AI-driven environmental performance index using deep autoencoders for monitoring sustainable development goals
Abstract
This study develops an autoencoder-based environmental performance index (AE_EPI) to assess environmental performance across 38 countries in five regions. Using a deep autoencoder, AE_EPI captures complex nonlinear relationships in high-dimensional, multi-region data and is benchmarked against a principal component analysis-based index (PCA_EPI). Linear Regression, Random Forest (RF), and XGBoost, combined with SHAP analysis, are used to identify key environmental drivers. Results show that AE_EPI outperforms PCA_EPI in reconstruction accuracy, stability, and robustness to noise, while RF yields the most consistent predictions. Globally, electricity access and renewable energy are the dominant drivers. Regionally, governance and innovation are most important in East Asia; urbanization in South Asia; renewable energy in Central Asia; digitalization in MENA; and economic capacity, electricity access, and governance in Sub-Saharan Africa. These findings support region-specific, data-driven policies that advance energy transitions, strengthen institutional capacity, and address local environmental priorities for achieving the Sustainable Development Goals.
Related
- The integrated environmental zoning-based regulation further enhances sustainable development in China
- Corporate governance and Environmental, Social, and Governance practices as drivers of alignment with the sustainable development goals: Evidence from Bangladesh
- Balancing economic growth and environmental protection: A sustainable approach to Malaysia's rare-earth industry
- Sustainable behavior among students: the role of pollution risk perception and environmental policies in Romania and the United States
- Energy, environmental, and economic assessment of sustainable fuels from wet biomass via hydrothermal, sorbent-enhanced gasification, and Fischer-Tropsch route
BibTeX
@article{khan2026aidriven,
title = {{AI-driven environmental performance index using deep autoencoders for monitoring sustainable development goals}},
author = {Abbas Khan and Li Shuangjie and Abdulrahman Alomair and Naif Almusallam and Zia Ur Rehman},
journal = {iScience},
year = {2026},
month = oct,
doi = {10.1016/j.isci.2026.117454},
url = {https://www.cell.com/iscience/fulltext/S2589-0042(26)02832-4},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - JOUR
TI - AI-driven environmental performance index using deep autoencoders for monitoring sustainable development goals
AU - Abbas Khan
AU - Li Shuangjie
AU - Abdulrahman Alomair
AU - Naif Almusallam
AU - Zia Ur Rehman
PY - 2026
DA - 2026/10/14
JO - iScience
DO - 10.1016/j.isci.2026.117454
UR - https://www.cell.com/iscience/fulltext/S2589-0042(26)02832-4
AB - This study develops an autoencoder-based environmental performance index (AE_EPI) to assess environmental performance across 38 countries in five regions. Using a deep autoencoder, AE_EPI captures complex nonlinear relationships in high-dimensional, multi-region data and is benchmarked against a principal component analysis-based index (PCA_EPI). Linear Regression, Random Forest (RF), and XGBoost, combined with SHAP analysis, are used to identify key environmental drivers. Results show that AE_EPI outperforms PCA_EPI in reconstruction accuracy, stability, and robustness to noise, while RF yields the most consistent predictions. Globally, electricity access and renewable energy are the dominant drivers. Regionally, governance and innovation are most important in East Asia; urbanization in South Asia; renewable energy in Central Asia; digitalization in MENA; and economic capacity, electricity access, and governance in Sub-Saharan Africa. These findings support region-specific, data-driven policies that advance energy transitions, strengthen institutional capacity, and address local environmental priorities for achieving the Sustainable Development Goals.
SN - 2589-0042
ER -