Data-Driven Energy Efficiency Assessment and Environmental Performance Optimization in Industrial Systems

Mohammad Ali Niroomand; Rojeen Mamershafai; Fereshteh Khalili Peykani; Omid Shabkhizan; Meysam Mahmoudi

Authors

Mohammad Ali Niroomand Faculty of Engineering, Shahrood University, Shahrood, Iran Author https://orcid.org/0009-0005-3339-2893
Rojeen Mamershafai Department of Energy Systems Engineering, Amirkabir University of Technology, Tehran, Iran Author https://orcid.org/0009-0004-9322-2063
Fereshteh Khalili Peykani Department of Economics and Management, SR.C., Islamic Azad University, Tehran, Iran Author https://orcid.org/0000-0001-9312-7796
Omid Shabkhizan Department of Management, SR.C., Islamic Azad University, Tehran, Iran Author https://orcid.org/0009-0005-7287-502X
Meysam Mahmoudi Department of Mechanical Engineering, Faculty of Engineering, Velayat University, Iranshahr, Iran https://orcid.org/0009-0004-4902-2312

Keywords:

energy efficiency, industrial systems, machine learning, CO₂ emissions intensity, feature importance, multi-objective optimization, predictive analytics, sustainability

Abstract

Industrial energy use accelerates environmental damage and increases greenhouse gas emissions. This data-oriented approach requires the collaboration of energy efficiency and environmental performance. The objective of this study was to design a data-driven methodology to evaluate energy efficiency and enhance environmental performance in industrial systems. Suppose we analyzed a fictitious dataset of 1,250 monthly observations from 125 industrial facilities. In our study, we would have focused on the following indicators: energy consumption, production output, electricity costs, fuel consumption, CO₂ emissions, operating hours, ambient temperature, and production intensity. Energy efficiency was defined as the ratio of production output to total energy consumption, while CO₂ emissions intensity represented the measure of environmental performance. Actually, predicting energy consumption and environmental performance of a facility with Random Forest, XGBoost and Support Vector Regression models was undertaken alongside feature-importance and multi-objective optimization. It was observed that average energy consumption was 4,820 MWh per facility per year and average CO₂ emission intensity was 0.74 tCO₂/MWh. XGBoost was better than all others, with an R2 of 0.91 and RMSE of 0.18, when it came to predicting energy consumption. Feature-importance analysis showed that production intensity (31.4%), operating hours (22.7%), equipment efficiency (18.5%), and ambient temperature (11.2%) were the most influential factors affecting energy consumption. An optimization analysis showed that the energy needs for production could be lowered by 13.6% with the better scheduling of production coupled with improved equipment and the better management of the production equipment’s energy dissipating needs. Improving scheduling and management of production equipment would further lower CO₂ emissions by 15.2% without significantly decreasing the production output. Integrating predictive analytics with the assessment of environmental performance is useful to industrial managers to understand opportunities to decrease energy consumption and develop operational strategies that sustain energy consumption.

References

Azubuike, U. G., Njoku, H. O., Ekechukwu, O. V., & Jen, T. C. (2025). A Comprehensive Review on Advanced Exergy, Exergoeconomic, and Exergoenvironmental Analyses for Sustainable Thermal Energy Systems. International Journal of Exergy, 48(1), 16-53. https://doi.org/10.1504/IJEX.2025.148556

Doğan, F., Oyucu, S., Unsal, D. B., Aksöz, A., & Vafaeipour, M. (2025). Impact of Environmental Conditions on Renewable Energy Prediction: An Investigation Through Tree-Based Community Learning. Applied Sciences, 15(1), 336. https://doi.org/10.3390/app15010336

Feng, Y., & Xu, R. (2025). Advancing Global Sustainability: The Role of the Sharing Economy, Environmental Patents, and Energy Efficiency in the Group of Seven's Path to Sustainable Development. Sustainability, 17(1), 322. https://doi.org/10.3390/su17010322

Ferro, M., Silva, G. D., de Paula, F. B., Vieira, V., & Schulze, B. (2023). Towards a Sustainable Artificial Intelligence: A Case Study of Energy Efficiency in Decision Tree Algorithms. Concurrency and Computation: Practice and Experience, 35(17), e6815. https://doi.org/10.1002/cpe.6815

Haghjou, M., Hayati, B., & Pishbahar, E. (2020). Factors Affecting Consumers' Awareness of Pesticides-Free Fruits and Vegetables. In The Economics of Agriculture and Natural Resources: The Case of Iran (pp. 125-139). Springer Singapore. https://doi.org/10.1007/978-981-15-5250-2_9

Hayati, B., Pishbahar, E., & Haghjou, M. (2012). Analyzing Determinants of Consumers' Willingness to Pay a Premium for Pesticide-Free Fruit and Vegetables in Marand City.

Krzywanski, J., Sosnowski, M., Grabowska, K., Zylka, A., Lasek, L., & Kijo-Kleczkowska, A. (2024). Advanced Computational Methods for Modeling, Prediction and Optimization—A Review. Materials, 17(14), 3521. https://doi.org/10.3390/ma17143521

Long, F., Xu, M., Liao, W., & Liu, H. (2025). Machine Learning for Predicting and Optimizing the Performance of a Commercial-Scale Anaerobic Digester with Diverse Feedstocks and Operating Conditions. Bioresource Technology, 435, 132940. https://doi.org/10.1016/j.biortech.2025.132940

Nozari, H., Szmelter-Jarosz, A., & Samadi, S. (2025). Machine Learning Models for Energy Optimization and Resource Consumption in Smart Factories. In Artificial Intelligence of Everything and Sustainable Development (pp. 175-189). Springer Nature Singapore. https://doi.org/10.1007/978-981-96-7202-8_10

Nsangou, J. C., Kenfack, J., Nzotcha, U., Ekam, P. S. N., Voufo, J., & Tamo, T. T. (2022). Explaining Household Electricity Consumption Using Quantile Regression, Decision Tree and Artificial Neural Network. Energy, 250, 123856. https://doi.org/10.1016/j.energy.2022.123856

Ordonez, J. C., Cavalcanti, E. J., & Carvalho, M. (2022). Energy, Exergy, Entropy Generation Minimization, and Exergoenvironmental Analyses of Energy Systems—A Mini-Review. Frontiers in Sustainability, 3, 902071. https://doi.org/10.3389/frsus.2022.902071

Peng, M., Tian, Z., Xia, G., & Wang, H. (2025). The Exergy Analysis Method of Thermodynamic Process. In Thermal Analysis of Nuclear Power Plants (pp. 71-126). Springer Nature Singapore. https://doi.org/10.1007/978-981-96-8983-5_3

Prošek, T., Keil, P., & Popova, K. (2025). Corrosion Protection and Sustainability: Why Are the Two Concepts Inherently Intertwined. Corrosion and Materials Degradation, 6(3), 38. https://doi.org/10.3390/cmd6030038

Strielkowski, W., Vlasov, A., Selivanov, K., Muraviev, K., & Shakhnov, V. (2023). Prospects and Challenges of the Machine Learning and Data-Driven Methods for the Predictive Analysis of Power Systems: A Review. Energies, 16(10), 4025. https://doi.org/10.3390/en16104025

Tang, T. (2026). Predicting Energy Prices and Renewable Energy Adoption Through an Optimized Tree-Based Learning Framework with Explainable Artificial Intelligence. Scientific reports, 16(1), 6771. https://doi.org/10.1038/s41598-026-35706-z

Zaki, A. M., Zayed, M. E., Bargal, M. H., Saif, A. G. H., Chen, H., Rehman, S., & El-deen, E. S. H. N. (2025). Environmental and Energy Performance Analyses of HVAC Systems in Office Buildings Using Boosted Ensembled Regression Trees: Machine Learning Strategy for Energy Saving of Air Conditioning and Lighting Facilities. Process Safety and Environmental Protection, 198, 107214. https://doi.org/10.1016/j.psep.2025.107214

Zhang, R., Yin, K., Wei, R., Ruan, J., Yang, J., Wang, S., & Wang, Y. (2025). Comprehensive Analysis of Life Cycle Energy Consumption and Environmental Impact of Hydrogen Production Process via Plasma Co-Gasification of Coal and Biomass. Energy, 324, 135976. https://doi.org/10.1016/j.energy.2025.135976