Dynamic Prediction of Manufacturing Energy Allocation Efficiency Based on Transformer Time-Series Model
Main Article Content
Abstract
Manufacturing is a major sector of energy consumption and carbon emissions, improving energy allocation efficiency is essential for promoting green transformation and low-carbon development. To dynamically predict manufacturing energy allocation efficiency, this study proposes an integrated framework combining efficiency measurement and Transformer-based time-series forecasting. First, an energy allocation efficiency evaluation index system is constructed from the perspectives of energy input, capital input, labor input, desirable output, and undesirable output. The SBM-DEA model with undesirable outputs is then applied to measure manufacturing energy allocation efficiency under economic and environmental constraints. Based on the measured efficiency series and related influencing factors, including industrial structure, energy structure, technological innovation, environmental regulation, and carbon intensity, a multivariate Transformer time-series model is developed to forecast future efficiency trends. Results show that manufacturing energy allocation efficiency presents an overall upward trend, but regional and sectoral heterogeneity remains significant. Compared with ARIMA, SVR, Random Forest, XGBoost, LSTM, GRU, and TCN models, the Transformer model achieves better prediction performance in terms of MAE, RMSE, MAPE, and R2. Further interpretability analysis indicates that technological innovation, energy structure, industrial structure, environmental regulation, and carbon intensity are the key factors affecting the dynamic evolution of energy allocation efficiency. This study provides a useful methodological framework for manufacturing energy efficiency prediction and offers decision-making support for energy optimization, carbon reduction, and green industrial transformation.
Downloads
Article Details
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
References
Y. Chen and B. Lin, “Understanding the green total factor energy efficiency gap between regional manufacturing: insight from infrastructure development,” Energy, vol. 237, p. 121553, 2021.
Y. Chen and Y. Liu, “How biased technological progress sustainably improve the energy efficiency: an empirical research of manufacturing industry in China,” Energy, vol. 230, p. 120823, 2021.
A. Zhou and J. Li, “Investigate the impact of market reforms on the improvement of manufacturing energy efficiency under China’s provincial-level data,” Energy, vol. 228, p. 120562, 2021.
W. Cao and X. Wei, “Regional differences and catch-up analysis of energy efficiency in China’s manufacturing industry under environmental constraints,” Energy Informatics, vol. 7, no. 1, p. 103, 2024.
W. Pan, T. X. Lin, H. Long, et al., “China’s energy economic efficiency evaluation based on novel dynamic network DEA,” Scientific Reports, vol. 15, p. 22406, 2025.
J. Walther and M. Weigold, “A systematic review on predicting and forecasting the electrical energy consumption in the manufacturing industry,” Energies, vol. 14, no. 4, p. 968, 2021.
I. Rojek, D. Mikołajewski, A. Mroziński, et al., “Green energy management in manufacturing based on demand prediction by artificial intelligence: a review,” Electronics, vol. 13, no. 16, p. 3338, 2024.
B. Ioshchikhes, M. Frank, and M. Weigold, “A systematic review of expert systems for improving energy efficiency in the manufacturing industry,” Energies, vol. 17, no. 19, p. 4780, 2024.
H. Ekwaro-Osire, D. Bode, J. H. Ohlendorf, et al., “Manufacturing process energy consumption modeling: a methodology to identify the most appropriate model,” Journal of Intelligent Manufacturing, vol. 36, no. 8, pp. 5673–5693, 2025.
B. Lim, S. O. Arik, N. Loeff, et al., “Temporal fusion transformers for interpretable multi-horizon time series forecasting,” International Journal of Forecasting, vol. 37, no. 4, pp. 1748–1764, 2021.
H. Zhou, S. Zhang, J. Peng, et al., “Informer: beyond efficient Transformer for long sequence time-series forecasting,” in Proc. AAAI Conf. Artif. Intell., vol. 35, no. 12, pp. 11106–11115, 2021.
H. Wu, J. Xu, J. Wang, et al., “Autoformer: decomposition transformers with auto-correlation for long-term series forecasting,” in Advances in Neural Information Processing Systems (NeurIPS), vol. 34, pp. 22419–22430, 2021.
T. Zhou, Z. Ma, Q. Wen, et al., “FEDformer: frequency enhanced decomposed Transformer for long-term series forecasting,” in Proc. 39th Int. Conf. Mach. Learn. (ICML), vol. 162, pp. 27268–27286, 2022.
Y. Nie, N. H. Nguyen, P. Sinthong, et al., “A time series is worth 64 words: long-term forecasting with Transformers,” in Proc. Int. Conf. Learn. Representations (ICLR), 2023.
Y. Liu, T. Hu, H. Zhang, et al., “iTransformer: inverted Transformers are effective for time series forecasting,” in Proc. Int. Conf. Learn. Representations (ICLR), 2024.