Assessment of the decarbonization efficiency in the European Union : machine learning approach

Abstract

Background: The European Union has established a strategic objective to attain carbon neutrality across the continent by the year 2050; however, this complex undertaking is shaped by a variety of influencing factors. It is particularly important to monitor the effects of such a long-term strategy, as it will influence all aspects of the European Union’s sustainable energy development as well as the welfare of its citizens. Since no universally accepted methodology exists for tracking the effects of decarbonization, the use of machine learning as a method of artificial intelligence is proposed - not only to generate concrete results but also to evaluate its applicability for this purpose. The main objective of this research is to assess the trends of 13 selected energy indicators that are vital to the decarbonization initiative. The research was conducted on a sample of 27 countries for the period from 2013 to 2030 using a novel predictive model developed in the Python runtime environment. Results: The primary findings of the research indicate that the EU is likely to experience significant fluctuations in the values of specific indicators. The anticipated progressive rise in electricity prices is expected across all EU countries, accompanied by an increase in consumption. In addition, the projected growth in energy imports presents a significant challenge that will affect the competitiveness of the European economy and the social standing of its citizens. Particularly disadvantaged in the implementation of the decarbonization strategy will be landlocked countries that are highly dependent on energy imports and therefore vulnerable to fluctuations in prices and security of supply. Also at risk are countries facing difficulties in the deployment and exploitation of renewable energy sources, as well as those with weaker socioeconomic indicators. The results further indicate a rising risk to energy security, even in the wealthiest EU countries. Overall, the projections suggest an increase in CO₂ levels up to 2030, followed by a gradual decline thereafter. A particular challenge for managing the decarbonization strategy lies in the significant fluctuations of the monitored parameters, which hinder planning in every respect. Conclusions: In light of the geopolitical and supply chain shifts post-2022, it is clear that a comprehensive reassessment of the strategies for managing the decarbonization of the European Union economy is necessary. The research findings demonstrated the effectiveness of the proposed machine learning approach, which has potential for enhancement due to its scalability and adaptability. The study provides governance and methodological recommendations.