Abstract
National sustainability is pivotal in bolstering resilience against global health crises by fostering robust public health systems, economic stability, and social equity. Understanding the interplay between sustainability and pandemic performance is crucial for pinpointing systemic strengths, vulnerabilities, and policy priorities. This study introduces an innovative analytical framework integrating Multi-Criteria Decision-Making (MCDM), Machine Learning (ML), and Design of Experiments (DoE) to assess how pre-existing sustainability conditions shaped COVID-19 outcomes across 170 countries. By combining MCDM methods with Random Forest and SHAP (SHapley Additive exPlanations) analyses, alongside Response Surface Methodology (RSM), the framework evaluates sustainability indicators from 2003 to 2018. Explainable AI (XAI) tools reveal key resilience drivers, while RSM explores nonlinear relationships and interactions among critical variables, such as access to electricity and Internet usage. The findings highlight stark regional disparities, with Europe exhibiting the highest sustainability and COVID-19 performance scores, driven by robust infrastructure and governance, while Africa and South America face greater challenges due to infrastructural and institutional deficits. Access to electricity (% of population) emerged as the most influential factor, with a high SHAP importance score, underscoring its role in enabling healthcare delivery and economic continuity, followed by financial and digital connectivity (ATMs and Internet usage) and educational attainment (primary completion and secondary enrollment). These insights emphasize the critical role of infrastructural and educational equity in enhancing pandemic resilience. The study provides a transferable analytical framework that can be extended in future research to examine emerging post-pandemic challenges, enabling scenario-based policy testing for future global crises.
| Original language | English |
|---|---|
| Article number | 100350 |
| Number of pages | 14 |
| Journal | World Development Sustainability |
| Volume | 9 |
| DOIs | |
| Publication status | Published - Dec 2026 |
Keywords
- COVID-19 response performance
- Explainable artificial intelligence (XAI)
- Machine learning (ML)
- National sustainability
- Pandemic resilience
- Response surface methodology (RSM)
Fingerprint
Dive into the research topics of 'National sustainability and pandemic resilience: How explainable AI uncovers global patterns and policy insights'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver