How IoT and machine learning could help factories use less energy

Manufacturing is one of the world's largest energy-consuming sectors, and industries are under growing pressure to improve efficiency while reducing energy use and environmental impact. As production systems become increasingly connected, factories are generating large amounts of operational data. A review published in Cyber-Physical Energy Systems examines how combining the Internet of Things (IoT) with machine learning (ML) can turn these data into strategies for improving energy efficiency in advanced manufacturing. The findings provide insights into how data-driven technologies can contribute to more sustainable and intelligent manufacturing while supporting future research and industrial applications.

The review, authored by Mohsen Soori of Final International University, Türkiye, and Behrooz Arezoo of Amirkabir University of Technology (Tehran Polytechnic), Iran, examines recent methodologies and applications of IoT- and ML-based approaches for energy saving in manufacturing. IoT technologies enable real-time monitoring of equipment operation, energy consumption, and production conditions, while ML algorithms analyze these data to identify patterns, predict future trends, and support automated decision-making.

The authors highlight several major applications of these technologies. ML-based predictive maintenance can forecast equipment failures and identify abnormal energy consumption, helping manufacturers prevent inefficient operation and reduce unnecessary energy losses. Energy consumption forecasting models can support production planning and energy management by predicting future demand, while different ML approaches provide advantages in different scenarios depending on data availability, computational requirements, and process complexity.

In addition, ML-driven optimization methods can adjust process parameters such as temperature, pressure, and speed to improve energy efficiency while maintaining production requirements.

Beyond individual production processes, the authors discuss the broader applications of IoT and ML in supply chain management, resource allocation, demand response, smart grid integration, and renewable energy utilization. In particular, hybrid models combining ML with physics-based simulation can forecast renewable energy availability more accurately than standalone algorithms.

The authors also identify several challenges that may affect wider adoption of IoT and ML for energy-efficient manufacturing, including data quality, model interpretability, cybersecurity, scalability, integration with existing industrial systems, and the need for specialized expertise.

###

Author Details:

Mohsen Soori*, Assistant Professor in Mechanical Engineering, Final International University, Kyrenia, Cyprus

Corresponding author: m.soori@aut.ac.ir

(Affiliation at the time of manuscript submission; current affiliation: Istanbul Aydın University, Istanbul, Türkiye.) 

Behrooz Arezoo, CAD/CAPP/CAM Research Center, Department of Mechanical Engineering, Amirkabir University of Technology (Tehran Polytechnic), Tehran 15875-4413, Iran

The publisher KeAi was established by Elsevier and China Science Publishing & Media Ltd to unfold quality research globally. In 2013, our focus shifted to open access publishing. We now proudly publish more than 300 world-class, open access, English language journals, spanning all scientific disciplines. Many of these are titles we publish in partnership with prestigious societies and academic institutions, such as the National Natural Science Foundation of China (NSFC).