Assessing non-linear socio-spatial dynamics in urban resilience using interpretable machine learning: evidence from aging communities

Aging residential communities face compound socio-ecological risks, yet renewal programs may prioritize physical retrofits over social conditions associated with resilience. We examine non-linear associations between social participation and physical infrastructure using data from four aging communities in Dalian, China (687 baseline surveys, spatial audits, and stakeholder interviews). We constructed a 48-indicator Comprehensive Resilience Index (CRI) and analyzed it with Random Forest (RF), Shapley Additive Explanations (SHAP), and panel threshold regression. Permutation importance assigned the largest relative share to social participation (31.2%). SHAP estimates showed larger model-specific contributions for selected social factors than for static physical attributes. The panel threshold model estimated a context-specific threshold of 54.80 on the normalized social participation scale (0–100). Below the threshold, the infrastructure coefficient was not statistically significant ( β = 0.062, p = 0.168); above it, the estimated coefficient was 0.485 ( p < 0.001). These observational results are consistent with a thresholded association between participation and infrastructure-resilience relationships; they do not establish a causal multiplier mechanism or a universal readiness cutoff.

Authors

Institutions

Publication Details

Journal
Scientific Reports
Published
2026-11-17
DOI
https://doi.org/10.1038/s41598-026-79302-7
Primary Topic
Disaster Management and Resilience
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Assessing non-linear socio-spatial dynamics in urban resilience using interpretable machine learning: evidence from aging communities

Xiaofei Zhou, Bingjiang Wang, Xue Wu
3 links
Scientific Reports
Disaster Management and Resilience
article

Assessing non-linear socio-spatial dynamics in urban resilience using interpretable machine learning: evidence from aging communities

Xiaofei Zhou, Bingjiang Wang, Xue Wu
article en

Abstract

Aging residential communities face compound socio-ecological risks, yet renewal programs may prioritize physical retrofits over social conditions associated with resilience. We examine non-linear associations between social participation and physical infrastructure using data from four aging communities in Dalian, China (687 baseline surveys, spatial audits, and stakeholder interviews). We constructed a 48-indicator Comprehensive Resilience Index (CRI) and analyzed it with Random Forest (RF), Shapley Additive Explanations (SHAP), and panel threshold regression. Permutation importance assigned the largest relative share to social participation (31.2%). SHAP estimates showed larger model-specific contributions for selected social factors than for static physical attributes. The panel threshold model estimated a context-specific threshold of 54.80 on the normalized social participation scale (0–100). Below the threshold, the infrastructure coefficient was not statistically significant ( β = 0.062, p = 0.168); above it, the estimated coefficient was 0.485 ( p < 0.001). These observational results are consistent with a thresholded association between participation and infrastructure-resilience relationships; they do not establish a causal multiplier mechanism or a universal readiness cutoff.

Scientific Reports
Huzhou College (CN), Dalian Art College (CN), Daegu University (KR)
Daegu University
Sustainable cities and communities
Disaster Management and Resilience
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon