A Multicriteria Approach to the Study of the Energy Transition Results for EU Countries

Abstract

The article presents a multicriterial approach to evaluating the efficiency of the energy transition in EU countries, emphasizing the relationship between resource efficiency and the results of transition. The study uses a data analysis methodology (DEA) to evaluate how effectively countries use resources (inputs), such as energy consumption, investment and innovative development, to achieve the desired results (outputs), including the renewable energy sources, reduction of CO₂ and labour trends. The use of DEA with Python 3.10 software made it possible to obtain objective performance and compare them with the energy transition index (ETI). The DEA and ETI based efficiency matrix has identified four clusters of countries: high efficiency and high transition readiness; high efficiency and low transition readiness; low efficiency and high transition readiness; low efficiency and transition readiness. Validation by means of a solution (DS) confirmed the reliability of the results. The conclusions emphasize that the higher efficiency of resource use does not automatically meet the higher transition indicators, which indicates the need to improve management, innovation spread and investment distribution. The study helps to develop evidence policy by offering a system for monitoring and comparative analysis of the efficiency of the energy transition in EU countries.

1. Introduction

The global energy transition has become a strategic imperative in response to the accelerating impacts of climate change, energy insecurity and fossil fuel dependency. For the European Union, achieving carbon neutrality by 2050 requires a profound structural transformation of its energy systems, encompassing technological innovation, the integration of renewable energy, energy efficiency and sustainable consumption patterns. Measuring progress in this transformation is crucial for evidence-based policymaking, ensuring the effectiveness and equity of strategies across Member States. However, despite the increasing availability of statistical and policy data, assessing the effectiveness of the energy transition remains methodologically challenging. Existing approaches often rely on aggregate indicators that describe the level of development but do not assess how effectively countries are using available resources to achieve the transition outcomes. In addition, socio-economic heterogeneity and differences in institutional readiness make direct comparisons across countries difficult. This creates a gap between descriptive and performance-based assessments.

To address this gap, this study applies the Data Encompassment Analysis (DEA) [1] method to assess the efficiency of the energy transition in EU countries by relating inputs (energy resources, investments and innovations) to outputs (share of renewable energy, CO₂ emission reductions, labor market conditions). The analysis is complemented by the Energy Transition Index (ETI) [2] and validated by the Demarcation Score (DS), which together improve the multidimensional understanding of the efficiency of the energy transition.

The results reveal significant asymmetries across EU countries. Namely, Northern and Western Europe demonstrate both DEA and ETI efficiency, some Central and Eastern European countries remain below average due to structural inefficiencies and delayed modernization. These findings highlight the importance of targeted policy measures that promote the diffusion of innovation, a balanced allocation of investments, and adaptive governance to achieve a fair and efficient energy transition in the EU.

2. Literature Review

2.1. Research on Factors Affecting Energy Transition

Research on the effectiveness of energy transition changes is an important tool for strategic planning, sustainable development policy formulation, and rational resource allocation. Such analysis allows not only to assess the achievements of countries in implementing energy reforms but also to identify systemic weaknesses, understand why some countries achieve significantly higher results with fewer resource costs. The relevance of studying the effectiveness of the energy transition is determined by several key factors. Firstly, this process is extremely resource-intensive and requires significant financial, technological, and human investments. Therefore, assessing efficiency helps to determine how rationally available resources are used [3,4]. Secondly, energy system development in Europe is uneven–countries start from different positions, have different institutional capacities, and varying access to technologies [5]. Thirdly, with the increasing accountability of governments and the need for evidence-based decision-making, efficiency becomes a criterion for policy quality [6]. Finally, to ensure sustainable progress, it is necessary to distinguish between actions that genuinely cause positive shifts and superficial or declarative changes [7].

The study of the energy transition effectiveness is an essential tool for strategic planning, forming sustainable development policies, and rational resource use. Researchers [8] define resources as both material and immaterial elements that interact with humans and their environment. These include physical resources (e.g., natural resources and materials), as well as social, psychological, cultural, and intellectual resources that determine an individual or organization’s ability to achieve their goals, and can be mobilized through various processes, including exchange, use, and transformation of these resources to overcome external and internal demands and threats. Based on this generalization, we consider various resource factors, which will later be analysed within the framework of the DEA methodology for achieving the goals of the energy transition.

The study [9] emphasizes the strategic role of effective resource use, highlighting energy efficiency as a multidimensional concept that ensures competitiveness, reduces environmental harm, and supports the creation of sustainable jobs through optimal use of materials and production waste, particularly in resource-intensive industries. Article [10] discusses the importance of mobilizing materials for the energy transition, specifically through efficient resource use, and identifies four main paths for material mobilization: increasing primary production, redirecting existing production, repurposing used reserves, and re-mining waste and emissions. The interaction of these factors determines the maximum possible speed of material mobilization and the energy transition [11]. This issue is especially relevant for economies that consume a significant portion of natural resources and produce substantial CO₂ emissions [12]. The publication [13] examines the impact of technological choices, investment requirements, and socio-economic contexts on shaping the world’s trajectory toward a sustainable, resilient, and inclusive energy future. The article [14] highlights the key role of digitalization and innovative technologies in increasing the resilience and efficiency of energy systems [15]. Authors in publication [16] discuss the importance of international law and organizations as institutional mechanisms for regulating energy decisions at the global level. Article [17] discusses alternatives to accelerate the energy transition in the Global South, focusing on three dimensions: technology, society, and policy. It highlights the potential of these dimensions to support the transition and examines the transformation of energy consumption through energy efficiency in the context of the global energy transition. Study [18] emphasizes the role of energy efficiency in reducing dependence on fossil fuels, decreasing energy imports, and mitigating negative environmental impacts, particularly in the context of energy issues in Ukraine due to the conflict. Article [19] emphasizes the importance of a culture of energy-saving consumption as a necessary condition for modernity.

The impact of urbanization on the energy transition also deserves attention. Households’ access to various forms of energy depends on the level of urbanization, which influences their energy consumption preferences. Studies [20] have shown that in sparsely populated areas, wood/biomass and natural gas reduce energy poverty, while in cities, electricity consumption is the key factor. Study [21] emphasizes the importance of localized approaches and long-term strategies to achieve a sustainable energy transition in cities. Thus, income levels significantly affect the effectiveness of the energy transition, as higher incomes provide more opportunities for implementing technical solutions and developing appropriate measures. Optimizing resource use is key to achieving maximum efficiency in the energy transition and reducing the environmental burden, particularly in low-income locations.

In addition to studying individual factors within the context of the energy transition, studies that comprehensively examine these factors also deserve attention. A multidimensional analysis of EU countries between 2000 and 2019 identified four energy transition profiles and showed that economic growth, trade indicators, innovation systems, and the design of policy measures significantly affected progress. However, some developed economies, such as Germany and the Netherlands, lagged in achieving their climate goals [22]. In study [23], it is emphasized that improving energy efficiency and achieving sustainable economic development requires favourable environmental conditions, fostering technological innovation, aligning with regional specifics, and enhancing collaboration in the field of ecological knowledge through targeted industrial and policy strategies.

Thus, publications highlight the key resource factors influencing the energy transition, among which we focus on the level of consumption of material resources, their reuse, energy efficiency, energy consumption, job potential, investments, the impact of urbanization, and income levels. Additionally, in the works of scientists, this list is supplemented with factors determining the energy transition, including new energy sources [24,25], shifts in the fuel source for energy production [26], a shift in the nature or pattern of how energy is utilized within a system [27], decarbonization [28,29,30].

2.2. Methods for Studying the Effectiveness of the Energy Transition

Various methods are employed to study the impact of factors on the energy transition, including the Analytical Hierarchy Process [31], clustering [32], statistical analysis of expert opinions [33], and fuzzy logic modeling [34]. Each of these methods has its advantages, but they also have limitations, such as the subjectivity of assessments and weights, or difficulty in working with large datasets. The DEA method stands out due to its objectivity, which allows for the simultaneous comparison of multiple units, considering various resources and performance outcomes. One of the key advantages of DEA is its ability to identify potential improvements and optimize resource usage, making it useful for decision-making in competitive environments with changing parameters. Thus, DEA is a relevant tool for environmental analysis, ensuring high accuracy and objectivity when assessing effectiveness in different contexts. Researchers consider the DEA method suitable for studying the efficiency of resource utilization to achieve goals. As shown in Xu et al.’s study (2020), DEA is widely used to assess energy efficiency as a method for evaluating overall factor efficiency, considering the changing technical condition and structure of sectors [35]. A systematic review of the literature on the use of DEA in energy and environmental studies highlights its ability to identify systemic weaknesses and potential for improving country performance through comparative analysis of DMUs [36]. The study [37] demonstrated the importance of integrating DEA with additional tools for more precise strategic analysis at the national level. Research [38] is dedicated to the application of DEA for analysing the effectiveness of renewable energy in developed countries and the role of green finance in promoting sustainable energy usage. The DEA method was used to analyse energy efficiency in 30 provinces of China, revealing regional differences and the need for targeted sustainable development policies [39]. Based on a comprehensive review of DEA applications in energy efficiency studies, key measurement variables have been highlighted, models and practical scenarios expanded, and limitations, such as dependency on structured and clearly defined data, have been identified, with suggestions for future methodological developments [40].

To improve the results of applying the DEA methodology, researchers have combined it with other methods. Authors in [41] proposed a hybrid methodology combining DEA Window analysis and Fuzzy TOPSIS to assess the renewable energy production potential of 42 countries, based on population size, energy usage, power capacity, and broader sustainable development criteria, offering a strategic tool for guiding policies and investments in clean energy. In study [42], interval DEA and a super-efficiency model were presented for evaluating sustainable tourism, effective processing of data uncertainty, and identifying inefficiencies and improvement goals in DMUs using interval input data and greenhouse gas emissions as undesirable outcomes. Study [43] introduces a new DEA model that enables sequential comparison of the efficiency of all management units and subunits, ensures linearity, and avoids assumptions regarding multiplicative or hierarchical efficiency decomposition, enhancing its applicability to complex systems. Authors in the article [44]. consider the combination of the DEA and MCDA techniques for solving problems related to efficiency and productivity

2.3. Combining DEA and ETI for Energy Transition Effectiveness Evaluation

Evaluating the effectiveness of the energy transition requires considering both institutional preconditions and the actual results of energy system transformations. As the analysis of barriers prevalent in both developed and developing countries reveals [45], obstacles to the energy transition encompass not only financial and infrastructural challenges but also socio-institutional and regulatory aspects. In this context, it is beneficial to combine the index approach (ETI) with the tool for assessing actual effectiveness (DEA) for a comprehensive analysis of the situation. Specifically, ETI enables the quantitative evaluation of governance quality, policy stability, access to capital, infrastructure availability, and the level of human capital—factors that are typically beyond the scope of traditional productivity models. However, the drawbacks of the ETI include subjectivity in forming the aggregate indicator and the failure to account for resource usage efficiency, as ETI focuses more on potential than productivity [46].

The importance of institutional support for the energy transition has been discussed in several publications. In particular, the article [47] states that integrating transition studies with policy provides a comprehensive foundation for managing energy transformations, emphasizing public involvement in governance, reflexive political mechanisms, and innovation ecosystems. Using Germany’s energy transition (Energiewende) as an example, the study demonstrates how inclusive multi-level governance fosters legitimacy, adaptability, and stakeholder engagement, ensuring socially just and sustainable transition processes. The study [48] highlights that a successful and fair energy transformation depends on multidimensional policies, identifying five main thematic areas—institutional, ecological, financial, socio-cultural, and technical—which together form the development of renewable energy. A systematic review in the source [49] identifies four key categories—economic, technological, social, and political—as foundational for accelerating the transition to low-carbon energy, with researchers emphasizing the critical role of environmental policy and stakeholder participation, while noting the limited attention to developing countries and low-income nations in existing studies. In the article [28], explicit and implicit aspects of the energy transition are distinguished: explicit aspects include observable changes in energy consumption and infrastructure, while implicit aspects involve deeper transformations such as governance, technologies, and geopolitical dynamics. By highlighting the interconnection of energy with social, economic, and ecological systems, the study emphasizes the need for future research to examine multi-scale consequences and support a fair and sustainable transition. In the study [50], advanced bibliometric and systematic analysis methods are employed to highlight the significance of the scientific landscape in transitioning to renewable energy sources, with a focus on key economic, political, managerial, social, and climate-related aspects.

Thus, the institutional component is a crucial factor in the energy transition, which is systematically addressed in the ETI methodology. At the same time, DEA enables the assessment of actual effectiveness in transforming resources (including investments, energy consumption, and human potential) into desired results—such as increasing the share of renewable energy, reducing emissions, and fostering innovative economic growth. DEA accounts for the heterogeneity of factors and identifies countries that, with the same or fewer resources, achieve better results, which is particularly important in regions with high costs or weak institutional support. ETI and DEA play complementary roles: the former diagnoses institutional capability, while the latter determines real effectiveness. Combining both approaches enables not only the identification of barriers that slow progress but also the identification of benchmark countries that can optimize policies and resources. This opens opportunities for forming comprehensive, strategically grounded decisions in the energy transition that consider both conditions and outcomes.

Therefore, the aim of this article is to establish and test the conceptual foundations for studying the factors influencing the effectiveness of energy transition, based on a multi-criteria approach. This approach considers the application of the DEA method and its integration with ETI for benchmarking the research results. The authors aim to propose a methodological framework that allows for an objective assessment of the effectiveness of countries or regions in implementing energy transformation, considering both quantitative and qualitative criteria. The research tasks include considering a multi-criteria approach for choosing factors of energy transition results study; developing and adapting a DEA model to assess the effectiveness of energy transformation in EU countries; comparing the results with other existing rankings (e.g., ETI); and forming an efficiency matrix for conducting benchmarking and identifying the potential of countries for energy transition.

3. Materials, Data and Methods

To achieve the article’s goal, the use of the DEA methodology is proposed. The necessity of applying DEA in energy transition studies is supported by a growing body of academic literature that emphasizes the importance of multidimensional evaluation frameworks. Prior studies have shown that energy transition is influenced not only by technological capacity and resource availability but also by governance quality, socio-economic structure, and innovation potential. The DEA provides an empirical basis for incorporating these multiple factors into a unified evaluation model. A key strength of the DEA method lies in its ability to rank the evaluated DMUs. Furthermore, combining DEA with complementary indices, such as ETI, allows researchers to overcome the limitations of one-dimensional rankings and better reflect the complexity of real-world transition dynamics.

The methodological framework of this study is based on integrating DEA into a multicriteria assessment model to evaluate the effectiveness of national energy transition efforts across European countries. The research employs a non-parametric, input-oriented DEA model under the assumption of variable returns to scale (CCR model), allowing for the comparison of countries (DMUs) in terms of their efficiency in converting multiple energy, economic, and environmental resources into desired outcomes.

The analysis is based on a panel of 27 European countries for which relevant quantitative indicators were collected from Eurostat databases. Variables were selected to reflect both the resource inputs and performance outputs of the energy transition process. The DEA model uses the following indicators.

In the DEA model, three key indicators have been selected as output variables to reflect the results of the energy transition. The share of renewable energy sources (y₁) represents the level of ecological transformation in the energy balance. The CO₂ Emissions in comparison with economic growth are presented by the decoupling Score (DS) indicator (y₂). The decoupling of GHG emissions from GDP growth in the article is determined using the Tapio decoupling index (ε), a quantitative measure designed to evaluate the relationship between economic growth and environmental impact, particularly carbon emissions (Formula (1)).

$$\mathsf{\epsilon }={\displaystyle \frac{\mathrm{e}}{\mathrm{g}}},$$

(1)

where e—$\mathrm{G}\mathrm{H}\mathrm{G} \mathrm{g}\mathrm{r}\mathrm{o}\mathrm{w}\mathrm{t}\mathrm{h} \mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e};$g—$\mathrm{G}\mathrm{D}\mathrm{P} \mathrm{g}\mathrm{r}\mathrm{o}\mathrm{w}\mathrm{t}\mathrm{h} \mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e}$.

The index ε is derived from the elasticity concept, which, in this context, represents the responsiveness of carbon emissions to changes in GDP. It characterizes the success of decarbonization measures and measures the country’s ability to reduce CO₂ emissions while maintaining economic growth and reflects how successfully technologies that reduce environmental impact are being implemented, contributing to the sustainable development of the economy. The higher this coefficient, the more effectively the country implements decarbonization and energy transformation strategies [51].

Decoupling Index can take different values. For example, the literature discusses eight types of it [52]. In this study, we consider four cases that reflect different combinations of changes in GDP (g) and emissions (e). Taking these into account allows us to distinguish the following types of decoupling:

(1)

Absolute decoupling (g > 0, e ≤ 0): we assign DS = 1—this is the strategically best state: the economy is growing, emissions are decreasing;

(2)

The worst case (g ≤ 0, e ≥ 0): we assign DS = 0—recession is accompanied by an increase in GHGs;

(3)

Relative decoupling (g > 0, e > 0)—GHGs and GDP are growing, we assign DS, calculated with the formula:

$$DS={\displaystyle \frac{1}{1+\epsilon }},$$

(2)

The meaning of this calculation is that ε, which shows how fast emissions grow relative to economic growth, determines better decoupling at lower ε and worse at higher ε.

(4)

Both indicators fall (g < 0, e < 0): we assign DS, calculated with the formula:

$$DS={\displaystyle \frac{\left|e\right|}{\left|e\right|+\left|g\right|}},$$

(3)

The content of the calculation is to determine the share of the “total reduction” that is due to the reduction in GHGs. The goal is to demonstrate the decline in emissions, but not to equate the case with the best-case scenario, if the reduction is achieved at the cost of a recession. If GHGs fall more strongly than GDP → the share is closer to 1. If the economy falls more strongly than GHGs → the share is closer to 0. The coefficient k = 0.6, κ ∈ (0, 1) guarantees that even with a large reduction in GHGs in a recession, it will not exceed the fall in GDP.

Thus, for DS calculation, we use Formula (4):

$$\mathrm{D}\mathrm{S}\left(\mathrm{g},\mathrm{e}\right)=\left\{\begin{array}{c} 1, \mathrm{g}>0, \mathrm{e}\le 0 (\mathrm{a}\mathrm{b}\mathrm{s}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{t}\mathrm{e} \mathrm{d}\mathrm{e}\mathrm{c}\mathrm{o}\mathrm{u}\mathrm{p}\mathrm{l}\mathrm{i}\mathrm{n}\mathrm{g}, \mathrm{t}\mathrm{h}\mathrm{e} \mathrm{b}\mathrm{e}\mathrm{s}\mathrm{t} \mathrm{c}\mathrm{a}\mathrm{s}\mathrm{e})\hfill \\ {\displaystyle \frac{1}{1+\epsilon }}, \mathrm{g}>0, \mathrm{e}>0 (\mathrm{r}\mathrm{e}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{v}\mathrm{e} \mathrm{d}\mathrm{e}\mathrm{c}\mathrm{o}\mathrm{u}\mathrm{p}\mathrm{l}\mathrm{i}\mathrm{n}\mathrm{g})\hfill \\ 0, \mathrm{g}\le 0, \mathrm{e}\ge 0 (\mathrm{r}\mathrm{e}\mathrm{c}\mathrm{e}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{w}\mathrm{i}\mathrm{t}\mathrm{h} {\mathrm{C}\mathrm{O}}_2\mathrm{e}\mathrm{m}\mathrm{i}\mathrm{s}\mathrm{s}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{g}\mathrm{r}\mathrm{o}\mathrm{w}\mathrm{t}\mathrm{h}, \mathrm{t}\mathrm{h}\mathrm{e} \mathrm{w}\mathrm{o}\mathrm{r}\mathrm{s}\mathrm{t} \mathrm{c}\mathrm{a}\mathrm{s}\mathrm{e}) \\ {\displaystyle \frac{\left|e\right|}{\left|e\right|+\left|g\right|}}, \mathrm{g}<0, \mathrm{e}<0 (\mathrm{t}\mathrm{h}\mathrm{e} \mathrm{b}\mathrm{o}\mathrm{t}\mathrm{h} \mathrm{f}\mathrm{a}\mathrm{l}\mathrm{l}, \mathrm{t}\mathrm{o} \mathrm{p}\mathrm{r}\mathrm{i}\mathrm{o}\mathrm{r}\mathrm{i}\mathrm{t}\mathrm{a}\mathrm{s}\mathrm{e} {\mathrm{C}\mathrm{O}}_2\mathrm{e}\mathrm{m}\mathrm{i}\mathrm{s}\mathrm{s}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{d}\mathrm{e}\mathrm{c}\mathrm{r}\mathrm{e}\mathrm{a}\mathrm{s}\mathrm{e}, \mathrm{k}=0.6)\end{array}\right.,$$

(4)

The proposed Decoupling Score normalizes Tapio’s elasticity to the [0, 1] interval and provides a ‘higher = better’ interpretation. Under relative decoupling, smaller values of ε (GHG grows more slowly than GDP) yield a DS that approaches 1. When GDP and GHG both decline, DS increases with the larger contribution of GHG reduction to the total decline, but is capped at κ = 0.6 so that recession scenarios do not outweigh cases with economic growth. Thus, the indicator aligns with the policy logic of promoting decoupling and does not ‘over-reward’ emission cuts achieved at the expense of economic contraction.

Job vacancies in the power sector transition (JV_PST) were included as an output (y₃) in the DEA model to capture the social dimension of the energy transition. While traditional outputs such as RES share, CO₂ reduction, or energy intensity represent technological and environmental effects, JV_PST reflects the socio-economic benefits of transition policies. Including this variable allows assessing which EU countries are not only technically efficient but also capable of generating green employment, in line with the objectives of the European Green Deal and the Just Transition Mechanism.

-

x₁: Energy import dependency rate (%)—high values indicate vulnerability to external energy supply;

-

x₂: Share of fossil fuels in total energy supply (%)—a key indicator of carbon-based dependency;

-

x₃: Investments in climate change mitigation (million EUR)—resources consumed to promote sustainability;

-

x₄: EIS is considered a reliable and comprehensive source for analysing innovation potential, making it ideal for studies on technological innovations and regional development [53];

-

x₅: Income by degree of urbanization (million EUR)—considered a resource associated with urban energy consumption and infrastructure capacity;

-

x₆: Energy consumption per capita—reflects intensity of resource use; lower values imply greater efficiency.

The goal of DEA is to achieve the goal of maximizing the efficiency ratio—that is, how proportionally can output be increased without reducing the level of input resources [1]. Objective of the task—Maximize θ—DEA efficiency ratio. The task is based on the following constraints:

-

input constraints: weighted sum of inputs for all DMUs is less than or equal to the input for the evaluated DMU:

$${\sum }_{j=1}^n{\lambda }_j{x}_{ij}\le x_{i0},{\forall }_i=1,\dots m,$$

(5)

-

output constraints: weighted sum of outputs for all DMUs is greater than or equal to θ times the output for the evaluated DMU):

$${\sum }_{j=1}^n{\lambda }_j{xy}_{rj}\le x_{r0},{\forall }_r=1,\dots s,$$

(6)

Non-negativity of weights:

$${\lambda }_j\ge 0,{\forall }_j=1,\dots n.$$

(7)

Variables:

λ_j—Weights for other DMUs;

θ—Output scaling coefficient (DEA efficiency ratio);

Notation Explanation:

x_ij—i-th input for j-th DMU;

y_rj—r-th output for j-th DMU;

x_io, y_ro—Input/output for the evaluated DMU;

n—Number of countries (DMUs);

m—Number of inputs;

s—Number of outputs.

This formulation is an output-oriented CCR. Output-oriented because the objective is to maximize output (θ) while keeping inputs constant. The CCR model, named after its developers (Charnes, Cooper, Rhodes), is the basic DEA model. It always assumes constant returns to scale (Constant Returns to Scale, CRS). There is no convexity constraint (∑λj = 1), implying constant returns to scale, which means: the CCR model is designed to assess efficiency assuming that efficiency is independent of the scale of operations; the absence of an additional constraint that the sum of the weights (λj) is equal to one is a mathematical reflection of this assumption of constant returns to scale. This allows the model to compare DMUs of different sizes without considering scale effects, if the optimal scale does not matter for achieving efficiency; in practice, this makes the CCR model more rigid and may lead to fewer efficient units, since to achieve efficiency a DMU must be efficient in an absolute sense, regardless of its scale. The analysis is performed for a single base year to provide evidence of the application of the DEA methodology under stable structural conditions. This design minimizes the effects of temporal fluctuations in the data and allows a clear assessment of the internal consistency and discriminatory power of the model before extending the approach to multi-period or dynamic analysis.

In the DEA model, all inputs are treated as resources consumed in the process of achieving energy transition outcomes. Import dependency (x₁) and the share of fossil fuels (x₂) indicate the level of vulnerability and carbon dependence of the energy system. Investments in climate initiatives (x₃), the level of innovation and technological progress (x₄), and income by urbanization level (x₅) are interpreted as socio-economic resources that form the potential for transformation, but if used irrationally, may indicate low efficiency. Energy consumption per capita (x₆) reflects the overall energy intensity of the economy. Though some variables, such as investments or innovation and technological progress, might appear as stimulators of progress, within the DEA logic, they are treated as inputs (i.e., resources consumed) and thus classified as de-stimulators to align with the model’s efficiency interpretation.

All variables were normalized to ensure comparability and avoid scale distortions, using min-max normalization where necessary. The DEA model was implemented in Python 3.10, offering transparency, flexibility, and scalability. Python allows the researcher to fully control the construction of the DEA model, adjust parameters, integrate weighting mechanisms if needed, and replicate results efficiently.

Accordingly, considering these indicators as resource provision for the energy transition, we will normalize them using the formula of linear transformation of disincentives:

$${\mathrm{x}}_{\mathrm{i}\mathrm{j}}^{\mathrm{n}\mathrm{o}\mathrm{r}\mathrm{m}}={\displaystyle \frac{{\mathrm{m}\mathrm{a}\mathrm{x}}_{\mathrm{x}\mathrm{j}}-{\mathrm{x}}_{\mathrm{i}\mathrm{j}}}{{\mathrm{m}\mathrm{a}\mathrm{x}}_{\mathrm{x}\mathrm{j}}-{\mathrm{m}\mathrm{i}\mathrm{n}}_{\mathrm{x}\mathrm{j}}}},$$

(8)

where

-

x_ij—actual value of the disincentive for object i by indicator j;

-

${\mathrm{m}\mathrm{a}\mathrm{x}}_{\mathrm{x}\mathrm{j}}$, ${\mathrm{m}\mathrm{i}\mathrm{n}}_{\mathrm{x}\mathrm{j}}$—respectively, the largest and smallest value of the indicator among all objects;

-

x_ij^norm ∈ [0;1]—a normalized value where lower values of the disincentive are converted to higher values on the efficiency scale (i.e., values closer to 1 are better).

In the DEA model, all output indicators are interpreted as the results of the efforts expended. Together, these indicators provide a comprehensive assessment of the effectiveness of the use of energy resources and are characterized as stimulants. The formula for their normalization is as follows:

$${\mathrm{x}}_{\mathrm{i}\mathrm{j}}^{\mathrm{n}\mathrm{o}\mathrm{r}\mathrm{m}}={\displaystyle \frac{{\mathrm{x}}_{\mathrm{i}\mathrm{j}}-{\mathrm{m}\mathrm{i}\mathrm{n}}_{\mathrm{x}\mathrm{j}}}{{\mathrm{m}\mathrm{a}\mathrm{x}}_{\mathrm{x}\mathrm{j}}-{\mathrm{m}\mathrm{i}\mathrm{n}}_{\mathrm{x}\mathrm{j}}}},$$

(9)

Thus, all variables—both input and output—are reduced to a single scale within [0;1], which allows for the correct application of DEA analysis without distorting the results due to different scales, units of measurement or directions of influence. This approach ensures metric consistency, analytical transparency and interpretative unambiguity in comparing countries in terms of their energy efficiency. This transformation allows for the preservation of the evaluation logic, according to which high values of stimulants and low values of disincentives are interpreted as a sign of greater efficiency.

DEA method for assessing relative efficiency considering a set of input and output variables. DEA allows for the comparison of countries as Decision Making Units (DMUs) operating under the same external circumstances, but with different resources and outcomes. To ensure the reliability of the results, an empirical rule regarding the minimum number of DMUs required for the correct construction of the model was followed. According to it:

$$\mathrm{D}MU\ge max\left\{m\times s, 3·(m+s)\right\}$$

(10)

where m—the number of input variables; s—the number of output variables.

This ratio ensures that the model will have sufficient statistical power to distinguish between efficient and inefficient countries. For example, when using 6 input and 3 output variables, the required number of countries (DMUs) is at least:

$$max\left\{6\times 3, 3·(6+3)\right\}=max\left\{\mathrm{18,27}\right\}=27 countries$$

Thus, for a model that includes 6 input and 3 output variables, at least 27 countries were included in the study to ensure the statistical validity of the constructed efficiency.

In this study, the Python 3.10 programming language was used to implement DEA, which provides full transparency, flexibility and reproducibility of calculations. Thanks to the open source and wide capabilities of Python, the researcher can adapt the model to the specifics of the input and output variables, add weight constraints, scale calculations to large samples, and integrate DEA with other analysis methods (clustering, regression, visualization, etc.). This ensures not only the accuracy and reliability of the results but also allows us to consider the complexity and multidimensionality of energy transition processes.

To validate the robustness of the DEA efficiency results, OAT sensitive analysis [54] was applied by increasing input/output variables by 5%. This step enabled the identification of the most influential outputs, which not only confirm the model’s stability but also provide empirical justification for the selection of these indicators.

To enhance interpretability and address multidimensionality, DEA results were combined with the Energy Transition Index (ETI), a composite indicator provided by the World Economic Forum that captures broader institutional, technological, and market readiness dimensions. A correlation analysis between DEA efficiency scores and ETI rankings was performed, revealing a relationship, which justifies the development of a two-dimensional efficiency matrix (DEA Score vs. ETI Rank).

The Efficiency Matrix for Strategic Positioning enables the classification of countries into strategic quadrants, supporting more nuanced policy recommendations. This approach is widely accepted and for this research it allows classifying countries by their strategic behavior in the energy transition; identifying asymmetry between the level of institutional readiness and the actual results of energy reforms; identifying benchmark countries—those that achieve high efficiency even under unfavorable conditions; and identifying potential reserves in countries that, despite a high ranking position, demonstrate lower performance in utilizing available opportunities. Together, these indicators allow us to assess how effectively countries use available resources to achieve the desired results of sustainable development.

To determine the presence of a statistical relationship between the ETI and the efficiency indicator according to the DEA model, the Pearson correlation coefficient was calculated using the classical formula:

$$\mathrm{r}={\displaystyle \frac{{\sum }_{i=1}^n(x_i-\overline{x })(y_i-\overline{y })}{\sqrt{\sum _{i=1}^n{(x_i-\overline{x })}^2}·\sqrt{\sum _{i=1}^n{(y_i-\overline{y })}^2}}}$$

(11)

where x_i—DEA Efficiency Score value for country i;

y_i—ETI value for country i;

$\overline{x }$, $\overline{y }$—average values of DEA and ETI, respectively.

For a better understanding of the research methodology, its graphic visualization is offered in Figure 1.

4. Results

Based on the results of the literature review, we identified groups of input and output indicators, which are examined within the DEA methodology in this study. The description and characteristics of each indicator are provided in

Table A1. Justification of the feasibility of using input variables in the DEA model.

№	Indicator Name	Variable Code	DEA-Role	Economic Interpretation as a Resource (Disincentive)
1	Level of energy import dependence (%)	x1	Input	Reflects the country’s strategic vulnerability; high values indicate the need for external resources, which reduces the stability of the energy system.
2	Share of fossil fuels in the total energy balance (%)	x2	Input	Carbon dependency indicator: a decrease in this share indicates a more environmentally efficient energy consumption structure.
3	Investment in climate change mitigation measures (million euros)	x3	Input	Despite the positive effect, investments are treated as a resource that is consumed to achieve sustainable development results.
4	EU Innovation Index (EIS)	x4	Input	The EU Innovation Index (EIS) is an important indicator for assessing the effectiveness of the energy transition, as it reflects the level of innovation and technological progress in a country. A high EIS indicates the development of new renewable energy technologies, the optimization of energy systems through innovations such as smart grids, and active investment in research and development. This allows countries to adapt more quickly to the demands of the energy transition, reduce CO2 emissions, and improve resource efficiency, contributing to a successful transition to sustainable energy systems.
5	Income by level of urbanization (million euros)	x5	Input	The economic capacity of urbanized areas is considered as a resource that promotes transformation, but with high consumption does not always provide proportional efficiency
6	Energy consumption per capita	x6	Input	Reflects the energy intensity of the economy; lower consumption with the same results indicates greater energy efficiency.

and

Table A2. Justification of the output variables in the DEA model.

№	Indicator Name	Variable Code	DEA-Role	Value for Assessing the Efficiency of the Energy Transition
1	Total share of energy from renewable sources (% in gross final energy consumption)	y1	Output	The basic indicators of sustainable development. It reflects the actual result of the transition to clean energy sources. The growth of this indicator indicates the achievement of the targets of the ecological transformation.
2	Level of vacancies in the scientific and technical sector (%)	y2	Output	High demand for innovative labour may indicate inefficient use of human capital, as an underutilized resource.
3	Decoupling Score (DS)	y3	Output	The Decoupling Score (DS) is a single, “higher = better” index (0 to 1) that summarizes how a country’s economy and its greenhouse-gas emissions move together in each period. It uses annual GDP growth and GHG change as inputs. DS equals 1 when the economy grows while emissions are flat or falling (absolute decoupling); it lies between 0 and 1 when both grow but GDP grows faster (relative decoupling, the closer to 1, the cleaner the growth); it gives mid-range value when both GDP and emissions fall (recognizing emission cuts during downturns without over-rewarding recessions); and it is 0 in the worst case, when the economy contracts while emissions still rise. The score is easy to compare across countries and years, aligns with policy goals of cleaner growth, and should be read alongside multi-year trends and sectoral evidence.

of the Appendix A, located at the end of the article. The input indicators x₁–x₆ were obtained from the official Eurostat website. As for the output indicators, for y₁ and y₃, the sources of information were also obtained from the Eurostat website. The y₂ indicator was calculated, and its calculation was based on Formula (4). The results of the calculation are presented in

Table 1. Value of the DS indicator.

Country	GHG Growth Rate (%) [55]	GDP Growth Rate (%)	$\mathit{\epsilon }$ (GHG-GDP Elasticity)	Regime	DS
Belgium	5.1	1.1	4.667	Relative decoupling (GDP ↑, GHG ↑)	0.176
Bulgaria	4.0	4.1	0.981	Relative decoupling (GDP ↑, GHG ↑)	0.505
Czech Republic	2.2	1.8	1.233	Relative decoupling (GDP ↑, GHG ↑)	0.448
Denmark	2.6	3.9	0.655	Relative decoupling (GDP ↑, GHG ↑)	0.604
Germany	1.6	−0.4	−4.040	Worst-case (GDP ↓, GHG ↑)	0.000
Estonia	−11.3	1.2	−9.390	Absolute decoupling (GDP ↑, GHG ↓)	1.000
Ireland	3.1	9.2	0.333	Relative decoupling (GDP ↑, GHG ↑)	0.750
Greece	8.1	2.7	3.006	Relative decoupling (GDP ↑, GHG ↑)	0.250
Spain	2.4	3.2	0.757	Relative decoupling (GDP ↑, GHG ↑)	0.569
France	0.4	1.1	0.331	Relative decoupling (GDP ↑, GHG ↑)	0.751
Croatia	6.1	3.9	1.566	Relative decoupling (GDP ↑, GHG ↑)	0.390
Italy	4.1	1.0	4.126	Relative decoupling (GDP ↑, GHG ↑)	0.195
Cyprus	4.4	2.6	1.674	Relative decoupling (GDP ↑, GHG ↑)	0.374
Latvia	1.4	−0.4	3.453	Both falling (GDP ↓, GHG ↓)	0.467
Lithuania	10.8	4.0	2.704	Relative decoupling (GDP ↑, GHG ↑)	0.270
Luxembourg	−2.3	1.8	−1.292	Absolute decoupling (GDP ↑, GHG ↓)	1.000
Hungary	4.7	0.4	11.783	Relative decoupling (GDP ↑, GHG ↑)	0.078
Malta	1.7	2.8	0.615	Relative decoupling (GDP ↑, GHG ↑)	0.619
Netherlands	4.1	1.9	2.163	Relative decoupling (GDP ↑, GHG ↑)	0.316
Austria	−1.7	−0.5	3.382	Both falling (GDP ↓, GHG ↓)	0.464
Poland	1.6	4.1	0.400	Relative decoupling (GDP ↑, GHG ↑)	0.714
Portugal	2.4	2.9	0.813	Relative decoupling (GDP ↑, GHG ↑)	0.552
Romania	0.3	0.5	0.662	Relative decoupling (GDP ↑, GHG ↑)	0.602
Slovenia	9.4	1.5	6.266	Relative decoupling (GDP ↑, GHG ↑)	0.138
Slovakia	2.6	1.7	1.514	Relative decoupling (GDP ↑, GHG ↑)	0.398
Finland	−6.1	0.9	−6.789	Absolute decoupling (GDP ↑, GHG ↓)	1.000
Sweden	−2.3	1.9	−1.232	Absolute decoupling (GDP ↑, GHG ↓)	1.000

.

The results analysis obtained allow us to conclude that the countries have different progress in energy transition:

(1)

Absolute decoupling (best case, DS = 1.000) was achieved in Estonia, Luxembourg, Finland, and Sweden, which expanded their economies while reducing emissions. In these cases, growth and mitigation coexisted, indicating that efficiency gains, fuel switching, clean power supply, or structural shifts outweighed any emission pressures from added output. These countries provide the benchmark pattern the EU ultimately seeks to scale.

(2)

Relative decoupling (mixed case, 0 < DS < 1) demonstrates a tendency—GDP ↑, GHG ↑, and indicates the economy grows, and emissions still increase, but the DS tells us how strongly growth outpaces emissions:

-

stronger relative decoupling (DS ≥ 0.60): France (0.751), Ireland (0.750), Poland (0.714), Malta (0.619), Denmark (0.604), Romania (0.602). These profiles show meaningful mitigation pressure within growth: emissions are rising more slowly than GDP, suggesting that efficiency policies, renewable deployment, or cleaner sectoral composition are taking hold;

-

moderate (0.50 ≤ DS < 0.60): Spain (0.569), Portugal (0.552), Bulgaria (0.505). Progress is visible but narrower; incremental improvements could tip these cases into the stronger bracket;

-

weak (DS < 0.50): Czechia (0.448), Slovakia (0.398), Croatia (0.390), Cyprus (0.374), Netherlands (0.316), Lithuania (0.270), Greece (0.250), Italy (0.195), Belgium (0.176), Slovenia (0.138), Hungary (0.078). In these economies, emissions rose nearly as fast as—or faster than—GDP. That pattern signals either carbon-intensive growth, delayed clean-energy adoption, or rebound effects that outweigh efficiency gains. Within this set, the lowest DS values correspond to ε > 1 (emissions growing faster than GDP), which serves as a red flag to address first.

(3)

Green recession (both falling, capped DS ≈ 0.46–0.47) follows a tendency—GDP ↓, GHG ↓, and was achieved in Latvia (0.467) and Austria (0.464) reduced emissions alongside economic contraction. The score acknowledges that a larger share of the joint decline stems from falling emissions but applies an explicit cap to ensure that recession-driven cuts do not overshadow countries that achieve reductions through growth. Policy-wise, the priority is to convert these short-run declines into structural improvements (efficiency, fuel switching, clean heat/power) that persist when growth resumes.

(4)

Worst case (DS = 0) with a tendency—DP ↓, GHG ↑, was achieved in Germany, which is the sole instance where the economy shrank while emissions increased. By design, DS = 0 flags this as the least desirable combination, typically indicating adverse shocks in low-carbon supply, weather-driven demand spikes, or sector-specific setbacks that increase emissions intensity during a downturn. The remedy is targeted and sectoral—e.g., stabilizing clean generation and heat, removing bottlenecks for electrification, and cushioning efficiency investments during slowdowns.

In general, it can be concluded that in some countries, economic growth is not accompanied by adequate reductions in greenhouse gas emissions, which may signal the need to strengthen environmental policies and seek more effective pathways for energy transition.

Thus,

Table 2. Input and Output Indicators for DEA Analysis.

EU Countries	RES_Share (%) [56]	Imp_Dep (%) [57]	Fossil_Share (%) [57]	CCM_Inv (EUR Million) [58]	JV_PST (%) [59]	DS	EIS [53]	IDU (EUR Million) [60]	EC_Per Capita (kWh/Person·Year) [61]
Austria	40.84	61.10	66.42	2756.33	7.40	0.46	119.90	34,130	145
Belgium	14.74	76.10	73.62	3682.74	8.50	0.18	125.80	29,244	200
Bulgaria	22.58	39.70	66.32	194.90	0.70	0.51	46.70	10,048	112
Croatia	28.05	55.70	67.28	353.61	1.93	0.39	69.60	120,690	96
Czech Republic	48.59	41.70	71.47	981.11	8.00	0.45	94.70	16,665	149
Denmark	44.92	38.90	57.23	5039.79	3.35	0.60	137.60	37,853	117
Estonia	40.95	3.50	68.51	172.73	2.38	1.00	98.60	18,216	138
Finland	50.75	29.60	38.26	1769.69	3.70	1.00	134.30	32,690	253
France	22.28	44.90	48.15	21,606.03	2.95	0.75	105.30	29,445	138
Germany	21.55	66.40	78.77	18,589.47	6.80	0.00	117.80	31,091	130
Greece	25.27	75.60	82.18	339.83	3.20	0.25	79.50	12,144	34
Hungary	17.36	62.10	69.43	1424.54	3.70	0.08	70.40	9548	106
Ireland	15.25	77.90	87.67	425.47	1.68	0.75	115.80	38,642	115
Italy	19.56	78.40	78.32	4976.76	2.08	0.20	90.30	24,143	103
Latvia	43.22	32.70	57.08	378.11	2.70	0.47	52.50	14,390	98
Lithuania	31.93	68.00	64.46	804.90	1.78	0.27	83.80	16,115	107
Poland	16.50	48.00	88.00	2899.62	0.90	0.71	62.80	12,740	112
Portugal	35.16	66.90	68.27	150.14	2.43	0.55	85.60	15,705	90
Romania	25.76	27.90	72.47	1258.56	0.83	0.60	33.10	9263	68
Slovakia	16.99	57.70	63.78	417.20	0.43	0.40	65.60	9781	127
Slovenia	25.07	49.30	60.93	141.72	4.10	0.14	95.10	20,487	120
Spain	24.85	68.40	72.38	2911.79	0.98	0.57	89.20	22,278	109
Sweden	66.39	26.40	31.39	5921.94	3.90	1.00	134.50	31,849	187
Luxembourg	11.62	90.60	78.69	137.00	5.00	1.00	117.20	64,863	234
Netherlands	17.15	40.45	89.13	2485.59	4.53	0.32	128.70	32,404	179
Malta	15.08	97.56	96.28	64.66	2.78	0.62	85.80	24,848	247
Cyprus	20.21	92.21	88.83	16.93	2.85	0.37	105.40	23,907	125

presents all the indicators used for analyzing energy transition efficiency using the DEA method.

Based on the use of Formulas (8) and (9), the values were normalized (

Table 3. Normalized DEA Analysis Indicators for EU Countries.

EU Countries	RES_Share (%) [56]	Imp_Dep (%) [57]	Fossil_Share (%) [62]	CCM_Inv (EUR Million) [58]	JV_PST (%) [59]	DS	EIS [53]	IDU (EUR Million) [60]	EC_Per Capita (kWh/Person·Year) [61]
Austria	0.534	0.388	0.381	0.873	0.864	0.740	0.169	0.777	0.493
Belgium	0.057	0.228	0.254	0.830	1.000	0.527	0.113	0.821	0.242
Bulgaria	0.200	0.615	0.383	0.992	0.033	0.582	0.870	0.993	0.644
Croatia	0.300	0.445	0.366	0.984	0.186	0.394	0.651	0.000	0.717
Czech Republic	0.675	0.594	0.292	0.955	0.938	0.861	0.411	0.934	0.475
Denmark	0.608	0.624	0.544	0.767	0.362	0.824	0.000	0.743	0.621
Estonia	0.536	1.000	0.344	0.993	0.242	0.004	0.373	0.920	0.525
Finland	0.714	0.723	0.879	0.919	0.405	0.708	0.032	0.790	0.000
France	0.195	0.560	0.704	0.000	0.312	0.448	0.309	0.819	0.525
Germany	0.181	0.331	0.163	0.140	0.789	0.000	0.189	0.804	0.562
Greece	0.249	0.233	0.103	0.985	0.343	0.628	0.556	0.974	1.000
Hungary	0.105	0.377	0.328	0.935	0.405	0.732	0.643	0.997	0.671
Ireland	0.066	0.209	0.006	0.981	0.155	0.639	0.209	0.736	0.630
Italy	0.145	0.204	0.171	0.770	0.204	0.720	0.453	0.866	0.685
Latvia	0.577	0.690	0.546	0.983	0.281	0.699	0.814	0.954	0.708
Lithuania	0.371	0.314	0.416	0.964	0.167	1.000	0.515	0.939	0.667
Poland	0.089	0.527	0.000	0.866	0.058	0.774	0.716	0.969	0.644
Portugal	0.430	0.326	0.349	0.994	0.248	0.368	0.498	0.942	0.744
Romania	0.258	0.741	0.274	0.942	0.050	0.674	1.000	1.000	0.845
Slovakia	0.098	0.424	0.428	0.981	0.000	0.718	0.689	0.995	0.575
Slovenia	0.246	0.513	0.478	0.994	0.455	0.137	0.407	0.899	0.607
Spain	0.242	0.310	0.276	0.866	0.068	0.472	0.463	0.883	0.658
Sweden	1.000	0.757	1.000	0.726	0.430	0.398	0.030	0.797	0.301
Luxembourg	0.000	0.074	0.164	0.994	0.566	0.754	0.195	0.501	0.087
Netherlands	0.101	0.607	−0.020	0.886	0.508	0.622	0.085	0.792	0.338
Malta	0.063	0.000	−0.146	0.998	0.291	0.835	0.496	0.860	0.027
Cyprus	0.157	0.057	−0.015	1.000	0.300	1.000	0.308	0.869	0.584

).

To determine the weight coefficient for each country (Formulas (2)–(4)), a linear programming method was applied using the Python programming language. The program code for solving this task is presented in Algorithm A1 Appendix B. Following the DEA methodology, the calculation of DEA Score was provided based on Formulas (5)–(7).

The results of the DEA were prepared in cartographic form and presented in Figure 2.

As a result of the DEA conducted to assess the efficiency of countries in achieving energy transition goals, the conversion of input resources into desired results was evaluated, particularly regarding the use of renewable energy sources, economic growth, and the reduction in greenhouse gas emissions.

The results obtained show significant variability in the efficiency levels of the energy transition across EU countries. Specifically, most countries, such as Belgium, the Czech Republic, Estonia, Finland, Germany, France, Ireland, Poland, Austria, Croatia, Sweden, Denmark, Luxembourg, the Netherlands, Cyprus, and Malta have a DEA efficiency score of 1.0000, indicating their high efficiency in the context of energy transition. These countries are self-sufficient in their achievements and do not require comparisons with others to evaluate their efficiency. This reflects the successful implementation of energy and environmental policies that ensure high results in energy transition. Slovenia, Portugal, Latvia, Greece,

In contrast, some countries have lower efficiency scores, indicating the need to improve their energy transition strategies. Portugal, with a score of 0.8951, Lithuania (0.7847), Greece (0.7498), Spain (0.7076), and Latvia (0.7037) have even lower results, indicating the need to enhance their energy strategies and improve the effectiveness of environmental initiatives.

Countries with low efficiency scores, including Romania (0.6448), Bulgaria (0.4903), Italy (0.4682), Slovakia (0.3576), Slovenia (0.4492), and Hungary (0.3680), have significant potential for improvement in their energy policies. Comparing these countries with leaders helps identify areas for improvement, including more active implementation of renewable energy sources, CO₂ reduction, and increased innovation in environmental measures.

For a comprehensive analysis of the energy transformation results and the identification of relationships between technical efficiency and institutional readiness, a combination of two key indicators—DEA Efficiency Score and ETI Score—was used. While the first indicator reflects a country’s ability to efficiently utilize available resources to achieve specified outcomes, the second characterizes the overall strategic, political, and market readiness for transitioning to new energy models.

For visual representation and further analytical interpretation, these two indicators were consolidated into a joint table (

Table 4. Values of the ITE and DEA Indicators.

#	Country	DEA Efficiency Index	ETI
1	Austria	1	69.3
2	Germany	1	67.5
3	Malta	1	54.9
4	Netherlands	1	68.8
5	Luxembourg	1	64.2
6	Sweden	1	78.5
7	Poland	1	59.7
8	Belgium	1	59.2
9	Ireland	1	59.3
10	Cyprus	1	56.4
11	France	1	70.6
12	Finland	1	72.8
13	Estonia	1	68.2
14	Croatia	1	62.0
15	Czech Republic	1	58.6
16	Denmark	1	76.1
17	Portugal	0.8951	65.8
18	Lithuania	0.7847	61.2
19	Greece	0.7498	60.9
20	Spain	0.7076	65.0
21	Latvia	0.7037	63.4
22	Romania	0.6447	56.8
23	Bulgaria	0.4903	57.2
24	Italy	0.4682	60.6
25	Slovenia	0.4493	62.6
26	Hungary	0.3680	64.3
27	Slovakia	0.3576	58.8

), which enables comparative evaluation of countries across both dimensions simultaneously. This approach not only allows for ranking countries but also identifies the gaps between potential (ETI) and actual effectiveness (DEA), which is extremely valuable for making informed management and strategic decisions.

The discrimination power of the DEA model was evaluated using efficiency score statistics. Out of 27 EU countries analyzed, 16 were identified as fully efficient (DEA = 1), accounting for 59.3% of the sample. The average efficiency score was 0.8377, with a minimum of 0.3576 and a standard deviation of 0.2293. Although the proportion of fully efficient DMUs is relatively high, the dispersion of values (0.3576–1.00) confirms the model’s ability to differentiate between countries. The ratio of DMUs to input–output variables (n/k = 27/9 ≥ 3) meets the recommended threshold, indicating an acceptable level of discrimination capacity and robustness of the DEA model.

The results of the OAT sensitive analysis are demonstrated in Figure 3.

The OAT sensitivity analysis shows that DEA efficiency is most affected by variations in Investments and Decoupling Score (DS), moderately influenced by Import Dependency, Renewable Share, and Job Vacancy, while indicators such as IDU and EC_per_capita remain relatively robust, confirming the overall stability of the model.

To deepen the interpretation of DEA analysis results and identify strategic profiles of countries, it was deemed appropriate to construct an efficiency matrix that combines two key indicators—DEA Score and ETI Rank. In this context, the DEA Score is considered a generalized indicator of technical efficiency, calculated based on the DEA model, while the ETI Rank serves as an indicator of the input conditions of energy transformation: institutional, political, and market readiness, where a lower value indicates a higher position in the global ranking.

To justify the feasibility of constructing an efficiency matrix, we check the dependency between technical efficiency (DEA Efficiency Score) and the overall readiness for energy transition (ETI), the Pearson correlation coefficient was calculated (Formula (11)). The presence of a weak direct correlation (r = 0.331) with a not statistically significant p-value (p ≥ 0.05) makes the construction of a two-vector efficiency matrix particularly appropriate and correct. This decision is justified by the fact that such a matrix turns a quantitative indicator (weak correlation) into a powerful tool for qualitative and strategic analysis. The weak correlation between technical efficiency (DEA) and energy transition readiness (ETI) means that these two parameters, although related, do not move synchronously. That is, high technical efficiency does not guarantee high readiness for change, and conversely, companies ready for the transition may not yet have achieved a high level of efficiency. Thus, the relationship exists, but it is not linear or predictive. That is why the matrix will allow us to get a “map” that shows how different entities (enterprises, industries) are distributed along two key dimensions and to isolate and classify countries into four clearly defined groups, each of which requires a unique approach.

To systematize the results, a 2 × 2 classification matrix was applied, constructed by dividing the space into four quadrants. The classification is based on a matrix built using two indicators—DEA Score and ETI Rank—with the determination of average values for each axis, allowing the space to be divided into four quadrants and countries to be classified based on the combination of technical efficiency and institutional readiness for the energy transition.

The construction of such a matrix allows for moving from abstract country ranking to a practically oriented typology, which can serve as the basis for recommendations on policy improvement strategies, adaptation of institutional approaches, and optimization of resource utilization in the context of the energy transition. In the future, this classification could be expanded by integrating additional indicators (such as environmental pressure, investment rates, or social justice), which would allow for the formation of a more comprehensive typology of countries for sustainable development purposes.

Thus, constructing the efficiency matrix based on DEA Score and ETI Rank is the logical next step in the research, providing a deeper understanding of the relationship between efficiency, preparedness, and outcomes of the energy transition. The results of the research were visualized in the form of a two-dimensional matrix combining two key parameters: DEA Score (on the X-axis), which reflects the technical efficiency of resource use in the energy transition process, and ETI Rank (on the Y-axis), an integral indicator of a country’s overall readiness for energy changes, taking into account political, market, institutional, and technological aspects (Figure 4). This approach allows for identifying not only absolute leaders but also countries that demonstrate relative efficiency under limited resources or, conversely, suboptimal results despite a favorable environment.

The presented scatter plot, hereinafter referred to as energy transition effectiveness matrix, visually displays the effectiveness of different European countries based on two key dimensions: the DEA efficiency score (on the X-axis) and the ETI rank (on the Y-axis). This matrix is divided into four different quadrants based on median values for both efficiency and ETI, allowing for clear categorization and comparison of national indicators.

The X-axis (DEA Efficiency Score) quantitatively assesses the relative effectiveness of countries in converting input resources into output results, with a score of 1.0 indicating full efficiency relative to the observed group of peers (the efficiency frontier). Countries positioned to the right are more efficient. The Y-axis (ETI Rank) reflects the country’s progress and readiness for energy transition, with higher values (further up the axis) indicating better scores or more advanced transitions. The matrix uses a quadrant classification system, where the dashed lines indicate the median DEA efficiency score and median ETI rank, thus segmenting the chart into four areas. The horizontal dashed line indicates the median ETI score, while the vertical dashed line corresponds to DEA Score = 1.0. These lines divide the matrix into four quadrants, each characterized by a specific combination of efficiency and ETI score.

According to the legend, countries are grouped by colour reflecting their position in these four quadrants:

High DEA, High ETI (marked in blue): This quadrant represents countries that demonstrate high relative efficiency (DEA = 1.0) and high ETI scores. These are the leaders and “benchmarks” in the energy transition sample (Sweden, Finland, Denmark, Germany, Luxembourg, the Netherlands, France, Estonia, and Austria). Their task is to maintain the edge: build out storage and hydrogen infrastructure, electrify hard-to-abate sectors, pursue deep renovation of buildings, and develop local energy communities. They can serve as practice donors for neighbors and pilot the next wave of innovations (carbon contracts for difference, digital grids, retail flexibility markets) to remain in the upper-right quadrant.

High DEA, Low ETI (marked in green): countries in this quadrant are also on the efficiency frontier (DEA = 1.0), but their ETI score is below the median. This indicates the “latent reserve”: technically efficient systems that lag on aggregate outcomes and policy indicators (Poland, Croatia, Belgium, Ireland, Czech Republic, Cyprus, and Malta). They need stable, predictable delivery frameworks: long-term auctions for renewables, permitting reform, development of flexibility markets, scaling corporate demand for green electricity, and a “green” taxonomy for capital. The key is to move upward—i.e., convert existing efficiency into a higher ETI through better policies, investment, and social support.

Low DEA, High ETI (marked in red): This quadrant includes countries with relatively lower efficiency (DEA < 1.0) but with ETI scores above the median. These countries have relatively strong policies and institutions, but with “structural costs” on efficiency (Hungary, Latvia, Portugal, Spain, and partly Slovenia). The focus should be on micro-reforms of markets and sectoral reorientation: boosting system flexibility (storage, demand response), accelerating decarbonization of heat and transport, clearing local network bottlenecks, and speeding up innovation diffusion in industry. The goal is to shift rightward without losing altitude.

Low DEA, Low ETI (marked in yellow): This is the quadrant for countries that exhibit lower scores for both relative efficiency (DEA < 1.0) and ETI. These countries lack both technical efficiency and transition outcomes (typically: Italy, Bulgaria, Slovakia, Lithuania, Romania, and Greece). Their priority is the “basic infrastructure of the transition”: grid and generation upgrades, lowering energy intensity, tackling energy poverty, and transparent incentives for renewables and energy efficiency in housing and SMEs. Quick wins come from thermal retrofits, dynamic tariffs/DSM (demand-side management), and removing permitting barriers for distributed generation.

The matrix effectively visualizes the reference links between countries with lower DEA scores and those on the efficiency frontier. These reference links, indicated by dashed gray lines, illustrate potential benchmarking relationships, where less efficient countries “refer” to more efficient ones as models of best practices.

Thus, the effectiveness matrix categorizes countries based on their DEA and ETI indicators. It allows for the quick identification of leaders for considered case (“Best”—Sweden) and countries that need the most improvement (“Worst”—Slovakia), as well as highlighting groups of countries facing similar challenges. Additionally, the matrix is complemented by reference links. These links, obtained from the DEA analysis, indicate benchmarking relationships, where less effective countries “refer” to more effective ones as models of best practices. The central element of the presented visualization is the network of lines connecting countries with a DEA score below 1.0 to countries with a DEA score of 1.0. These lines are not random; they are a direct result of the DEA and visualize the concept of “reference groups” or “benchmarks.” Each line indicates a specific effective country serving as a reference for a less effective country, demonstrating how the latter could improve its efficiency by following the practices of the “best in class.” For each inefficient country, DEA identifies one or more effective counterparts that form its reference group (Figure 5).

In essence, the DEA method enables the determination of a country’s relative efficiency by identifying the “efficiency frontier.” This frontier is formed by the most efficient countries in the sample, which have a DEA score of 1.0. Countries located on this vertical frontier (e.g., Sweden, Finland, Germany, and several others) are considered benchmark or reference countries as they demonstrate the best practices in achieving efficiency. The gray dashed lines departing from countries with a DEA score below 1.0 visually represent these reference links. They connect countries with the potential to improve their efficiency in DEA to their effective counterparts or references on the efficiency frontier. Each such link suggests that a country with lower relative efficiency according to DEA could improve its efficiency by approaching the level of its reference countries. This can be achieved by studying and implementing the best practices applied in reference countries, regardless of their ETI score. Countries like Slovakia, which demonstrate a low DEA score (around 0.3576), have multiple reference links. These links are directed towards countries with higher DEA efficiency, such as France, Germany, Croatia, and Luxemburg which are on the efficiency frontier. Another link involves Hungary, with a DEA score of around 0.3680, which also has an extensive network of reference links. Its references are high-performing countries in terms of DEA, such as Germany (which has the highest ETI score among the effective countries), as well as Luxembourg, Croatia and Malta.

Thus, the reference links on this graph are not just lines, but a visual roadmap for improving relative efficiency. They enable the identification of specific benchmarks for countries with the potential to improve their DEA scores, promoting the exchange of knowledge and best practices in energy transition and enhancing operational efficiency. This enables the development of more targeted policies and investment strategies for achieving a sustainable energy future. Thus, the visualization of reference groups through connecting lines transforms DEA results into concrete, actionable recommendations, clearly demonstrating that improving efficiency and accelerating the energy transition are not isolated tasks, but interconnected processes where the experience of leaders can serve as a catalyst for transforming others.

Table 5. Classification of EU countries according to the efficiency matrix based on the integrated use of DEA and ETI, and distribution of publications dedicated to the study of current energy transition results in EU countries.

#	Arrears of Integrated Use of DEA and ETI	Countries and Sources of Publication	EU Green Deal Instruments and Funding Mechanisms for Energy Transition Acceleration
1	High DEA, High ETI	Sweden [63,64,65]	Priorities should be given to the Innovation Fund and the deployment of activities within Horizon Europe together with the Connecting Europe Facility (CEF). This is an EU funding instrument that co-finances cross-border infrastructure that provides EU level. The InvestEU program brings a new wave of funding for innovation and job creation in Europe. The EU LIFE programme is a financing instrument for environmental and climate change measures, contributing to fundamental changes in the environment and people’s lives.
		Denmark [66]
		Finland [67,68]
		France [69,70,71]
		Austria [72]
		Netherlands [73]
		Estonia [74,75]
		Germany [76,77]
		Luxembourg [78,79,80]
2	High DEA, Low ETI	Croatia [81]	REPowerEU and the Recovery and Resilience Facility (RRF) as well as the Modernisation Fund should be used to accelerate the deployment of RES and electrification of heat, while the Technical Support Instrument (TSI) addresses permitting bottlenecks and market reforms.
		Poland [82,83,84]
		Belgium [85]
		Ireland [86]
		Czech Republic [87]
		Cyprus [88,89]
		Malta [90]
3	Low DEA, High ETI	Portugal [91,92]	It is proposed to combine efforts to comply with the Renewable Energy Directive (RED) and the European Endowment for Democracy (EED) with the InvestEU programme and LIFE projects that increase operational efficiency through digitalisation, smart metering and ESCO aggregation models.
		Spain [93]
		Latvia [94,95]
		Slovenia [96]
		Hungary [97]
4	Low DEA, Low ETI	Italy [98]	It is recommended to use the European Regional Development Fund and the Cohesion Fund (ERDF/CF) to build institutional capacity and strengthen networks, to continue mobilising the Modernisation Fund to replace assets and ensure flexibility, and to use the Social Climate Fund to protect vulnerable consumers and ensure a just transition
		Slovakia [99,100]
		Lithuania [101]
		Bulgaria [102]
		Romania [103]
		Greece [104,105,106]

highlights the study of the energy transition issue in EU countries, providing detailed insights that allow for identifying priorities for best practices in energy change.

Overall, the results of the matrix are significant for the formation of state policy, strategic planning, and monitoring. First, they allow for the identification of reference countries whose practices can serve as models for others. Second, the matrix helps to identify countries where strategic preparation does not translate into effective results, which provides a basis for revising policies and management decisions. Third, it can be used as a tool for assessing the investment attractiveness of countries based on the efficiency of resource use, rather than just formal positions in rankings. Finally, the combination of DEA and ETI strengthens the analytical foundation for adaptive energy transition management in line with the Sustainable Development Goals.

5. Discussion

This study developed a conceptual approach for integrating DEA methods and the ETI to evaluate the effectiveness of energy transformation in European countries. At the same time, it is important to highlight several discussion points that may limit the generality of the conclusions and create a basis for future research.

5.1. Discussions of the Research

Both DEA and ETI are indices based on selected inputs and outputs. Changing the list of these inputs or their weighting can significantly change the final indicators of efficiency and readiness for transition.

The four quadrants in the efficiency matrix are demarcated by the median values of DEA and ETI. The median is only a statistical measure of the middle of the sample, not an absolute criterion of “good” or “bad”. Therefore, a country close to the median may be classified as “efficient”, while another, just below, may be classified as “less efficient”, although the actual difference between them is minimal. This creates a “border effect” that may seem unfair.

All countries that are near the median lines or on the borders between quadrants are considered “gray areas.” Their belonging to one group or another can be a matter of debate, as small changes in the data can move them from one quadrant to another.

Two basic specifications are commonly used in DEA: CCR (CRS) and BCC (VRS). Their fundamental difference is in the assumption of returns to scale: CCR assumes constant returns and builds a single “global” frontier, estimating overall technical efficiency; BCC allows for variable returns, adds a convexity condition, forms a “local” frontier, and measures net technical efficiency. Comparing the results of CCR and BCC enables the separation of scale effects, represented by the difference between overall and net efficiency.

We deliberately use the output-oriented CCR (constant returns to scale) because the research question concerns absolute transformation efficiency—how well countries transform inputs into energy transition outputs—regardless of scale. Most of the variables in our panel are shares, intensities, and per capita rates, which already normalize for size and weaken the need for the variable returns assumption. The single CCR frontier provides a unified “best practice” benchmark for cross-country comparison, offering higher discrimination power even under relatively small sample sizes and rich indicator sets. For robustness, we also present the BCC results in

Table A3. Completed DEA results, sorted by CCR and BCC.

#	Country	CCR	BCC
1	Austria	1	1
2	Germany	1	1
3	Malta	1	1
4	Netherlands	1	1
5	Luxembourg	1	1
6	Sweden	1	1
7	Poland	1	1
8	Belgium	1	1
9	Ireland	1	1
10	Cyprus	1	1
11	France	1	1
12	Finland	1	1
13	Estonia	1	1
14	Croatia	1	1
15	Czech Republic	1	1
16	Denmark	1	1
17	Portugal	0.8951	0.9001
18	Lithuania	0.7847	0.8655
19	Greece	0.7498	0.8531
20	Spain	0.7076	0.8242
21	Latvia	0.7037	0.7864
22	Romania	0.6447	0.7751
23	Bulgaria	0.4903	0.6556
24	Italy	0.4682	0.8704
25	Slovenia	0.4493	0.6609
26	Hungary	0.3680	0.6267
27	Slovakia	0.3576	0.6068

of Appendix C; the resulting ranks are stable.

5.2. Limitations of the Research

Firstly, the analysis was conducted based on data from only one year. This approach was driven by the need to create and test a conceptual framework for assessing the effectiveness of the energy transition environment. However, future studies should consider the dynamics of changes in input and output parameters of the model, taking into account time series data. This would allow tracking the sustainability of efficiency over time, evaluating the impact of political or economic changes, and identifying the delayed effects of implemented reforms. Secondly, despite the ability of the DEA to consider various factors, the multidimensionality of the indicators used (economic, technical, and social) makes it difficult to normalize data between countries with different economic scales and energy consumption structures. This limits the ability to consider external factors, such as geographical, war’s consequences, political, etc. The integration of DEA analysis with the ETI partially mitigates these limitations. DEA allows the assessment of technical efficiency based on multiple parameters without the need for prior functional form assumptions, while ETI serves as an aggregated indicator of institutional and market readiness for energy changes. Combining these approaches enables the identification of countries that, despite a low ETI rating, demonstrate effective resource use, or conversely, reveal inefficiency despite a high overall rating. Thirdly, the DEA method has certain limitations related to the sensitivity of the choice of input and output variables. Although the variables were justified based on a literature review and previous research results, different combinations of parameters can lead to different outcomes. This requires further validation of the model to correct any potential statistical distortions. Fourthly, the ETI, which is used as a component for comparison, has its own internal calculation methodology, which may not account for all contextual or regional specifics, such as institutional capacity or social readiness. This creates a risk of oversimplification when comparing countries that differ structurally or historically. Fifthly, effectiveness in this study is primarily considered in terms of resource use and macroeconomic parameters. The social aspects of the energy transition, such as energy poverty, the inclusivity of policies, and the level of public participation and stakeholder engagement, are insufficiently covered [107]. Future research should integrate qualitative indicators and just transition measures.

While such a matrix is a powerful tool for visualization and strategic analysis, it is only a model of reality. Its classification serves as a starting point for discussions, not a final verdict. Users of the analysis should consider the context, methodological limitations, and dynamics of change to avoid simplistic conclusions.

In conclusion, although this study provides a solid foundation for the integrated assessment of the energy transition, it also opens several avenues for model refinement, including the incorporation of time dynamics, multidimensional criteria, error-correction methods, and an expanded variable set that captures regional specificities.

We restricted the baseline DEA model to EU countries, where harmonized indicators are available and the heat sector is structurally comparable. Pooling non-EU regions into a single frontier would violate the common-technology assumption, given substantial technological and institutional differences, and could bias the estimated frontier and mislead policy interpretation. Accordingly, the external generalizability of our results is limited to the European context. Extending the proposed evaluation framework to assess how the energy-system environment affects the effectiveness of the energy transition in Asia, Africa, and Latin America is planned as future work, with appropriate adaptations to the specific development conditions of these regions.

6. Conclusions

As a result of the conducted research, a few theoretical and practical conclusions were formulated, which define its scientific novelty and practical significance. The relevance of DEA was substantiated due to the need to assess the rationality of resource expenditures, particularly during the energy transition process, to identify systemic weaknesses, and to conduct comparative analyses of countries based on real efficiency outcomes. The methodological integration of DEA with multi-criteria analysis in the field of energy policy has been further developed, enabling a comprehensive assessment of the effectiveness of the energy transition that considers a wide range of parameters relevant to the energy transition. An innovative methodology for comparative analysis was developed, allowing for the inclusion of a wide variety of input and output factors, ensuring a more accurate interpretation of countries’ effectiveness in the dynamics of transition to sustainable energy systems.

The research expands the potential of DEA as a strategic-level analytical tool capable of supporting well-grounded policy decisions. It was demonstrated that operational research tools can serve as an effective platform for developing adaptive energy policies that focus on achieving energy transition results. The relevance of evaluating the effectiveness of the energy transition is determined not only by the high resource intensity of this process but also by the unequal starting conditions of countries, differing levels of institutional capacity, access to technologies, and political will.

The integration of DEA and ETI results allows for the creation of a two-dimensional analytical matrix that reflects not only the level of resource efficiency but also countries’ readiness for energy transformation. Based on the constructed matrix, all countries were grouped into four types according to DEA efficiency scores (on the X-axis) and ETI ratings (on the Y-axis).

• High DEA, High ETI (blue points): Countries in this quadrant, such as Sweden, Denmark, Finland, Germany, France, Estonia, Austria, Luxembourg, and the Netherlands, demonstrate high efficiency and successful energy transition. They are leaders who effectively use resources and actively develop RES, serving as benchmarks for other countries.
• High DEA, Low ETI (green points): Croatia, Poland, Belgium, Ireland, Czech Republic, Cyprus, and Malta, while effectively using their resources, show low progress in energy transition. This may indicate their focus on optimizing existing energy systems rather than rapidly transitioning to new energy sources.
• Low DEA, High ETI (red points): Hungary, Latvia, Spain, Portugal, and partly Slovenia shows significant progress in the energy transition, but their internal processes and resource use are still suboptimal. This highlights the need to improve efficiency to achieve greater results in the transition.
• Low DEA, Low ETI (yellow points): countries such as Italy, Slovakia, Bulgaria, Romania, Greece, and Lithuania face challenges both in efficiency and readiness for energy transition. These countries require a comprehensive approach to improving resource efficiency and accelerating the transition to sustainable energy systems.

Thus, combining the DEA and ETI methods not only allows for the analysis of the effectiveness of energy changes but also identifies barriers and policy priorities for each country. The resulting typology enables more accurate targeting of strategies, considering both current effectiveness and readiness for implementing changes, which is crucial for achieving a just and inclusive energy transition in countries with varying starting points.

6.1. Verification of the Results Obtained

A comparison of the DEA × ETI matrix results with Decoupling Score (DS) typology shows broad alignment in identifying profiles, while each metric captures a different facet of transition performance. DS leaders with absolute decoupling (DS = 1)—Estonia, Luxembourg, Finland, and Sweden—also sit in the matrix’s upper-right quadrant (high technical efficiency and high transition outcomes), indicating that growth is supported by structurally efficient systems and mature policies. The high-decoupling DS group—Denmark, Ireland, France, Poland, and Malta—falls in the right-hand quadrants of the matrix, where both outcomes and efficiency remain high, consistent with improving DS dynamics. By contrast, Romania lies in the lower-left quadrant, indicating structural efficiency without a correspondingly high ETI. The “green recession” cases—Belgium, Lithuania, the Netherlands—are located towards the leaders (at the DEA median), consistent with the idea that their emission reductions rely on sustained improvement potential. While Slovakia, Slovenia, Italy, Greece agree with the idea that their emission reductions should be converted into sustained intensity increases so that when growth resumes, they consolidate on the right side of the matrix. The sharpest divergence is observed in Germany, where DS = 0 due to a one-year pattern of GDP contraction accompanied by rising emissions. In contrast, the DEA = 1 indicator for Germany reflects its position at the frontier of best practices in converting resources into energy transition outcomes. After a sharp reduction in Russian gas supplies in 2022, a fuel switch to coal occurred, which increased emissions despite a decline in energy consumption and falling GDP in late 2022—early 2023 [108]. Thus, DS reflects the situational effect of the supply crisis, while DEA reflects a more stable structural capacity of the system. This underlines the complementarity of the two lenses: DS is sensitive to shocks and short-run disruptions (weather, war, fuel switching, outages), whereas the matrix reflects a more structural picture of efficiency and policy quality. In practice, DS offers a dynamic barometer of how clean growth was in the period observed, while DEA × ETI indicates how reliably the system can translate investments and governance into sustained results. Used together, they clarify priorities: countries with high DS but a “left-side” position should invest in grids, the heat sector, and flexibility to convert momentum into durable performance, while those in the lower-right should strengthen policy frameworks and social support to move upward.

The analysis of the correlation matrix of input and output variables for DEA in

Table A4. Correlation matrix of DEA inputs and outputs.

RES_share	1	−0.62	−0.73	−0.02	0.18	0.38	0.26	−0.02	0.07
JV_PST	−0.62	1	0.60	−0.07	0.10	−0.36	0.08	0.14	0.08
DS	−0.73	0.60	1	−0.24	−0.03	−0.33	−0.19	−0.04	−0.11
Imp_dep	−0.02	−0.07	−0.24	1	0.24	−0.08	0.31	0.03	0.04
Fossi_share	0.18	0.10	−0.03	0.24	1	−0.23	0.59	0.12	0.42
CCM_inv	0.38	−0.36	−0.33	−0.08	−0.23	1	0.21	0.14	0.43
EIS	0.26	0.08	−0.19	0.31	0.59	0.21	1	0.27	0.58
IDUo	−0.02	0.14	−0.04	0.03	0.12	0.14	0.27	1	0.21
EC_per_capita	0.07	0.08	−0.11	0.04	0.42	0.43	0.58	0.21	1
	RES_share	JV_PST	DS	Imp_dep	Fossi_share	CCM_inv	EIS	IDU	EC_per_capita

of Appendix D showed several key dependencies, which allow us to justify the choice of indicators. The strong negative correlation between the share of fossil fuels and the share of renewable sources (r = −0.73) and between import dependence and renewable sources (r = −0.62) logically reflects the interchangeability of these factors in the structure of the energy balance. At the same time, there is a positive relationship between import dependence and fossil fuels (r = 0.60), which confirms the vulnerability of countries with a high share of fossil energy. The Innovation Index (EIS) has a moderate positive correlation with energy consumption per capita (r = 0.58) and the level of vacancies in the sector (r = 0.59), which indicates the relationship of technological development with energy demand and the labor market. The negative relationship between DS and energy consumption per capita (r = −0.43), as well as with the share of RES (r = −0.38) indicates that even in countries with high levels of energy consumption and development of green generation, a complete gap between economic growth and emission reduction is not always achieved. Low or weak correlations for investment and infrastructure development indicators confirm their independent nature, which makes it advisable to include them in DEA as factors that can reveal hidden efficiency. Thus, the selected system of indicators provides a balance between closely related and independent variables, which allows an adequate assessment of the efficiency of the energy transition.

6.2. Applications of the Research Findings

The practical significance of the results lies in their potential use in management, policy, and investment. Specifically, the research allows identifying best practices for dissemination, adjusting national strategies based on the actual effectiveness of resource use, forming a well-reasoned investment appeal, and conducting monitoring and policy reviews within international cooperation frameworks such as EU programs, OECD, or other international organizations. The use of reference groups for countries with lower efficiency scores enables a focus on countries with higher results, which can serve as models for improving energy transition strategies in less efficient countries. This also enables the transfer of best practices from countries that have already succeeded in the energy transition to other lagging countries.

From a practical standpoint, the results underscore the need for policymakers to prioritize the deployment of renewable energy and labor market transformation in their national transition strategies. Since efficiency reacts most strongly to these dimensions, targeted interventions in these areas are likely to deliver the highest marginal impact on overall transition performance. The practical significance of the findings is summarized in

Table A5. Practical relevance of the research results to specific EU energy transition challenges.

EU Challenges	Measures	DEA Variables and Its Usage Explanation
Import dependence	Grants and concessional loans for fast-track projects that reduce consumption of natural gas and oil; modernization and expansion of electricity and heat networks; deployment of renewable energy capacities; installation of electricity storage systems (battery energy storage); development and decarbonization of district heating; improvements in industrial energy efficiency.	Core:
		Imp_dep, %—energy import dependency;
		Fossil_share, %—a lower share implies less import demand;
		RES_share, %—domestic supply substitutes imports;
		EIS—innovation potential reduces import needs;
		EC_per_capita—energy lower use causes lower imports.
		Supporting:
		CCM_inv, million euro—insulation, heat pumps, district heating reduce demand for imported fuels;
		Cross-cutting/context:
		DS—indicates whether reducing use/emissions occurs without harming the economy;
		ETI—overall policy readiness and quality.
European Green Deal	Scaling portfolios of thermal renovation for residential and public buildings; development and upgrade of district-heating systems; launch and support of energy-service models; implementation of national renovation programs; deployment of digital tools (building renovation passports, consumption monitoring systems); support for vulnerable households; stimulation of low-carbon technologies in industry; rollout of innovative solutions in buildings and heat supply.	Core:
		RES_share, %—renewable rollout;
		EIS—efficiency gains;
		EC_per_capita—lower consumption in buildings/sectors due to renovation;
		CCM_inv, million euro—investments in carbon-abatement
		measures, renovation, and heat supply.
		Supporting:
		JV_PST, %—workforce for the renovation wave, engineering, heat-pump service;
		IDU, million euro—cities’ ability to finance renovation/heat networks).
		Cross-cutting/context:
		Imp_dep, %,—energy efficiency targets and electrification cut the import
		Fossil_share, %—the Green Deal reduces them over time;
		ETI—alignment of actions with energy transition goals;
		DS—framework indicators of policy quality and dynamics.
Post-crisis recovery	Priority to regions moving away from coal and natural gas; programs for worker reskilling and up skilling; accelerated investment in network infrastructure, heat generation and heat systems; measures to increase power-system flexibility (demand response, thermal and electricity storage); support for localization of equipment and component manufacturing (e.g., heat pumps and their components); financing of urban-resilience projects (microgrids, thermal and electricity storage, modernization of critical infrastructure).	Core:
		DS—ability to cut emissions/consumption without sacrificing growth (shock resilience);
		Imp_dep, %—lower import dependence means less exposure to price/logistics shocks.
		Supporting:
		EIS – innovation potential creation to provide energy transition
		EC_per_capita—efficiency reduces household/business vulnerability;
		RES_share, %—diversification and local supply that lower supply-risk;
		CCM_inv, million euro—steady crisis-time investment in heat/grids/storage raises reliability;
		IDU, million euro—fiscal capacity of urban areas to sustain resilience programs;
		JV_PST, %—available workforce to scale green solutions quickly.
		Cross-cutting:
		ETI—trajectory and institutional readiness, also relevant for resilience.

of Appendix E, which maps each key EU energy-policy challenge to the corresponding measures, the energy-transition performance indicators used in the article (KPIs), and potential funding sources summed up from analyzed literature.

Thus, multi-factor data analysis based on the efficiency matrix allows not only to assess the current state of energy transition in EU countries, but also to formulate comprehensive and targeted strategic recommendations. The comparison of technical efficiency (DEA) and energy transition readiness (ETI) allows each country to determine its unique place in the context of energy transition and obtain a visualized “roadmap”. This helps countries develop individual strategies to optimize their energy systems and achieve sustainable development goals within the framework of the energy transition.