Applications of Machine Learning Techniques in Improving Industrial Energy Efficiency: A Systematic Literature Review

Resumen

Abstract– This systematic literature review compiles and analyzes the main research developments, emerging approaches, and trend lines related to the use of Machine Learning techniques aimed at improving energy efficiency in industrial settings. Using the PRISMA and PICO methodologies, 42 selected studies are evaluated based on inclusion criteria such as recency, thematic relevance, and practical application. The results show that Machine Learning techniques—such as neural networks, decision trees, and hybrid models—have been primarily used for energy consumption prediction, anomaly detection, and industrial process optimization. However, a fragmentation of knowledge is evident, with a predominance of studies conducted in simulated environments and a lack of validation in real-world scenarios. Limitations in the interoperability and scalability of the implemented models are also identified. In this regard, this review proposes a unified conceptual framework to guide future research and integrated industrial applications, promoting intelligent and sustainable energy management aligned with green manufacturing and green information technologies. The findings support the potential of Machine Learning as a key tool in the digital transformation of the industrial sector toward greater energy efficiency and emissions reduction. Keywords — Machine Learning, Energy consumption, Data analysis, Industrial sustainability, Energy optimization