Scholarlink

Background <p>The European Union considers the decarbonization of Europe by 2050 a strategic objective. This necessitates finding the solutions that will support the intricate process of formulating energy policies and decisions with enduring implications for the economy, environment, and social welfare of European Union citizens and beyond. The primary objective of this study is to evaluate the suitability of an artificial intelligence (AI)-based approach in policymaking (control stage) with the aim of achieving a more optimal formulation of the common European energy policy and the policies of individual member states.</p> Results <p>Given the scope and volatility of the data, as well as the research objective, data processing was conducted using clustering—a technique within artificial intelligence (AI)—which is suitable for producing more precise, explainable, and provable (PEP) outcomes. In doing so, the study addresses one of the main obstacles to the broader use of AI in policymaking: the current lack of trust in AI-based solutions. The research was conducted in three stages. In the first stage, energy security indicators were determined based on 13 selected indicators using the aggregate approach. In the second stage, clustering was executed as an unsupervised machine learning method, utilizing the K-means algorithm as the designated learning model. In the third stage, a classifier model utilizing an artificial neural network was proposed. The research findings have revealed that countries exhibiting the highest levels of energy security, and consequently the most favorable conditions for sustainable growth, have different energy portfolios, unique economic frameworks, and differing energy prices.</p> Conclusions <p>The research findings are significant in the domains of energy and environmental policies, decision theory, and AI (with special emphasis on the EU AI Act). The research highlights the efficacy of an interdisciplinary approach and contributes to the studies about the use of AI in policymaking, with emphasis on the improvements that will lead to its greater power and precision—one of the milestones for efficient policymaking. Policymaking based on PEP AI outcomes can be seen as one of the most efficient methods for strategic planning and control of decarbonization of Europe; therefore, the paper also proposes recommendations in this context.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Sustainable and secure energy development of the European Union: artificial intelligence-based approach for policymaking

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年11月22日

查看全文
  • Mirjana Radovanović,
  • Goran Šimić,
  • Jámbor Attila

摘要

Background

The European Union considers the decarbonization of Europe by 2050 a strategic objective. This necessitates finding the solutions that will support the intricate process of formulating energy policies and decisions with enduring implications for the economy, environment, and social welfare of European Union citizens and beyond. The primary objective of this study is to evaluate the suitability of an artificial intelligence (AI)-based approach in policymaking (control stage) with the aim of achieving a more optimal formulation of the common European energy policy and the policies of individual member states.

Results

Given the scope and volatility of the data, as well as the research objective, data processing was conducted using clustering—a technique within artificial intelligence (AI)—which is suitable for producing more precise, explainable, and provable (PEP) outcomes. In doing so, the study addresses one of the main obstacles to the broader use of AI in policymaking: the current lack of trust in AI-based solutions. The research was conducted in three stages. In the first stage, energy security indicators were determined based on 13 selected indicators using the aggregate approach. In the second stage, clustering was executed as an unsupervised machine learning method, utilizing the K-means algorithm as the designated learning model. In the third stage, a classifier model utilizing an artificial neural network was proposed. The research findings have revealed that countries exhibiting the highest levels of energy security, and consequently the most favorable conditions for sustainable growth, have different energy portfolios, unique economic frameworks, and differing energy prices.

Conclusions

The research findings are significant in the domains of energy and environmental policies, decision theory, and AI (with special emphasis on the EU AI Act). The research highlights the efficacy of an interdisciplinary approach and contributes to the studies about the use of AI in policymaking, with emphasis on the improvements that will lead to its greater power and precision—one of the milestones for efficient policymaking. Policymaking based on PEP AI outcomes can be seen as one of the most efficient methods for strategic planning and control of decarbonization of Europe; therefore, the paper also proposes recommendations in this context.