EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Characteristics and driving factors of urban resilience and sustainability evolution: an XGBoost-SHAP model approach

Fuyang Yu, Liqiu Xin

Frontiers in Environmental Science · 25.06.2026 · 14

DOI: 10.3389/fenvs.2026.1741475 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Environmental Science, online erschienen am 25.06.2026. Verfasst von Fuyang Yu und 1 weiteren Autor:innen. Die Arbeit nennt 76 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Urban Resilience and Sustainability (USR) are pivotal in addressing the challenges arising from rapid urbanization, climate uncertainty, and socio-economic transformation. Urban sustainability (US) emphasizes balanced long-term development alongside the efficient utilization of resources. Conversely, urban resilience (UR) focuses on a city’s capacity to absorb shocks, recover, and adapt. Despite their distinct emphases, these dimensions are interdependent and mutually reinforcing. This study introduces an integrated analytical framework designed to investigate the spatiotemporal evolution, driving mechanisms, and internal coupling dynamics of USR across 300 Chinese cities from 2000 to 2023. A comprehensive indicator system, incorporating economic, social, and ecological dimensions, has been developed to facilitate this analysis. The USR levels are subsequently quantified using an enhanced entropy-weighted TOPSIS model. Regional disparities and dynamic transitions were examined through Dagum Gini decomposition, kernel density estimation, and a spatial Markov chain. Furthermore, the relative importance and interactions of multidimensional drivers were quantified using XGBoost-SHAP. The study yielded five principal findings. Firstly, USR levels have consistently increased nationwide. This progression has altered the former “east-high, west-low” pattern into a more balanced multi-core structure, signifying notable convergence. Economic development, fiscal self-sufficiency, and social wellbeing serve as significant positive drivers. Conversely, employment pressure and air pollution exert notable negative effects. SHAP dependence and interaction analyses reveal pronounced non-linearities and synergies. These findings underscore the critical importance of coordinated economic and fiscal policies, alongside strategic investment in innovation, to bolster resilience. Integrating interpretable machine learning with system-based evaluation enhances the understanding of the sustainability-resilience nexus. It also deepens the analysis of urban governance. PLS-SEM elucidates the linear pathways through which economic, social, and ecological sustainability foundations contribute to resilience. This approach reveals the internal mechanisms of the US-UR coupling. These insights suggest that enhancing resilience demands a coordinated strategy in socio-economic development. Additionally, it necessitates the creation of strong fiscal buffers and the implementation of targeted innovations specifically adapted to local contexts.

Zitieren

APA

Yu, F., & Xin, L. (2026). Characteristics and driving factors of urban resilience and sustainability evolution: an XGBoost-SHAP model approach. Frontiers in Environmental Science, 14. https://doi.org/10.3389/fenvs.2026.1741475

Vancouver

Yu F, Xin L. Characteristics and driving factors of urban resilience and sustainability evolution: an XGBoost-SHAP model approach. Frontiers in Environmental Science. 2026;14. doi:10.3389/fenvs.2026.1741475

BibTeX
@article{Yu2026,
  author = {Yu, Fuyang and Xin, Liqiu},
  title = {Characteristics and driving factors of urban resilience and sustainability evolution: an XGBoost-SHAP model approach},
  journal = {Frontiers in Environmental Science},
  year = {2026},
  volume = {14},
  doi = {10.3389/fenvs.2026.1741475}
}
Kostenlos starten →