All Issue

2025 Vol.16, Issue 3 Preview Page

General Article

Predicting residential CO2 emissions: Machine learning and deep learning approaches for short-term forecasting
Maryam Pournaghi Keykele1*
,
Mehdi Ravanshadnia2
1Master of Construction Management, Department of Civil Engineering, Islamic Azad University, Science and Research Branch, Tehran, Iran
2Associate Professor, Department of Civil Engineering, Islamic Azad University, Science and Research Branch, Tehran, Iran
*Corresponding Author
30 September 2025. pp. 373-387
PDF XML Citation
Abstract

The residential sector is a major contributor to global greenhouse gas emissions, especially in high- energy-consuming countries like the United States. While previous studies have typically focused on long-term or annual emission trends, this research addresses the urgent need for accurate short-term forecasting by predicting daily CO2 emissions. The study compares the performance of six traditional machine learning models—Decision Tree, Random Forest, Ridge Regression, Gradient Boosting, Support Vector Regression, and XGBoost—with a Long Short-Term Memory (LSTM) deep learning model. All models were trained and evaluated using three years of daily U.S. residential CO2 emission data, with careful attention to temporal consistency and seasonal variability. Among the models tested, LSTM demonstrated the highest predictive accuracy, effectively capturing both short-term dependencies and seasonal fluctuations. Unlike traditional models, which heavily relied on engineered lag features, the LSTM model learned temporal patterns directly from sequential data. SHAP analysis further supported these findings, showing that traditional models depended predominantly on the previous day’s emissions, while LSTM dynamically distributed importance across a seven-day window. These results emphasize the effectiveness of deep learning for time-series emission forecasting and offer practical guidance for developing responsive environmental policies. The study also suggests that short-term historical data may be more impactful than extensive archives for forecasting near-term emissions in residential settings.

Keywords
residential sector
CO2 emissions
time-series forecasting
machine learning
deep learning
References
1

I.D.O. Barbosa Júnior, A.N. Macêdo, and V.W.B. Martins, Construction industry and its contributions to achieving the SDGs proposed by the UN: an analysis of sustainable practices. Buildings. 13(5) (2023), 1168.

10.3390/buildings13051168
2

W. Qi, J. Zuo, G. Li, and L. Yao, Residential carbon emission flows embedded in population migration over time in China: A geospatial dynamics analysis. Resources, Conservation and Recycling. 212 (2025), 107919.

10.1016/j.resconrec.2024.107919
3

A.A. Ajala, O.L. Adeoye, O.M. Salami, and A.Y. Jimoh, An examination of daily CO2 emissions prediction through a comparative analysis of machine learning, deep learning, and statistical models. Environmental Science and Pollution Research. (2025), pp. 1-26.

10.21203/rs.3.rs-4648686/v1
4

Z. Huang, J. Wang, L. Bing, Y. Qiu, R. Guo, Y. Yu, M. Ma, L. Niu, D. Tong, and R.M. Andrew, Global carbon uptake of cement carbonation accounts 1930-2021. Earth System Science Data. 15(11) (2023), pp. 4947-4958. DOI: 10.5194/essd-15-4947-2023 -15-4947-2023.

10.5194/essd-15-4947-2023
5

X. Guan, S. Guo, J. Xiong, G. Jia, and J.L. Fan, Energy-related CO2 emissions of urban and rural residential buildings in China: A provincial analysis based on end-use activities. Journal of Building Engineering. 64 (2023), 105686.

10.1016/j.jobe.2022.105686
6

M. Crippa, D. Guizzardi, E. Solazzo, M. Muntean, E. Schaaf, F. Monforti-Ferrario, M. Banja, J. Olivier, G. Grassi, and S. Rossi, GHG emissions of all world countries, EUR 30831 EN, Publications Office of the European Union, Luxembourg, (2021), ISBN 978-92-76-41546-6. DOI: 10.2760/173513.

10.2760/173513
7

S. Martinez, M.M. Delgado, R.M. Marin, and S. Alvarez, Identifying the environmental footprint by source of supply chains for effective policy making: The case of Spanish households consumption. Environmental Science and Pollution Research. 426 (2019), pp. 33451-33465.

10.1007/s11356-019-06296-3
8

B. Goldstein, D. Gounaridis, and J.P. Newell, The carbon footprint of household energy use in the United States. Proceedings of the National Academy of Sciences. 117(32) (2020), pp. 19122-19130.

10.1073/pnas.192220511732690718PMC7431053
9

L. Liu, J. Qu, T.N. Maraseni, Y. Niu, J. Zeng, L. Zhang, and L. Xu, Household CO2 emissions: Current status and future perspectives. International Journal of Environmental Research and Public Health. 17(19) (2020), 7077.

10.3390/ijerph1719707732992633PMC7579624
10

Y. Long, Y. Yoshida, R. Zhang, L. Sun, and Y. Dou, Policy implications from revealing consumption-based carbon footprint of major economic sectors in Japan. Energy Policy. 119 (2018), pp. 339-348.

10.1016/j.enpol.2018.04.052
11

T.N. Maraseni, J.S. Qu, Y. Bian, J. Zeng, and J. Maroulis, Dynamism of household carbon emissions (HCEs) from rural and urban regions of northern and southern China. Environmental Science and Pollution Research. 23 (2016), pp. 20553-20566.

10.1007/s11356-016-7237-5
12

United Nations Development Programme, Emissions Gap Report; UNEP: Nairobi, Kenya [Online], 2024. Available at: http://www.unenvironment.org/emissionsgap [Accessed 01/08/2024].

13

A.M. Nassef, A.G. Olabi, H. Rezk, and M.A. Abdelkareem, Application of artificial intelligence to predict CO2 emissions: critical step towards sustainable environment. Sustainability. 15(9) (2023), 7648.

10.3390/su15097648
14

S. Kumari and S.K. Singh, Machine learning-based time series models for effective CO2 emission prediction in India. Environmental Science and Pollution Research. 30(55) (2023), pp. 116601-116616.

10.1007/s11356-022-21723-8
15

M. Javanmard, Y. Emami, Z. Tang, W. Wang, and P. Tontiwachwuthikul, Forecast energy demand, CO2 emissions and energy resource impacts for the transportation sector. Applied Energy. 338 (2023), 120830.

10.1016/j.apenergy.2023.120830
16

J.B. Liu, X.Y. Yuan, and C.C. Lee, Prediction of carbon emissions in China’s construction industry using an improved grey prediction model. Science of the Total Environment. 938 (2024), 173351.

10.1016/j.scitotenv.2024.173351
17

A. Zavvari, M.B. Jelodar, and M. Sutrisna, Comparing two AI methods for predicting the future trend of New Zealand building projects: Decision Tree and Artificial Neural Network. IOP Conference Series: Earth and Environmental Science. 1101(8) (2022), 082016.

10.1088/1755-1315/1101/8/082016
Information
  • Publisher :Sustainable Building Research Center (ERC) Innovative Durable Building and Infrastructure Research Center
  • Publisher(Ko) :건설구조물 내구성혁신 연구센터
  • Journal Title :International Journal of Sustainable Building Technology and Urban Development
  • Volume : 16
  • No :3
  • Pages :373-387
  • Received Date : 2025-05-10
  • Accepted Date : 2025-07-12
  • DOI :https://doi.org/10.22712/susb.20250024
Journal Informaiton International Journal of Sustainable Building Technology and Urban Development International Journal of Sustainable Building Technology and Urban Development
Online Submission
  • scopus
  • NRF
  • KOFST
  • KISTI Current Status
  • KISTI Cited-by
  • crosscheck
  • orcid
  • open access
  • ccl
  • isc
Journal Informaiton Journal Informaiton - close