Research Analysis
Toward Urban Sustainability: Assessing SDG11.2 via Functional Zone Analysis in Five Chinese Cities
Authors: Lina Yuan, Xiaowen Zhang, Zijiang Song, Yuying Li, Pengyang Zang & Min Liu
Journal: npj Urban Sustainability
Published: February 18, 2026
DOI: 10.1038/s42949-026-00367-4
This study introduces an integrated framework, E-UFZ, combining VHR satellite imagery, geospatial data, and deep learning to evaluate SDG11.2 (public transport accessibility) at the Urban Functional Zone (UFZ) level. Achieving an overall accuracy of 87.25%, the model effectively captures fine-scale urban heterogeneity and reveals significant spatial disparities in accessibility across five Chinese cities, highlighting the need for targeted, zone-specific urban planning.
Executive Impact: Quantifying AI's Role in Urban Planning
Our AI-driven analysis of urban sustainability metrics provides unprecedented granularity, empowering data-informed strategic decisions for city development and resource allocation.
0
E-UFZ Model Overall Accuracy
0
Mean Intersection Over Union (MIOU)
0
Population with Public Transport Access
0
Accuracy for Key UFZ Categories
Deep Analysis & Enterprise Applications
Select a topic to dive deeper, then explore the specific findings from the research, rebuilt as interactive, enterprise-focused modules.
Enterprise Process Flow
E-UFZ Model for UFZ Extraction (CBAM-Deeplab, MRS, MRF)
→
SDG11.2 Indicator System Construction
→
UFZ-Scale SDG11.2 Assessment
→
Spatial Analysis & Relationship Investigation
87.25%
E-UFZ Model Overall Accuracy in UFZ Classification
| Feature | E-UFZ | U-Net | Xception |
|---|---|---|---|
| IOU (%) | 84.63 | 75.87 | 63.06 |
| OA (%) | 87.25 | 77.63 | 65.79 |
Urban Disparities in SDG11.2 Accessibility
Analysis across five Chinese cities (Shanghai, Suzhou, Hangzhou, Hefei, and Nanjing) reveals pronounced spatial disparities in SDG11.2 indicators. While residential, institutional, and commercial zones generally exhibit higher accessibility, slum zones and urban-fringe areas often fall into 'Very Low' categories, indicating constrained access to public transport. Inter-city variations highlight distinct urban development trajectories and functional specializations.
Key Learnings:
- SDG11.2 values are spatially heterogeneous within cities.
- Urban cores (residential, commercial, institutional) show higher accessibility.
- Slum zones and urban fringes often suffer from very low accessibility.
- Significant inter-city differences in functional zone composition and accessibility.
The integrated framework offers a robust, fine-grained approach to urban sustainability assessment, bridging macro-city and micro-grid analyses. Its transferability allows for application in diverse urban contexts globally, supporting standardized public transport accessibility measurement. For policy-makers, these insights underscore the necessity of UFZ-specific strategies that integrate transportation planning with land governance. Future work should focus on integrating granular demographic data, real-time monitoring, and addressing socio-spatial inequities to develop dynamic, equity-sensitive evaluation frameworks that extend to other SDG indicators like 11.3 (land-use efficiency) and 11.1 (poverty).
Calculate Your Potential AI Impact
Estimate the efficiency gains and cost savings your organization could achieve by implementing similar AI-driven analysis frameworks.
Annual Cost Savings
$0
Annual Hours Reclaimed
0
Your AI Implementation Roadmap
A phased approach to integrating advanced AI for urban sustainability, ensuring robust data foundations and impactful outcomes.
Phase 1: Data Foundation & Preprocessing (1-2 Months)
Gather VHR satellite imagery, road network data, population grids, and public transport station data. Clean and integrate multi-source geospatial information.
Phase 2: E-UFZ Model Development & Training (2-3 Months)
Customize and train the E-UFZ deep learning framework using CBAM-Deeplab, MRS, and MRF for accurate Urban Functional Zone (UFZ) extraction. Validate model performance against baseline methods.
Phase 3: SDG11.2 Assessment at UFZ Scale (1 Month)
Apply the trained E-UFZ model to map UFZs across target cities. Calculate SDG11.2 indicators (public transport accessibility) for each UFZ unit using integrated population and transport service data.
Phase 4: Spatial Analysis & Disparity Identification (1 Month)
Conduct global and local spatial autocorrelation analyses (Moran's I, LISA) to identify clustering patterns and hot/cold spots in SDG11.2 values. Analyze area proportions of SDG11.2 across different UFZ types to reveal disparities.
Phase 5: Strategic Planning & System Integration (2-3 Months)
Translate findings into actionable, zone-specific urban development and transportation strategies. Develop a roadmap for integrating granular demographic data and real-time monitoring to enhance equity-sensitive evaluations and extend to other SDG indicators.
Ready to Transform Your Urban Initiatives?
Our AI solutions are designed to provide the granular insights needed for sustainable urban development. Book a free consultation to see how we can tailor our framework to your city's unique challenges.
Ready to Get Started?
Book Your Free Consultation.
Let's Discuss Your AI Strategy!
Lets Discuss Your Needs
Select a Date
Sun
Mon
Tue
Wed
Thu
Fri
Sat