The Agricultural Regeneration of Salento (Apulia, Italy) After the Xylella fastidiosa Crisis: Managing the Shocks Through Multi-Criteria Decision-Making Methods

Abstract

In recent years, agriculture has increasingly faced shocks related to climate change, pathogen outbreaks, and geopolitical instability, highlighting the need for sustainable regeneration strategies. This study develops an innovative Multi-Criteria Decision-Making (MCDM) framework that integrates the Delphi method, the Analytic Network Process (ANP), and the Aggregated Decision-Making (ADAM) method—the first application of this combination in the context of agricultural regeneration. The framework was applied to the Apulia region (Italy), heavily affected by the Xylella fastidiosa epidemic, and evaluated alternative crops across 30 economic, environmental, and socio-cultural sub-criteria. Results indicate that carob, walnut, and pistachio outperform other options by combining strong economic viability, climate resilience, and cultural compatibility. To mitigate the risks of monoculture, crop diversification strategies based on high-ranked alternatives are recommended. Sensitivity analysis confirmed the robustness of results, and the framework demonstrates high scalability, offering a transparent tool for policymakers in regions facing similar agricultural crises.

1. Introduction

In recent decades, the agricultural sector has encountered a growing array of shocks arising from climate change, emerging pathogens and geopolitical conflicts [1,2,3]. Rising temperatures and shifting rainfall patterns threaten crop yields and livestock productivity, while the spread of novel diseases poses risks to both plant health and food security [4,5]. Recent geopolitical instabilities and COVID-19 pandemic have disrupted global supply chains, affecting the import and export of key agricultural inputs and heightening price volatility [6,7]. Such shocks exacerbate existing weaknesses in rural areas, where infrastructure and local economies often lack the resilience to respond promptly to crises [8].

Current research highlights that agricultural resilience depends on multiple and interrelated factors, including crop and income diversification, sustainable land-use and circular economy practices, the adoption of technological and organizational innovations, and the role of participatory and multi-level governance in aligning scientific knowledge with local needs [9]. Within this debate, the concept of agricultural regeneration has emerged as a transformative response, aiming not only to restore production systems aftershocks but also to redesign them toward long-term sustainability and socio-ecological balance [10]. Against this backdrop, rural regeneration has gained prominence as a means of strengthening local systems through strategies that balance environmental, economic, and social objectives [11,12]. The Common Agricultural Policy (CAP) has encouraged diversified cropping models, landscape conservation, and social inclusion, thus providing mechanisms to absorb shocks and sustain rural livelihoods [13,14]. Parallel initiatives, such as the European Green Deal and the Farm to Fork Strategy, have further underscored the importance of resource efficiency and circular approaches in agricultural and agri-food practices [15,16].

In Mediterranean rural contexts, regeneration involves a combination of land-use decisions, crop selection, landscape enhancement, and socio-economic revitalization, all of which need to align with long-term sustainability goals [17,18]. Participatory governance and collaborative planning are especially relevant, as they integrate scientific insights, technical expertise, and local community needs [19,20]. This multi-actor approach gains particular importance where monocropping has made extensive areas vulnerable to systemic risks [21]. In the case of regions facing phytosanitary or environmental crises, “post-crisis regeneration” represents an opportunity to redefine the production paradigm: on the one hand, through the reconversion or introduction of resilient and high value-added crops; on the other, through landscape requalification and the adoption of circular practices [22,23]. Such strategies, if supported by adequate territorial branding policies and financial support [24,25], can help boost local employment and preserve cultural heritage.

This is the context in which this contribution is set, aimed at developing and validating an innovative methodological approach—based on Multi-Criteria Decision Making (MCDM)—with the aim of identifying agri-food chains capable of promoting a sustainable regeneration process, balancing economic, landscape–environmental and social criteria.

1.1. Background and Scope

Within the broader framework of rural regeneration and resilience-building strategies, the case of the Apulia region, in Italy, represents a particularly relevant example of the complex challenges posed by phytosanitary emergencies. The Apulia region, historically known for its vast expanses of olive groves, has long been one of the leading global hubs for olive oil production, contributing approximately 40% of Italy’s total output [26]. Beyond its economic significance, olive cultivation has shaped the region’s cultural heritage, rural landscapes, and socio-economic fabric. However, since the first detection of Xylella fastidiosa subspecies pauca in Salento in 2013, Apulia has faced one of the most severe phytosanitary crises in European history, leading to an irreversible transformation of its agricultural sector [27,28].

The rapid spread of the pathogen has caused the progressive destruction of olive groves, impacting not only agricultural production but also tourism, biodiversity, and soil stability [29]. Recent studies estimate that the Apulian region has already lost more than 21 million olive trees, including thousands of monumental specimens, radically altering the rural landscape [30]. This transformation has reduced the attractiveness of Salento’s traditional “olive grove landscape,” which previously represented a cornerstone of cultural tourism. From an ecological perspective, the widespread tree mortality has decreased habitat availability for numerous bird and insect species, while also reducing canopy cover and organic matter inputs, with negative consequences for soil fertility and erosion control [31]. These impacts confirm that the epidemic is not only a local agricultural issue but a regional crisis with far-reaching socio-economic and environmental repercussions. The magnitude of the crisis has led to an urgent call for action, with national and regional authorities implementing various measures to contain the pathogen and promote agricultural recovery. The primary strategy has focused on replanting Xylella fastidiosa-resistant olive cultivars, such as Leccino and FS-17 Favolosa, in an attempt to restore olive oil production in the region [32]. However, while this approach ensures continuity in olive farming, it does not fully address the broader economic, environmental, and social challenges posed by the epidemic. The transition from a monoculture-based agricultural system to a more diversified and resilient model has emerged as a critical objective for ensuring the long-term sustainability of the Apulian agro-food sector.

In the broader literature, resilience in agriculture is often defined within the framework of social-ecological systems (SESs). It refers to the capacity of farming systems to absorb shocks, adapt to changing conditions, and, when necessary, transform into more sustainable configurations [33]. From this perspective, agriculture is understood as part of complex adaptive systems, where ecological processes and social institutions are deeply interconnected. Closely linked to this, agricultural regeneration has been increasingly conceptualized through transition theory, which views systemic change as a multi-level process that reorients socio-technical regimes towards sustainability [34]. Regeneration therefore goes beyond restoring pre-crisis conditions: it involves reconfiguring production models, land-use practices, and governance arrangements to promote long-term sustainability, resilience, and socio-economic revitalization. Agricultural regeneration strategies following catastrophic events have been widely explored in other global contexts, where regions affected by environmental shocks, plant diseases, or economic disruptions have implemented diverse recovery measures.

For instance, in California, where Huanglongbing (HLB) disease has devastated citrus orchards, researchers have examined crop diversification as a means of reducing economic vulnerability and enhancing ecosystem resilience [35]. Similarly, in Spain, where prolonged droughts have led to the decline of traditional olive and citrus farming, alternative crops such as almonds and pistachios have been introduced to adapt to changing climatic conditions [36]. In Latin America, the widespread destruction of banana plantations due to Panama disease has prompted research into sustainable crop substitution models that integrate economic, environmental, and social factors [37]. These international experiences highlight that effective agricultural regeneration requires a multidimensional approach, where agronomic and ecological measures are combined with socio-economic strategies and the preservation of cultural heritage. In practice, most regions have relied on a mix of crop substitution, targeted policy support, and community participation, with the balance among these components varying according to local conditions. Positioning the Apulia case within this broader landscape not only underscores its unique challenges but also shows how the methodological framework proposed here can be transferred to other contexts facing agricultural crises. Recent studies in the region have emphasized the need to explore alternative agro-food supply chains that go beyond olive replanting and introduce a more sustainable and diversified agricultural model. Various crops, such as almonds, pomegranates, kiwis, and figs, have been identified as potential alternatives due to their resilience to Xylella fastidiosa and their adaptability to the region’s climatic and soil conditions [38]. Additionally, researchers have begun to investigate the role of circular economy principles in enhancing agricultural regeneration, with studies exploring the valorization of by-products from new crops for bioactive compound extraction, animal feed production, and sustainable packaging materials [39]. However, despite the increasing focus on diversification, existing research tends to assess potential crops and supply chains based on either agronomic or economic parameters, without integrating these factors into a comprehensive decision-making framework that considers environmental sustainability, market potential, and socio-cultural acceptance. To address this gap, this study proposes a structured decision-making framework that evaluates and prioritizes alternative agro-food supply chains based on a multi-criteria approach.

Specifically, by integrating the Analytic Hierarchy Process (AHP), Analytic Network Process (ANP), and the Aggregated Decision-Making (ADAM) methodology, this research aims to identify the most viable options for regenerating the Apulian agricultural sector in a way that ensures economic resilience, environmental sustainability, and social acceptability. Moreover, through a participatory approach involving stakeholders such as farmers, policymakers, and local communities, the study seeks to align proposed regeneration strategies with the needs and perspectives of those most directly affected by the crisis. By adopting a comprehensive and systematic evaluation process, this research contributes to the ongoing discourse on post-Xylella fastidiosa regeneration in Apulia, while also providing a replicable model for other regions facing similar agricultural crises. The findings are expected to inform policy decisions at the local and national levels, guiding public funding and investment strategies toward sustainable and resilient agricultural solutions that can effectively respond to environmental and economic shocks.

1.2. Literature Review

Recent literature underscores the urgency of exploring alternative and innovative agro-food supply chains in the Salento region in response to the devastation caused by Xylella fastidiosa. The epidemic has severely impacted olive cultivation, historically the backbone of the local economy, necessitating diversification and regeneration of agricultural practices to ensure resilience and long-term sustainability [40]. Initial studies predominantly focused on the biological and epidemiological aspects of Xylella fastidiosa, examining its spread and the viability of replanting resistant olive varieties such as Leccino and Favolosa (F17). Research by Saponari et al. [41] demonstrated that these cultivars exhibit significant tolerance to Xylella fastidiosa, making them promising candidates for replanting efforts. However, as the epidemic evolved, the scope of research expanded beyond biological resilience to include socio-economic considerations, emphasizing the need for sustainable agricultural models capable of ensuring food security and economic stability [41,42]. Specifically, recent contributions point to agroecological diversification strategies, circular bioeconomy approaches, and integrated multi-crop systems as viable alternatives to monoculture. Within this literature, stability is understood in a multidimensional sense: ecological stability, by reducing vulnerability to pests and climatic variability; economic stability, by diversifying farmers’ sources of income; and social stability, by sustaining employment and rural livelihoods in the long term [21].

This shift laid the foundation for investigations into alternative crops and diversified supply chains as viable solutions. One of the earliest strategies to encourage agricultural diversification involved the introduction of public funding to support the cultivation of alternative crops such as almonds, kiwis, pomegranates, and figs, all of which are well-adapted to the region’s climatic conditions and possess significant market potential [38]. Studies have analyzed the economic and environmental benefits of these crops, highlighting their ability to reduce dependence on monoculture and enhance biodiversity [43]. In parallel, the concept of circular economy has gained traction in the Salento agro-food sector, with studies demonstrating how by-products from new crops, such as pomegranate peels, can be repurposed for bioactive compound extraction or animal feed, creating additional revenue streams and minimizing agricultural waste [39]. In terms of methodological approaches, GIS-based analyses have been widely used to assess land suitability for alternative crops, integrating climatic and soil parameters to optimize planting locations [44].

Moreover, sustainability assessments employing Life Cycle Assessment (LCA) and Life Cycle Costing (LCC) methods have provided insights into the ecological and economic impact of agricultural diversification, particularly in relation to biodiversity conservation and resource efficiency [45]. Despite these advancements, several critical gaps remain in the literature. The transition to alternative supply chains requires substantial investments in infrastructure, farmer training, and market development, while also necessitating strong social acceptance and alignment with cultural values. A key shortcoming is the lack of a systematic framework to evaluate and prioritize diversification strategies based on multiple, and often conflicting, criteria. Globally, MCDM methods have been extensively applied to challenges in agricultural decision-making, including crop selection, supply chain optimization, and resource allocation [46,47].

Among the most widely applied methods, the AHP has often been used for crop selection and land-use planning because of its transparency and relative ease of implementation. However, its strictly hierarchical structure does not capture the interdependencies that frequently exist among economic, environmental, and social criteria. The ANP was developed to address this limitation, allowing for feedback and interconnections between factors, though at the cost of higher complexity and a heavier cognitive burden for experts involved [48]. Other techniques, such as TOPSIS (Technique for Order of Preference by Similarity to Ideal Solution), have been employed in contexts such as disaster recovery and resource allocation, where they provide intuitive rankings but show sensitivity to data normalization and indicator scaling. PROMETHEE (Preference Ranking Organization Method for Enrichment Evaluation) has also been applied to agricultural planning problems, offering flexibility and robustness, but its outranking logic is often less intuitive for stakeholders with limited technical expertise [49]. This range of applications highlights both the potential and the methodological challenges of using MCDM in agriculture and helps to justify the integration of ANP and ADAM in this study, which seeks to combine the capacity to model interdependencies with a transparent and geometrically robust aggregation procedure. However, their application in the context of Xylella fastidiosa and the transition of the Salento agro-food sector remains limited.

Specifically, the existing literature does not sufficiently integrate socio-economic, environmental, and cultural factors into a comprehensive decision-making model for selecting alternative supply chains, nor does it employ participatory MCDM approaches to incorporate the preferences of diverse stakeholders such as farmers, policymakers, and researchers. Furthermore, current studies lack a robust comparative analysis of alternative crops or supply chains based on their resilience, market potential, and sustainability within the Salento context. This gap is particularly evident in the limited use of MCDM techniques to address the complex challenges associated with shifting away from olive monoculture. While some studies highlight the benefits of crop diversification and circular economy principles, they often rely on qualitative assessments or single-criterion evaluations, failing to capture the multidimensional complexity of real-world decision-making. Additionally, the scalability and long-term implications of these innovative supply chains remain largely unexplored, as existing models tend to overlook dynamic factors such as market fluctuations, climate change impacts, and evolving policy landscapes, all of which are critical for ensuring the resilience and sustainability of agro-food systems.

Addressing these gaps, the present study aims to develop a comprehensive MCDM framework tailored to the Salento region, integrating socio-economic, environmental, and cultural criteria to systematically evaluate and prioritize diversification strategies. By employing participatory approaches, this research will ensure that the perspectives of key stakeholders are incorporated into the decision-making process, fostering greater acceptance and feasibility of proposed solutions. Furthermore, it will provide a comparative analysis of alternative crops and supply chains, assessing their resilience, economic viability, and long-term sustainability. The application of MCDM techniques in this context represents a novel contribution to literature, offering a structured methodology to guide the transition of the Salento agro-food sector toward more diversified, resilient, and sustainable models in the wake of Xylella fastidiosa.

2. Methodological Strategy

To ensure a rigorous and structured assessment of potential alternatives to olive cultivation, the methodology was developed through a sequence of well-defined steps (Figure 1).

These steps progressively refined the evaluation framework, identified the most suitable crops, and ranked them based on their economic, environmental, and social sustainability.

The first phase focused on defining the criteria that would guide the evaluation. Since the selection of an alternative crop requires balancing multiple dimensions, the Delphi Method was used to identify the most relevant criteria through an expert-driven consensus process. A multidisciplinary panel participated in multiple iterative rounds, progressively refining the criteria based on structured feedback. The final set of criteria encompassed economic, environmental, and social factors, ensuring a comprehensive evaluation framework.

With these criteria in place, the second phase involved selecting potential crop alternatives. The expert panel reviewed a preliminary list of candidate species, derived from nursery data, market trends, and previous proposals for innovative agricultural practices. These crops were assessed based on their adaptability to Salento’s pedoclimatic conditions, economic feasibility, and environmental and social impact. Through a structured scoring process, where experts rated each crop against the established criteria, the initial list was refined to a final selection of ten crops.

The next step was to assign relative importance to the criteria and sub-criteria. Because different factors carry different weights and are often interdependent, the Analytic Network Process (ANP) was applied instead of a simple hierarchy. In this approach, experts conducted pairwise comparisons among the criteria and the sub-criteria, capturing both inner and outer dependencies within a network structure. By constructing an unweighted supermatrix, weighting the clusters, and subsequently raising the resulting weighted supermatrix to successive powers until convergence, the ANP provided a comprehensive quantification of priorities. This method ensured that the final ranking of alternative crops reflected a balanced and nuanced consideration of economic viability, environmental impact, and social benefit.

The fourth phase of the methodology was the ranking of alternative crops, performed using the ADAM method, a geometric multi-criteria decision-making approach. This method translated the evaluation scores into a three-dimensional model, where each crop was positioned within a mathematical space based on its performance in relation to the weighted criteria. By computing the volume of complex polyhedra formed by these data points, the ADAM method generated a final ranking, identifying the crop that achieved the highest aggregated score as the most suitable alternative to olive cultivation. This innovative method was chosen because it ensures a transparent geometric aggregation, reduces sensitivity to normalization and compensation effects, and thus provides a robust ranking for multidimensional criteria such as those applied in this study.

Lastly, to test the robustness of the findings, a sensitivity analysis was conducted. This step reinforced the methodological strength and practical relevance of the proposed MCDM approach. The integration of Delphi, ANP, and ADAM was specifically chosen to address the multidimensional and sometimes contradictory nature of the evaluation criteria: Delphi ensured consensus, ANP captured interdependencies, and ADAM provided a robust ranking mechanism.

2.1. Selection of Criteria

The selection of criteria to evaluate alternative crops to olive cultivation was conducted using the Delphi Method. This approach, recognized for its ability to synthesize diverse expert opinions and converge towards consensus through iterative feedback, was chosen for its robustness in developing evaluation frameworks for complex, multidimensional problems [50,51]. The goal of this phase was not to evaluate crops directly but to identify and validate the criteria that would later serve as the basis for a structured assessment of alternative crops. The process was conducted in three iterative rounds, allowing for the refinement and validation of the criteria based on input from a multidisciplinary panel of experts.

The panel consisted of 15 experts representing five Mediterranean countries (Italy, France, Greece, Algeria, and Tunisia), with three experts from each country. Their professional backgrounds covered Mediterranean agriculture, crop diversification, and sustainability. Participants included agronomists, plant nursery managers, agricultural consultants, and researchers, all selected to ensure a comprehensive representation of expertise. Eligibility required at least five years of professional experience or three years of interdisciplinary engagement in agriculture, with preference given to individuals who had contributed to crop innovation either through scientific research or practical application. This rigorous selection process followed best practices outlined in Delphi methodology studies [52,53], ensuring that the panel’s expertise was sufficient to address the complexity of the task.

In the first round, the experts were asked to propose criteria and sub-criteria they considered relevant for evaluating alternative crops. Using an open-ended questionnaire, qualitative insights were collected and synthesized into a preliminary list of 30 criteria and sub-criteria, which were grouped into three primary dimensions: economic, environmental/landscape, and social. In line with the Delphi approach, experts were explicitly asked to converge on an exhaustive list of 30 criteria considered most suitable for evaluating alternative crops, ensuring that the final set reflected their collective judgment. The open-ended nature of the first round is a fundamental characteristic of the Delphi Method, as it encourages the generation of a broad spectrum of ideas [54].

In the second round, experts were asked to rate the importance of each criterion on a five-point Likert scale, where 1 indicated low importance and 5 indicated high importance. Statistical analysis, including the calculation of medians and interquartile ranges (IQR), was used to assess the level of agreement among the panelists, following the standards established by von der Gracht [55]. Consensus was defined as 75% of responses falling within a single IQR. Criteria that did not meet this threshold were revised or excluded, resulting in a refined list of 25 criteria and sub-criteria. This process aligns with best practices in consensus-building exercises, which emphasize iterative rounds to refine and focus expert feedback [56].

The third round was dedicated to validating the refined criteria and achieving a high level of agreement among the experts. Participants were presented with the updated list and the statistical results from the second round, including median scores and the range of responses, and were invited to confirm their agreement or provide further input. The final framework achieved a Kendall’s W (coefficient of concordance) of 0.89, indicating a strong consensus among panelists. Kendall’s W was calculated using the formula:

$$W={\displaystyle \frac{12\cdot S}{m^2·(n^3- n)}}$$

where $S$ represents the sum of squared deviations of ranks from their mean, $m$ is the number of experts, and $n$ is the number of criteria evaluated. A high $W$ value reflects the robustness of the consensus-building process, as outlined by Kendall and Gibbons [57].

2.2. Selection of Alternatives

The selection of alternative crops was carried out through a structured process involving the same panel of experts who participated in the Delphi method (Section 3.1). This approach ensured continuity and consistency in the application of the multidimensional evaluation framework. After defining the criteria and sub-criteria, the group was reconvened to identify and validate specific crop species that align with the defined economic, environmental/landscape, and social dimensions.

The process unfolded as follows. First, the panel reviewed a preliminary list of non-traditional crops derived from regional nursery data, reflecting market demand, and proposals submitted in public funding calls for innovative cultivation and processing systems. Second, each crop was evaluated based on its alignment with the sub-criteria. Special attention was given to climatic adaptability, economic feasibility, environmental impact, and social acceptability. Scores were assigned using a scale from 1 (poor alignment with criteria) to 5 (excellent alignment with criteria). For each crop, the weighted average score for the sub-criteria was calculated, and a total score was obtained as the sum of these averages.

Third, the panel engaged in iterative discussions to address discrepancies in evaluations and refine the selection. This process ensured that the results were consistent and shared among all members. Finally, the level of agreement among experts was measured using Kendall’s coefficient of concordance ($W$).

2.3. Assigning Weights to Criteria and Sub-Criteria Through the ANP

The weights for the criteria and sub-criteria were determined using the ANP, an extension of the traditional Analytic Hierarchy Process (AHP) that accommodates both inner and outer dependencies among criteria and sub-criteria. A panel of 20 university professors specializing in agricultural economics, plant pathology, agronomy, planning, and forest ecology conducted pairwise comparisons to assess the relative influence among the factors, following the principles established by Saaty [58] and later expanded by Chung et al. [59]. The process began by defining a network model using SuperDecision software (version 2.2), which allowed us to represent interdependencies between clusters of factors, thereby capturing realistic mutual influences among criteria and sub-criteria. The experts used a standardized scale to express their judgments, ranging from “Equal Importance” (1) to “Extremely High” (9).

The results of these comparisons were then organized into an unweighted supermatrix, a block-partitioned matrix that aggregates the priority vectors corresponding to all possible influence relationships among the clusters. In general terms, if there are n clusters $C_1,C_1,\dots ,C_n$ within each cluster $k$ a set of elements $\{e_{k1},e_{k2},\dots ,e_{kn}$} the unweighted supermatrix $W$ is represented as

$$W=\left(\begin{array}{ccc}{W}_{11}& \dots & W_{1\mathrm{n}}\\ ⋮& \ddots & ⋮\\ W_{\mathrm{n}1}& \dots & W_{\mathrm{n}\mathrm{n}}\end{array}\right)$$

where each block $W_{ij}$ contains the priority vectors reflecting the influence of elements in cluster $i$ as affected by elements in cluster $j$. If no influence exists between two clusters, the corresponding block is a zero matrix.

To incorporate the relative importance of each cluster, additional pairwise comparisons were performed to determine weighting coefficients such that the sum of the cluster weights equals 1. These coefficients are organized in a diagonal matrix $D$ and used to transform the unweighted supermatrix into a weighted supermatrix $W_{\mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\mathrm{e}\mathrm{d}}$ as follows:

$$W_{\mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\mathrm{e}\mathrm{d}} = D \times W$$

where $D$ is a diagonal matrix whose entries are the weights of the clusters. To ensure that each column of the supermatrix is stochastic (i.e., sums to 1), normalization is applied. For each element $p_{ij}$ in a given column, the normalized value is calculated as:

$$r_{ij} = {\displaystyle \frac{p_{ij}}{\sum _{i=1}^m{p}_{ij}}},$$

and the weight vector for each sub-criterion is obtained by averaging the normalized values:

$$w_{j }= {\displaystyle \frac{1}{m}}{\sum }_{i=1}^m{r}_{ij }\mathrm{with} w_j\ge 0·{\sum }_{i=1}^m{w}_{j }$$

The final step involves raising the weighted supermatrix to successive powers until convergence is achieved, yielding the limit supermatrix:

$$W^{*}= \underset{k\to \infty }{\lim }{\left(1+W_{weighted}\right)}^k$$

The entries of the limit supermatrix $W^{\mathrm{*}}$ represent the final, stable weights of each subcriterion, encapsulating the global influence acquired through the network’s feedback loops. These final weights reflect the consensus of the expert panel, effectively translating qualitative assessments into quantitative measures for subsequent decision-making. In addition, to ensure the reliability of the pairwise comparisons, a Consistency Ratio (CR) was computed for each matrix, following the guidelines of Saaty [58]. A CR of 0.1 or lower was considered acceptable; if the CR exceeded this threshold, the judgments were revisited and refined to improve consistency [60,61].

2.4. Normalization, Aggregation, and Ranking Through the ADAM Method

The ADAM method belongs to the family of geometric multi-criteria decision-making (MCDM) techniques [49]. Its rationale is based on assessing alternatives by calculating the volumes of complex polyhedra constructed in a three-dimensional coordinate space. The vertices of these polyhedra are organized into three categories: the coordinate origin (O), the reference points (R), and the weighted reference points (P). The origin corresponds to the coordinates (0, 0, 0). Reference points lie in the x–y plane and are defined as (x, y, 0), representing the alternatives’ values depending on the four orientations of the considered criteria, with their distance from the origin capturing their relative position. Weighted reference points, instead, are expressed as (x, y, z), where the z-dimension incorporates the weights of the sub-criteria by measuring their distance from the x–y plane.

The volume of a complex polyhedron is obtained by summing the volumes of the m constituent polyhedra, where m corresponds to the number of alternatives analyzed. Once these volumes are computed, alternatives are ranked in descending order: the greater the volume associated with an alternative, the higher its position in the final ranking. The procedure for applying this method follows a structured sequence of steps.

Step 1: Construct the decision matrix $E$, whose generic element $e_{ij}$ represents the evaluation of alternative $i$ (A1, …, Am) with respect to sub-criterion $j.$ In other words, each element corresponds to the magnitude of the evaluation assigned to an alternative for a specific indicator:

$$E={\left[e_{ij}\right]}_{m\times n},$$

(1)

where $m$ is the number of alternatives and $n$ is the number of subcriteria.

Step 2: Arrange the decision matrix into a sorted matrix $S$, where the elements $e_{ij}$ are ordered in descending order according to the relative importance (weight) of the sub-criteria:

$$S={\left[s_{ij}\right]}_{m\times n},$$

(2)

Step 3: Normalize the sorted matrix to obtain the normalized matrix N. The normalized value $n_{ij}$ is computed differently depending on whether the sub-criterion is a benefit (set $B$) or a cost (set $C$):

$$n_{ij}=\left\{\begin{array}{c}{\displaystyle \frac{s_{ij}}{\underset{i}{\max }{s}_{ij}}}, forj\in B\\ {\displaystyle \frac{\underset{i}{\min }{s}_{ij}}{s_{ij}}}, forj\in C\end{array}\right.,$$

Step 4: Determine the coordinates ($x, y, z$) of the reference points $R_{ij}$ and the weighted reference points ($P_{ij}$) that define the polyhedron:

$$x_{ij}=n_{ij}\times \sin {\alpha }_j, \forall j=1,\dots ,n; \forall i=1,\dots ,m,$$

$$y_{ij}=n_{ij}\times \cos {\alpha }_j, \forall j=1,\dots ,n; \forall i=1,\dots ,m,$$

$$z_{ij}=\left\{\begin{array}{c}0, for{R}_{ij}\\ w_j, for{P}_{ij}\end{array}, \right.\forall j=1,\dots ,n; \forall i=1,\dots ,m,$$

where $w_j$ is the weight assigned to sub-criterion $j$ and ${\alpha }_j$ the angle that specifies the orientation of the vector.

Step 5: Calculate the volume of each complex polyhedron $V_i^C$ by summing the volumes of the pyramids composing it:

$$V_i^C=\sum _{k=1}^{n-1}{V}_k, \forall i=1,\dots ,m$$

Step 6: Establish the final ranking of the alternatives based on the descending order of the volumes $V_i^C$ ($i=1,\dots ,m$). The alternative associated with the largest volume is considered the most preferable.

2.5. Sensitivity Analysis

Within the framework of MCDM, sensitivity analysis serves to evaluate how stable and reliable the obtained results are when the weights assigned to criteria are altered [62]. This procedure makes it possible to understand whether modifications in the relative importance of criteria could produce different rankings or choices among the alternatives. In doing so, it highlights the extent to which each criterion contributes to the final outcome and helps to single out those with the strongest influence. By systematically adjusting the weights and observing the corresponding shifts in rankings, decision-makers gain clearer insights into possible trade-offs and can establish priorities more consistently. Sensitivity analysis, therefore, works as a safeguard against subjective judgments in weight assignment, promoting more transparent, balanced, and resilient decision-making processes [49].

In the present research, robustness checks were carried out by introducing three additional scenarios designed to test the effect of varying the relative weight of specific criteria. The analysis was applied individually to each of the ten criteria, producing three scenarios for every sub-criterion considered. In each case, the weight of the criterion identified as the most relevant was progressively reduced by 15%, 30%, and 60%. The portion of weight removed was then evenly redistributed across the remaining criteria, so that the overall sum of weights continued to equal 1. This structured procedure enabled a detailed exploration of how the most influential criteria affect the final evaluation, thus offering valuable insights into the robustness and stability of the obtained ranking when different weighting assumptions are introduced.

3. Results

3.1. Evaluation Criteria and Crop Alternatives

Based on the methodology outlined in Section 2, a structured framework for the evaluation of the most sustainable and strategic crop alternatives has been defined. The adopted model is divided into three main dimensions—economic, environmental/landscape and social—each of which includes ten specific sub-criteria that reflect the most relevant aspects for a holistic and multidimensional analysis of the crop choice (Figure 2).

The economic dimension included: (i) planting costs, (ii) maintenance costs, (iii) workforce costs, (iv) productivity, (v) market evaluation, (vi) national and foreign demand, (vii) added value, (viii) transformation potential, (ix) economic incentives, (x) territorial branding. The environmental/landscape dimension encompassed: (i) impact on the rural landscape, (ii) contribution to biodiversity, (iii) water consumption, (iv) required use of pesticides/phytopharmaceuticals, (v) delivery of ecosystem services, (vi) resilience to environmental shocks, (vii) adaptation to climate change, (viii) conservation of wildlife, (ix) impact on soil fertility, (x) pedoclimatic adaptability.

The social dimension comprised: (i) food security, (ii) food safety, (iii) cultural heritage preservation, (iv) attractiveness of innovative crops, (v) gastronomic innovation, (vi) employment creation, (vii) equitable value distribution along the supply chain, (viii) visual attractiveness for ecotourism purposes, (ix) mastery of agronomic techniques, (x) community involvement through cooperatives, consortia, and local markets.

Following the definition of the evaluation framework, and following the process described in Section 2.2, the ten alternative crops most suitable for replacing the olive tree in Salento were identified as follows: (i) avocado, (ii) papaya, (iii) mango, (iv) apple, (v) pistachio, (vi) walnut, (vii) chestnut, (viii) kiwi, (ix) citrus (lemon, tangerine, orange), (x) carob.

These alternatives represent a selection of non-traditional crops—that is, species not historically associated with Salento’s agricultural identity—but which have shown strong adaptability to the region’s soil and climatic conditions. Their inclusion reflects an innovative and forward-looking approach, aimed at diversifying local production while enhancing resilience and sustainability in response to current agronomic challenges.

3.2. Weighting of Criteria (ANP Results)

The results obtained through the ANP provided a clear prioritization of the most relevant sub-criteria identified for selecting alternative crops to olive cultivation (Figure 3).

Specifically, among the 30 sub-criteria assessed, food security (socio-cultural dimension) was ranked highest. This outcome reflects the experts’ focus on ensuring continuous, reliable access to sufficient and nutritious food in response to the disruption caused by the Xylella fastidiosa epidemic, which significantly compromised the traditional local agri-food systems. Ranked second was the maintenance costs criterion (economic dimension), defined as the expenses incurred by farmers to maintain optimal production conditions, including water management, fertilization, pesticide applications, and harvesting practices. The prominence assigned to this factor highlights the importance attributed by experts to the economic feasibility of alternative crops, particularly given the relatively low operational costs historically associated with olive cultivation. In third place, experts positioned the impact on the rural landscape (environmental dimension).

This criterion evaluates the potential transformative effects that new crops may have on areas historically dominated by olive groves. Such a finding underscores the significance attributed to preserving the landscape integrity, biodiversity, aesthetic qualities, and local cultural identity, reflecting the essential role landscapes play in Mediterranean rural contexts. The presence of these three sub-criteria, each belonging to a different macro-dimension (socio-cultural, economic, and environmental), emphasizes the necessity and value of adopting a multidimensional, balanced approach in assessing crop diversification strategies for the Apulian regional regeneration following the Xylella fastidiosa crisis.

3.3. Final Ranking of Crop Alternatives and Robustness Analysis

In order to identify the best alternative, i.e., the most suitable cultivation to implement in order to achieve landscape, economic, and environmental regeneration objectives, based on the identified subcriteria, the ADAM method has been applied. The alternatives are assigned a final ranking, arranged in descending order according to the volume values of the corresponding polyhedra (Figure 4,

Table 1. Schematization of ADAM results.

Alternative	Volume	Rank
Carob	0.014282	1
Walnut	0.013818	2
Pistachio	0.013189	3
Apple	0.009901	4
Citrus	0.009886	5
Chestnut	0.008713	6
Avocado	0.008365	7
Kiwi	0.007510	8
Papaya	0.006362	9
Mango	0.006208	10

).

The results of the ADAM analysis show that the top position in the ranking was assigned to the carob Ceratonia siliqua, which is considered the best crop to ensure the regeneration of the area affected by the shock.

Agronomically, carob demonstrates notable resilience, characterized by adaptability to arid climates, minimal water and fertilizer requirements, ease of maintenance, and suitability for mechanized harvesting techniques [63,64]. Economically, the species presents robust market potential, driven by strong demand in food, pharmaceutical, and livestock industries [65,66]. Notably, carob flour, produced from dried pods, has emerged as a valuable ingredient in specialized high-value supply chains, including gluten-free confectionery and bakery products, owing to its nutritional properties and absence of gluten [67,68].

From a landscape and environmental perspective, Ceratonia siliqua (carob) offers distinct advantages, particularly due to its dense, evergreen canopy that supports local biodiversity. Historically prevalent in the Salento landscape prior to the widespread cultivation of olive trees, carob’s morphological similarity to olive trees enhances its landscape integration and continuity [63]. Moreover, its resilience to arid conditions, limited water requirements, and adaptability to calcareous soils align effectively with the region’s environmental sustainability goals, particularly in terms of reduced resource inputs and climatic adaptability [64].

From an economic standpoint, Ceratonia siliqua demonstrated strong performance, primarily due to its low planting, maintenance, and labor costs. Maintenance costs for carob cultivation are significantly lower compared to other alternatives, given its minimal water requirements, reduced need for fertilizers and pesticides, and suitability for mechanized harvesting methods [63]. Additionally, carob productivity in intensive systems can reach up to 10,000 kg/ha, positioning this crop as economically attractive. Its commercial potential is robust, characterized by consistently high domestic and international market demand, especially for products derived from its pods [65]. Notably, carob flour, a gluten-free ingredient, holds significant promise within high-value-added supply chains, particularly in gluten-free confectionery and bakery sectors due to its nutritional qualities and absence of gluten [66,67]. Furthermore, carob cultivation presents relatively low operating costs, benefiting from minimal water and fertilizer inputs, limited vulnerability to phytosanitary issues, and suitability for mechanical harvesting, factors that significantly enhance its economic feasibility [63].

From a social perspective, carob cultivation aligns effectively with criteria such as food security, as it maintains reliable productivity under challenging environmental conditions, while also ensuring affordability and ease of access, thereby contributing to local food stability.

Ranked second in the analysis is walnut (Juglans regia), a deciduous species extensively cultivated across the Mediterranean basin and temperate regions globally [69,70]. Economically, walnut exhibits strong market potential, benefiting from growing domestic and international demand driven by increased consumer awareness and diverse industrial applications, especially within the food sector [71]. Europe remains reliant on imports to fulfill internal walnut consumption despite its historical prominence in walnut cultivation, highlighting an opportunity for enhanced local production [72]. Walnut productivity ranges from 3 to 4 metric tons per hectare under optimal conditions, with higher yields achievable through intensive cultivation of improved varieties [70]. Although initial planting costs are moderate, continuous irrigation requirements can increase maintenance expenditures; however, the mechanization potential during harvest significantly reduces labor costs [73]. Furthermore, the presence of various by-products derived from walnut processing creates additional avenues for economic value creation, particularly under territorial branding strategies linked to the “Made in Italy” market.

From an environmental perspective, walnut demonstrates considerable adaptability to local pedoclimatic conditions and resilience against environmental shocks, thereby sustaining long-term productivity and stability [73]. Its broad canopy and substantial biomass provide ecological benefits by supporting biodiversity and enhancing landscape heterogeneity [69]. Socially, walnut cultivation integrates well within the traditional agricultural landscape of the Salento region, facilitating acceptance among local growers who are already familiar with relevant agronomic practices.

The third-ranked alternative is pistachio (Pistacia vera L.), a perennial tree historically cultivated throughout the Mediterranean region, the Middle East, and parts of Asia and the Americas, primarily valued for its high-quality seeds [74]. Despite its high initial establishment costs and prolonged period (approximately five to seven years) before reaching optimal productivity, pistachio orchards offer extended productive lifespans [75]. Economic attractiveness is significantly bolstered by consistently high market prices and robust demand driven by gastronomic industries, including confectionery, patisserie, and processed food markets [76]. Nonetheless, the alternate bearing characteristic of pistachio cultivation necessitates careful management to optimize productivity [77]. Yields generally range from 1.5 to 2 metric tons per hectare, indicating viable commercial potential [78].

Environmentally, pistachio cultivation is advantageous due to its inherent drought resilience, limited water usage, and minimal reliance on pesticides, thereby aligning with sustainable agricultural practices. Its shrubby growth form contributes positively to landscape diversification and biodiversity, facilitating its integration within ecological buffer zones or boundary plantings. Socially, pistachio aligns with territorial branding initiatives, fostering job creation, gastronomic tourism, and enhanced local identity, thereby further strengthening its value in the broader regeneration strategy.

Ranked in intermediate positions are apple, citrus, and chestnut, which present moderate potential as alternative crops. Despite encountering some constraints related to climatic adaptability, particularly with respect to temperature extremes and water availability, these crops remain economically valuable and are culturally familiar within the region, thus benefiting from established market recognition and agronomic expertise.

Conversely, avocado, kiwi, papaya, and mango occupy the lowest positions in the analysis. These exotic species exhibit significant limitations concerning their integration into the regional agro-ecosystem, primarily due to high water requirements, reduced compatibility with local soil conditions, and limited regional familiarity with their cultivation techniques. Additionally, their introduction poses notable challenges for landscape preservation and socio-cultural acceptance, as their distinct visual characteristics diverge considerably from traditional landscape aesthetics and regional agricultural practices. Consequently, these crops are considered less feasible by experts, primarily because they represent substantial departures from the collective agricultural and cultural identity of the Salento area.

To ensure the solidity and reliability of the results obtained through the MCDM analysis, a sensitivity analysis was conducted. The results of these tests showed good model stability. As illustrated in the following graph (Figure 5), the middle positions of the alternatives underwent only minor shifts, whereas the first and last positions in the ranking remained essentially unchanged. This indicates that, despite variations in the weights, the final choices were not significantly affected, thereby confirming both the robustness of the decision-making process and the validity of the results obtained.

4. Conclusions

The present study provides a rigorous method, aimed at supporting the selection of optimal agricultural alternatives for regenerating rural areas affected by shocks, specifically addressing the Xylella fastidiosa epidemic in the Apulia region. The innovative integration of economic, environmental, and social criteria has allowed for a nuanced and multidimensional evaluation of alternative crop options, significantly enhancing traditional decision-making models. The results clearly identified carob (Ceratonia siliqua), walnut (Juglans regia), and pistachio (Pistacia vera L.) as the most promising alternatives, based on their superior performance across multiple dimensions. However, it should be noted that the optimal choice may not necessarily be limited to the single highest-ranked crop. To mitigate the well-documented risks associated with monoculture, such as heightened vulnerability to diseases, economic instability, and reduced biodiversity, policymakers and agricultural stakeholders should consider combinations of several high-ranked crops, strategically integrated to foster ecological resilience and economic diversification.

Translating these findings into practice requires targeted policy measures. Beyond simply identifying promising crops, regional and national authorities should encourage diversified farming systems where different species are strategically combined to reduce risks and enhance ecological resilience. This could be supported through incentive schemes—such as subsidies for intercropping, fiscal benefits for farms that diversify production, or dedicated funding programs that reward ecological services provided by diversified landscapes. At the same time, policy efforts should aim to strengthen cooperative structures, local processing chains, and territorial branding strategies, so that diversified products can find stable and profitable market outlets. By aligning financial incentives with market support, institutions can create the enabling conditions needed for farmers to adopt and sustain diversified models over the long term.

One significant advantage of this methodological approach lies in its scalability and transferability. Although developed in response to the specific phytosanitary emergency in Apulia, the framework can readily be adapted to address a wide array of agricultural crises, including climatic shocks, economic disruptions, or other pathogen outbreaks in different geographical and socio-economic contexts. Thus, it represents a valuable decision-support tool for enhancing agricultural resilience and sustainability globally.

From a policy perspective, this research offers concrete guidance for policymakers by providing a transparent, scientifically robust tool for prioritizing investments and policy actions. Its practical implications include aiding regional authorities in channeling financial support toward sustainable agricultural diversification, fostering effective territorial branding strategies, and supporting local economic revitalization. Moreover, the involvement of stakeholders through structured participatory methods ensures that proposed solutions align closely with local needs and preferences, thereby increasing their acceptance and practical feasibility.

Future research should further refine and test this methodological framework in different regions and under varying conditions, potentially incorporating additional dynamic criteria such as real-time climate modeling or economic market forecasts. Expanding the participatory dimension to include wider societal representation and integrating longitudinal analyses to monitor long-term outcomes of implemented strategies could also substantially enhance the decision-making robustness and practical applicability of this approach.