framework of different decision logics. The COPRAS method clearly separated benefit and cost criteria, highlighting the total impact of cost components namely import dependence, food waste, environmental impact, water use, and food insecurity. Therefore, it ranked countries with relatively low cost indicators, such as France and Greece, higher. The TOPSIS method, on the other hand, evaluated alternatives based on proximity to the positive ideal solution and distance from the negative ideal solution, leading France to stand out as the country that best achieved overall balance. Furthermore, Germany’s high infrastructure and resilience performance was more rewarded in this method. The VIKOR method, thanks to its compromise solution approach, considered both the total performance angle (S) and the regret stemming from the weakest criterion (R), revealing a contextually sensitive leadership shift, particularly between Greece and France. The shifting rankings of Germany and the Netherlands depending on the methodology indicate that these countries have strong trade-off structures between criteria. In contrast, Türkiye’s consistent last-place ranking in all three methods suggests that the results obtained are robust regardless of the methodology, and that Türkiye’s multidimensional structural disadvantages are evident regardless of the decision-making approach. While Germany is strong in utility criteria such as resilience and logistical capacity, it is at a disadvantage, particularly in the environmental impact criterion, which is a cost criterion. The negative environmental impact it creates due to being an industrialized country can be explained by its lower rankings in some methodologies. This comparative analysis reveals that COPRAS’s cost-oriented, TOPSIS’s balance-oriented, and VIKOR’s compromise-oriented structures offer complementary insights into complex problems such as sustainable food systems.

In conclusion, although COPRAS, TOPSIS, and VIKOR methods use different mathematical structures and decision logics, the methods produce a similar ranking logic and achieve a high degree of consistency between the methods, especially at the extremes (first and last ranks).

5.2 Comparison with prior studies

The findings of this study are consistent with previous research examining the sustainability and structural performance of agri-food systems at the national level. For example, FAO (2018) emphasizes that technological capacity, efficient resource management, and institutional agricultural policies are among the key drivers shaping sustainable food system performance across countries. Similarly, Garnett et al. (2013) highlight that differences in agricultural productivity and environmental management practices significantly influence the sustainability outcomes of national food systems. In the context of European agriculture, Pe’er et al. (2020) demonstrate that policy frameworks and technological innovation play a critical role in shaping agricultural sustainability and resource efficiency across EU countries.

Moreover, previous studies emphasize that the sustainability and resilience of food systems are influenced by a complex configuration of economic, institutional, and socio-environmental factors. In particular, Béné et al. (2020) argue that transforming food systems requires coordinated changes across agricultural production, supply chains, governance structures, and socio-economic contexts. Similarly, Béné et al. (2023) highlight that food systems consist of interconnected actors, institutions, infrastructures, and environmental resources that jointly shape food production, processing, distribution, and consumption processes. From this perspective, the resilience and performance of food systems emerge not only from the capacities of individual actors but also from the structural conditions and interactions within the system.

Within the European context, agricultural systems differ considerably across regions due to variations in environmental and socio-economic conditions. In particular, Mediterranean countries often display distinctive structural characteristics such as climatic constraints, fragmented agricultural land structures, and resource limitations (Pe’er et al., 2020). The results obtained in the present study are broadly consistent with these observations, as countries characterized by more technologically advanced agricultural systems and stronger institutional frameworks tend to perform better within the multi-criteria evaluation framework.

Overall, the findings reinforce the argument in the existing literature that the sustainability and performance of agri-food systems are shaped by a combination of technological capacity, policy environment, and structural agricultural characteristics. These findings contribute to the growing body of literature on agri-food system sustainability by providing a comparative multi-criteria assessment of national agricultural systems and by highlighting how structural, technological, and institutional differences across countries shape the sustainability performance of agri-food systems.

5.3 Policy implications

From a policy perspective, the integrated evaluation framework used in this study allows policymakers to identify priority areas for improvement within national agri-food systems. Rather than focusing on a single indicator, the approach highlights the need for balanced progress across multiple dimensions such as resource efficiency, environmental sustainability, and supply chain capacity. Such integrated insights may support more coherent policy strategies aimed at improving both the sustainability and resilience of food systems.

Examples of France and Greece reveal that countries with relatively low cost profiles can achieve a competitive position in sustainable food systems despite moderate benefit indicators. For Germany and the Netherlands, the findings show that high infrastructure and logistics capacity alone are insufficient; overall performance may remain limited unless supported by policies aimed at reducing environmental impact and external dependence. For Türkiye, the results point to the need for a multi-dimensional transformation in agro-food systems. It is understood that improving the country’s sustainability performance will be difficult unless strategies aimed at reducing import dependence, agricultural technologies that increase water efficiency, production models that reduce greenhouse gas emissions, and social policies targeting food insecurity are addressed simultaneously.

The impacts of climate change on agri-food supply chains vary by region, with high-latitude temperate regions (e.g., Northern Europe) being less affected during the production phase compared to lower-latitude tropical regions. Developed countries (Germany and the Netherlands), despite possessing advanced agricultural technologies, face higher production costs and food prices, and may also cause more environmental damage (more negative environmental impacts) as a negative consequence of industrialization. Developing countries, on the other hand, are burdened with low yields and worsening food security problems due to inadequate infrastructure. Indicating this situation, Türkiye’s ranking last in all three methods demonstrates a high degree of consistency among the methods. Consequently, many countries have launched relevant programs for sustainable agri-food supply (e.g., the EU’s “Farm to Table” Strategy and Low Carbon Agriculture Program). In Türkiye, the “Heirloom Seed” initiative has also been launched in this direction.

The six countries included in this study exhibit varying levels of development, policy priorities, and implementation performance in relation to SDGs directly related to agriculture and food systems: SDG 2 (Zero Hunger), SDG 6 (Clean Water and Sanitation), SDG 12 (Responsible Consumption and Production), and SDG 13 (Climate Action). Three countries located in the Mediterranean basin—Türkiye (ranking last in all three methods), Italy (ranking 5th), and Greece—share common structural characteristics such as vulnerability to climate change, water stress, and fragmented land ownership in agricultural production. These three countries need to make critical improvements, particularly in food security and sustainable agricultural production (under SDG 2), water use (under SDG 6), reducing food loss and waste (under SDG 12), and climate change adaptation policies (under SDG 13).

France, which ranks first in all three methods, and Germany, which ranks in the middle according to all three methods, are among the leading countries within the EU in terms of agricultural production volume, environmental regulations, and the implementation of sustainability policies. These countries are implementing strong practices in terms of circular economy applications, sustainable food supply chains, and the transformation of consumption habits under SDG 12; and in terms of reducing greenhouse gas emissions and low-carbon agriculture policies under SDG 13. In this respect, France and Germany are reference countries in terms of policy integration and implementation capacity for the SDG targets.

Despite its limited agricultural land, the Netherlands, which ranks in the middle according to the three methods, stands out globally as a model country for innovative agro-food systems aligned with the SDGs, thanks to its high productivity, advanced greenhouse technologies, and digital farming applications. Particularly in the context of resource efficiency, optimization of water and energy use, and food supply chain integration, the Netherlands is implementing best practices for the technology-driven application of SDGs 2 and 12.

In this context, the study’s findings highlight which indicators should be prioritized by policymakers and demonstrate that multi-criteria decision-making approaches can be used as an effective decision support tool in the design of national food and agricultural policies.

5.4 Limitations and future research

This study presents a comparative analysis of the sustainability and resilience performance of agri-food systems in representative European countries, but it has some limitations. First, the analysis was conducted for six countries and seven criteria. Although this country set was selected to reflect different agri-food system typologies within Europe, it does not cover the full diversity of European agri-food systems. Therefore, the findings should be interpreted as applying to a representative comparative sample rather than to all European countries. In addition, while the selected set of criteria (import dependence, household food waste, food-system greenhouse gas emissions, agricultural water withdrawal, food security, climate resilience, and logistics capacity) represents the sustainability-resilience axis, complementary dimensions such as biodiversity, soil health, nutritional quality, price/accessibility, agricultural productivity, energy dependence, and food loss at different supply chain stages (harvest, processing and retail) were not included. Several other factors may influence the sustainability and resilience of agri-food systems. In particular, soil quality plays a crucial role in long-term agricultural productivity and environmental sustainability. Similarly, geopolitical disruptions such as trade restrictions, regional conflicts, or supply chain interruptions can significantly affect food availability and price stability at national and international levels. Another critical dimension relates to food losses and waste occurring along the supply chain, including stages such as harvesting, storage, transportation, and trade. These factors have increasingly been emphasized in the food system sustainability literature as key determinants of system performance and resilience. However, integrating such dimensions into cross-country comparative frameworks remains methodologically challenging due to the limited availability of harmonized and consistently reported international datasets. Therefore, the results are limited to the scope of the selected criteria, and interpretations of “country performance” should be evaluated within this framework.

Future studies may expand the proposed analytical framework by incorporating criteria related to soil health, geopolitical risk exposure, and food loss across the supply chain as more comprehensive and comparable datasets become available. Addressing these additional dimensions would provide a more comprehensive understanding of agri-food system dynamics and represents an important direction for future research.

Secondly, the study presents a cross-sectional (static) comparison; the dynamic effects of policy changes, technological transformation, intensification of climate shocks, or trade disruptions over time are not directly modeled. Therefore, the findings represent a “snapshot of instantaneous performance” for a specific period/set of conditions.

Thirdly, the results of the Multi-Criteria Decision Making (MCDM) are sensitive to methodological preferences. In this study, weights were calculated objectively using Entropy and CRITIC, and consistency was strengthened by comparing the rankings with TOPSIS, VIKOR, and COPRAS. However, choices such as the normalization method, cost–benefit transformations, and VIKOR’s compromise parameter (v) can affect the results. Furthermore, since objective weighting is sensitive to variance in the data (e.g., Entropy suppressing low-variance criteria), it may reduce or increase the impact of specific criteria. Finally, the study results do not claim a causal relationship. The obtained rankings reflect “indicator-based relative performance” and require further analysis to verify the reasons behind inter-country differences.

Future research should include (i) expanding the scope of countries and, if possible, repeating the analyses with larger samples such as the EU-27; (ii) monitoring performance over the years using panel/dynamic MCDM approaches that include the time dimension; and (iii) modelling multiple shock conditions using scenario analysis (such as drought, energy price shock, trade constraint, logistics disruption) and stress tests. Furthermore, the scope of criteria could be broadened to include dimensions such as soil-biodiversity criteria, nutritional quality/health outcomes, price accessibility, agricultural productivity, energy intensity, and chain stages of food loss. Methodologically, it would be beneficial to test the weights not only with objective methods but also with expert opinion (AHP/BWM) or hybrid approaches. Finally, to strengthen the policy impact of MCDM outputs, validating the obtained rankings with causal inference approaches (e.g., with food security or emission outputs) will contribute to a more explanatory understanding of the sustainability-resilience interaction.

References

• 1

  ArvisJ. F.OjalaL.WiedererC.ShepherdB.RajA.DairabayevaK.et al. (2018). Connecting to Compete 2018: Trade Logistics in the Global Economy-The Logistics Performance index and its Indicators. Washington, DC: The International Bank for Reconstruction and Development/The World Bank.

  • Google Scholar

• 2

  AungkulanonP.AtthirawongW.LuangpaiboonP.ChanpuypetchW. (2024). Navigating supply chain resilience: a hybrid approach to Agri-food supplier selection. Mathematics12:1598. doi: 10.3390/math12101598

  • CrossRef
  • Google Scholar

• 3

  AwokuseT.LimS.SanteramoF.SteinbachS. (2024). Robust policy frameworks for strengthening the resilience and sustainability of Agri-food global value chains. Food Policy127:102714. doi: 10.1016/j.foodpol.2024.102714

  • CrossRef
  • Google Scholar

• 4

  AyyildizE. (2023). Interval valued intuitionistic fuzzy analytic hierarchy process-based green supply chain resilience evaluation methodology in post COVID-19 era. Environ. Sci. Pollut. Res.30, 42476–42494. doi: 10.1007/s11356-021-16972-y,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 5

  BarrettC. B. (2010). Measuring food insecurity. Science327, 825–828. doi: 10.1126/science.1182768

  • CrossRef
  • Google Scholar

• 6

  BénéC.FanzoJ.HaddadL.HawkesC.CaronP.VermeulenS.et al. (2020). Five priorities to operationalize the EAT–lancet commission report. Nat. Food1, 457–459. doi: 10.1038/s43016-020-0136-4,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 7

  BénéC.FrankenbergerT. R.NelsonS.ConstasM. A.CollinsG.LangworthyM.et al. (2023). Food system resilience measurement: principles, framework and caveats: Béné, Frankenberger, Nelson, Constas, Collins, Langworthy and Fox. Food Secur.15, 1437–1458. doi: 10.1007/s12571-023-01407-y

  • CrossRef
  • Google Scholar

• 8

  BeskeP.LandA.SeuringS. (2014). Sustainable supply chain management practices and dynamic capabilities in the food industry: a critical analysis of the literature. Int. J. Prod. Econ.152, 131–143. doi: 10.1016/j.ijpe.2013.12.026

  • CrossRef
  • Google Scholar

• 9

  CallensK.FontaineF.SanzY.BogdanskiA.LangeL.SmidtH.et al. (2022). Microbiome-based solutions to address new and existing threats to food security, nutrition, health and agrifood systems’ sustainability. Front. Sust. Food Syst.6:1047765. doi: 10.3389/fsufs.2022.1047765

  • CrossRef
  • Google Scholar

• 10

  Castillo-DiazF. J.Belmonte-UrenaL. J.Lopez-SerranoM. J.Camacho-FerreF. (2023). Assessment of the sustainability of the European Agri-food sector in the context of the circular economy. Sust. Prod. Consumpt.40, 398–411. doi: 10.1016/j.spc.2023.07.010

  • CrossRef
  • Google Scholar

• 11

  ChabouhS.SidhomL.ZammitiA.MamiA. (2024). “Assessing Agri-food supply chain multi-capital sustainability using Simple Multi-Attribute Rating Technique: the policy maker case study,” in 10th International Food Operations & Processing Simulation Workshop, FOODOPS.

  • Google Scholar

• 12

  ChatterjeeP.ChakrabortyS. (2016). A comparative analysis of VIKOR method and its variants. Dec. Sci. Lett.5, 469–486. doi: 10.5267/j.dsl.2016.5.004

  • CrossRef
  • Google Scholar

• 13

  ChenP. (2019). Effects of normalization on the entropy-based TOPSIS method. Expert Syst. Appl.136, 33–41. doi: 10.1016/j.eswa.2019.06.035

  • CrossRef
  • Google Scholar

• 14

  ChenX.QianX.WangY.JiaF. (2025). Enhancing sustainability in Agri-food supply chains amid climate change: research frontiers and future directions. Int J Log Res Appl1-44, 1–44. doi: 10.1080/13675567.2025.2576522

  • CrossRef
  • Google Scholar

• 15

  ClappJ. (2017). Food self-sufficiency: making sense of it, and when it makes sense. Food Policy66, 88–96. doi: 10.1016/j.foodpol.2016.12.001

  • CrossRef
  • Google Scholar

• 16

  ColucciaB.TunnoV.AgnusdeiG. P. (2025). The agricultural regeneration of Salento (Apulia, Italy) after the Xylella fastidiosa crisis: managing the shocks through multi-criteria decision-making methods. Sustainability17:8812. doi: 10.3390/su17198812

  • CrossRef
  • Google Scholar

• 17

  DerissenS.QuaasM. F.BaumgärtnerS. (2011). The relationship between resilience and sustainability of ecological-economic systems. Ecol. Econ.70, 1121–1128. doi: 10.1016/j.ecolecon.2011.01.003

  • CrossRef
  • Google Scholar

• 18

  EDGAR-FOOD. (2025). Available online at: https://edgar.jrc.ec.europa.eu/report_2025 (accessed November 2, 2025).

  • Google Scholar

• 19

  FAO (2008). An Introduction to the Basic Concepts of Food Security. Rome: FAO Food Security Programme.

  • Google Scholar

• 20

  FAO (2018). Sustainable Food Systems-Concept and Framework. Rome: Food and Agriculture Organization of the United Nations.

  • Google Scholar

• 21

  FAO (2021a). The State of Food Security and Nutrition in the World 2021: Transforming Food Systems for Food Security, Improved Nutrition and Affordable Healthy Diets for all. Rome: Food and Agriculture Organization of the United Nations.

  • Google Scholar

• 22

  FAO (2021b). The State of Food and Agriculture 2021: Making Agrifood Systems more Resilient to Shocks and Stresses. Rome: Food and Agriculture Organization of the United Nations.

  • Google Scholar

• 23

  FAO (2025). AQUASTAT database. Rome: FAO.

  • Google Scholar

• 24

  FAO, IFAD, UNICEF, WFP, and WHO (2025). The State of Food Security and Nutrition in the World 2025 – Addressing high Food price Inflation for Food Security and Nutrition. Rome: FAO, IFAD, UNICEF, WFP and WHO.

  • Google Scholar

• 25

  FAOSTAT (2025). Food and Agriculture Organization of the United Nations. Rome: FAOSTAT.

  • Google Scholar

• 26

  FarooqM. (2023). Conservation agriculture and sustainable development goals. Pakistan J. Agric. Sci.60, 291–298. doi: 10.21162/PAKJAS/23.170

  • CrossRef
  • Google Scholar

• 27

  FathiM. R.ZamanianA.KhosraviA. (2024). Mathematical modeling for sustainable agri-food supply chain. Environ. Dev. Sustain.26, 6879–6912. doi: 10.1007/s10668-023-02992-w

  • CrossRef
  • Google Scholar

• 28

  GarnettT. (2013). Food sustainability: problems, perspectives and solutions. Proc. Nutr. Soc.72, 29–39. doi: 10.1017/S0029665112002947

  • CrossRef
  • Google Scholar

• 29

  GarnettT.ApplebyM. C.BalmfordA.BatemanI. J.BentonT. G.BloomerP.et al. (2013). Sustainable intensification in agriculture: premises and policies. Science341, 33–34. doi: 10.1126/science.1234485,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 30

  Gésan-GuiziouG.AlaphilippeA.AubinJ.BockstallerC.BoutrouR.BucheP.et al. (2020). Diversity and potentiality of multi-criteria decision analysis methods for Agri-food research. Agron. Sust. Dev.40:44. doi: 10.1007/s13593-020-00650-3

  • CrossRef
  • Google Scholar

• 31

  GustavssonJ.CederbergC.SonessonU.Van OtterdijkR.MeybeckA. (2011). Global Food Losses and Food Waste: Extent, Causes and Prevention. Rome: FAO.

  • Google Scholar

• 32

  HaileslassieA.CraufurdP.ThiagarajahR.KumarS.WhitbreadA.RathorA.et al. (2016). Empirical evaluation of sustainability of divergent farms in the dryland farming systems of India. Ecol. Indic.60, 710–723. doi: 10.1016/j.ecolind.2015.08.014

  • CrossRef
  • Google Scholar

• 33

  HoekstraA. Y.MekonnenM. M. (2012). The water footprint of humanity. Proc. Nat. Acad. Sci.109, 3232–3237. doi: 10.1073/pnas.1109936109,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 34

  HyzA. (2024) “Conclusions: Charting the Course Forward,” in The Role of the Public Sector in Building Social and Economic Resilience: a Public Finance Approach, 339–343.

  • Google Scholar

• 35

  IPCC (2022). Climate Change 2022: Impacts, Adaptation and Vulnerability. Cambridge: Cambridge University Press.

  • Google Scholar

• 36

  JoshiS.SharmaM.EkrenB. Y.KazancogluY.LuthraS.PrasadM. (2023). Assessing supply chain innovations for building resilient food supply chains: an emerging economy perspective. Sustainability15:4924. doi: 10.3390/su15064924

  • CrossRef
  • Google Scholar

• 37

  KrishnanA. R. (2025). Research trends in criteria importance through intercriteria correlation (CRITIC) method: a visual analysis of bibliographic data using the tableau software. Inf. Disc. Delivery53, 233–247. doi: 10.1108/IDD-02-2024-0030

  • CrossRef
  • Google Scholar

• 38

  KrstićM.AgnusdeiG. P.TadićS.MigliettaP. P. (2023). Prioritization of e-traceability drivers in the Agri-food supply chains. Agric. Food Econ.11:42. doi: 10.1186/s40100-023-00284-5

  • CrossRef
  • Google Scholar

• 39

  KumarM.RautR. D.SharmaM.ChoubeyV. K.PaulS. K. (2022). Enablers for resilience and pandemic preparedness in food supply chain. Oper. Manag. Res.15, 1198–1223. doi: 10.1007/s12063-022-00272-w

  • CrossRef
  • Google Scholar

• 40

  LamineC. (2015). Sustainability and resilience in agrifood systems: reconnecting agriculture, food and the environment. Sociol. Ruralis55, 41–61. doi: 10.1111/soru.12061

  • CrossRef
  • Google Scholar

• 41

  LewA. A.NgP. T.NiC. C.WuT. C. (2016). Community sustainability and resilience: similarities, differences and indicators. Tour. Geograph.18, 18–27. doi: 10.1080/14616688.2015.1122664

  • CrossRef
  • Google Scholar

• 42

  LiuJ.WinklerJ. A.RossR. B.ViñaA.FrankK. A.KonarM.et al. (2025). Building sustainable and resilient Agri-food systems under multiple shocks. Front. Sust. Food Syst.9:1690853. doi: 10.3389/fsufs.2025.1690853

  • CrossRef
  • Google Scholar

• 43

  MadanchianM.TaherdoostH. (2023). A comprehensive guide to the TOPSIS method for multi-criteria decision making. Sust. Soc. Dev.1:2220. doi: 10.54517/ssd.v1i1.2220

  • CrossRef
  • Google Scholar

• 44

  MahmudovaS. (2020). Application of the TOPSİS method to improve software efficiency and to optimize its management. Soft. Comput.24, 697–708. doi: 10.1007/s00500-019-04549-4

  • CrossRef
  • Google Scholar

• 45

  ManglaS. K.KazançoğluY.YıldızbaşıA.ÖztürkC.ÇalıkA. (2022). A conceptual framework for blockchain-based sustainable supply chain and evaluating implementation barriers: a case of the tea supply chain. Bus. Strat. Environ.31, 3693–3716. doi: 10.1002/bse.3027

  • CrossRef
  • Google Scholar

• 46

  MekonnenM. M.HoekstraA. Y. (2014). Water footprint benchmarks for crop production: a first global assessment. Ecol. Indic.46, 214–223. doi: 10.1016/j.ecolind.2014.06.013

  • CrossRef
  • Google Scholar

• 47

  Muyulema-AllaicaJ. C.Menéndez-ZarumaC. M.Balseca-CastroJ. E.Aguirre-FloresF. X. (2025). Hybrid AHP-DEMATEL model for prioritizing key resilience and sustainability drivers and controllers in Agri-food supply chains. J. Eur. Syst. Automatisés58, 841–852. doi: 10.18280/jesa.580418

  • CrossRef
  • Google Scholar

• 48

  ND-GAIN (2025), ND-GAIN Country Index Database. Available online at: https://gain.nd.edu/our-work/country-index/download-data/ (accessed November 3, 2025).

  • Google Scholar

• 49

  OECD (2021). Making Better Policies for Food Systems. Paris: OECD Publishing.

  • Google Scholar

• 50

  Paredes-RodríguezA. M.Orejuela-CabreraJ. P.Osorio-GómezJ. C. (2024). Integrating sustainability and resilience in Agri-food supply chains. Int. J. Sust. Eng.17, 1122–1138. doi: 10.1080/19397038.2024.2430514

  • CrossRef
  • Google Scholar

• 51

  Pe’erG.BonnA.BruelheideH.DiekerP.EisenhauerN.FeindtP. H.et al. (2020). Action needed for the EU common agricultural policy to address sustainability challenges. People Nat.2, 305–316. doi: 10.1002/pan3.10080,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 52

  PooreJ.NemecekT. (2018). Reducing food’s environmental impacts through producers and consumers. Science360, 987–992. doi: 10.1126/science.aaq0216

  • CrossRef
  • Google Scholar

• 53

  PopescuG. C.PopescuM.PampanaS.KhondkerM.UmeharaM.HayashiH.et al. (2023). Introduction: sustainability as an agroecological strategy toward resilience in agricultural systems. Agron. J.115, 2657–2664. doi: 10.1002/agj2.21483

  • CrossRef
  • Google Scholar

• 54

  PrincipatoL.MattiaG.Di LeoA.PratesiC. A. (2021). The household wasteful behaviour framework: a systematic review of consumer food waste. Ind. Marketing Manage.93, 641–649. doi: 10.1016/j.indmarman.2020.07.010

  • CrossRef
  • Google Scholar

• 55

  PumaM. J.BoseS.ChonS. Y.CookB. I. (2015). Assessing the evolving fragility of the global food system. Environ. Res. Letters10, 1–14. doi: 10.1088/1748-9326/10/2/024007

  • CrossRef
  • Google Scholar

• 56

  RamosE.RabieeM.TareiP. K.ChavezM.ColesP. S. (2025). A diverse, unbiased group decision-making framework for assessing drivers of the circular economy and resilience in an Agri-food supply chain. Prod. Planning Control36, 1453–1473. doi: 10.1080/09537287.2024.2370988

  • CrossRef
  • Google Scholar

• 57

  Reig-MartínezE.Gómez-LimónJ. A.Picazo-TadeoA. J. (2011). Ranking farms with a composite indicator of sustainability. Agric. Econ.42, 561–575. doi: 10.1111/j.1574-0862.2011.00536.x

  • CrossRef
  • Google Scholar

• 58

  SaqlainM.KumamP.KumamW. (2025). Optimizing agricultural decision-making with integrated MCDM-MCDA methods: a case study on crop economics. Yugoslav J. Oper. Res.35, 857–874. doi: 10.2298/YJOR240915008S

  • CrossRef
  • Google Scholar

• 59

  SciortinoC.GiamporcaroG.SgroiF.ModicaF. (2025). Exploring the role of short food supply chains in agricultural sustainability and resilience: a literature review. Agric. Food Econ.13, 1–21. doi: 10.1186/s40100-025-00420-3

  • CrossRef
  • Google Scholar

• 60

  ScownM. W. (2024). “Sustainability and resilience for riverine landscapes,” in Resilience and Riverine Landscapes, eds. ThomsM.FullerI. (New York, NY: Elsevier), 287–303.

  • Google Scholar

• 61

  SharmaM.AntonyR.TsagarakisK. (2025). Green, resilient, agile, and sustainable fresh food supply chain enablers: evidence from India. Annal. Oper. Res.347, 13–39. doi: 10.1007/s10479-023-05176-x,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 62

  SharmaD.SridharS.ClaudioD. (2020). Comparison of AHP-TOPSIS and AHP-AHP methods in multi-criteria decision-making problems. Int. J. Ind. Syst. Eng.34, 203–223. doi: 10.1504/IJISE.2020.105291

  • CrossRef
  • Google Scholar

• 63

  SinghR.DwivediG. (2025). Resilience in Agri-food supply chains: a framework for risk assessment and strategy development. Int J Log Res Appl28, 1659–1690. doi: 10.1080/13675567.2024.2389050

  • CrossRef
  • Google Scholar

• 64

  StanescuS. G.IonescuC. A.ȘtefanM. C.IonescuL.BondacG. T.CristeaA. M. (2025). Digitalization and blockchain integration in Agri-food supply chains: towards a resilient, circular, and sustainable future. Sustainability17, 1–27. doi: 10.3390/su17209276

  • CrossRef
  • Google Scholar

• 65

  TainterJ. A.TaylorT. G. (2014). Complexity, problem-solving, sustainability and resilience. Building Res. Inf.42, 168–181. doi: 10.1080/09613218.2014.850599

  • CrossRef
  • Google Scholar

• 66

  TalukderB.HipelK. W.vanLoonG. W. (2018). Using multi-criteria decision analysis for assessing sustainability of agricultural systems. Sust. Dev.26, 781–799. doi: 10.1002/sd.1848

  • CrossRef
  • Google Scholar

• 67

  TendallD. M.JoerinJ.KopainskyB.EdwardsP.ShreckA.LeQ. B.et al. (2015). Food system resilience: defining the concept. Glob. Food Sec.6, 17–23. doi: 10.1016/j.gfs.2015.08.001

  • CrossRef
  • Google Scholar

• 68

  ThakkarJ. J. (2021a). “Technique for order preference and similarity to ideal solution (TOPSIS),” in Multi-Criteria Decision Making, ed. ThakkarJ. J. (Singapore: Springer Singapore), 83–91.

  • Google Scholar

• 69

  ThakkarJ. J. (2021b). “VIseKriterijumska Optimizacija I Kompromisno Resenje (VIKOR),” in Multi-Criteria Decision Making, ed. ThakkarJ. J. (Singapore: Springer Singapore), 129–138.

  • Google Scholar

• 70

  TomasielloS.AlijaniZ. (2021). Fuzzy-based approaches for Agri-food supply chains: a mini-review. Soft. Comput.25, 7479–7492. doi: 10.1007/s00500-021-05707-3

  • CrossRef
  • Google Scholar

• 71

  UNEP (2024). Food Waste Index Report 2024. Nairobi: United Nations Environment Programme.

  • Google Scholar

• 72

  Van der PloegJ. D.BarjolleD.BruilJ.BrunoriG.MadureiraL. M. C.DesseinJ.et al. (2019). The economic potential of agroecology: empirical evidence from Europe. J. Rural Stu.71, 46–61. doi: 10.1016/j.jrurstud.2019.09.003

  • CrossRef
  • Google Scholar

• 73

  VermaB. P.JonesJ. W.MigliaccioK.MoodyL.MadramootooC. A. (2021). The TFACS initiative: transforming food and agriculture to circular systems: envisioning multi-society objectives and initiatives. Res. Magazine28, 15–19.

  • Google Scholar

• 74

  WheelerT.Von BraunJ. (2013). Climate change impacts on global food security. Science341, 508–513. doi: 10.1126/science.1239402

  • CrossRef
  • Google Scholar

• 75

  WollniM.BohnS.Ocampo-ArizaC.PazB.SantaluciaS.SquarcinaM.et al. (2025). Sustainability standards in Agri-food value chains: impacts and trade-offs for smallholder farmers. Agric. Econ.56, 373–389. doi: 10.1111/agec.70005

  • CrossRef
  • Google Scholar

• 76

  World Bank (2025). Logistics Performance Index Database. Available online at: https://data.worldbank.org/indicator/LP.LPI.OVRL.XQ (accessed November 3, 2025).

  • Google Scholar

• 77

  ZavadskasE. K.PodvezkoV. (2016). Integrated determination of objective criteria weights in MCDM. Int. J. Inf. Technol. Dec. Making15, 267–283. doi: 10.1142/S0219622016500036

  • CrossRef
  • Google Scholar

• 78

  Zavala-AlcívarA.VerdechoM. J.Alfaro-SaizJ. J. (2020). “Resilient strategies and sustainability in agri-food supply chains in the face of high-risk events,” in Working Conference on Virtual Enterprises (560–570). Cham: Springer International Publishing.

  • Google Scholar

• 79

  ZhangM.YangJ. (2025). Agri-food supply chain resilience: an exploration of influencing factors based on fuzzy-DEMATEL-ISM analysis. PLoS One20:e0338492. doi: 10.1371/journal.pone.0338492,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 80

  ZhuY.TianD.YanF. (2020). Effectiveness of entropy weight method in decision-making. Mathematical Prob. Eng.2020, 1–5. doi: 10.1155/2020/3564835

  • CrossRef
  • Google Scholar

• 81

  ŽižovićM.MiljkovićB.MarinkovićD. (2020). Objective methods for determining criteria weight coefficients: a modification of the CRITIC method. Decision Making Appl. Manage. Eng.3, 149–161. doi: 10.31181/dmame2003149z

  • CrossRef
  • Google Scholar