Zoey Guo

Zoey Guo

she/her

Master of City Planning student at UPenn, specializing in Smart Mobility.

Publications

Journal Articles

Interpretable spatial machine learning insights into urban sanitation challenges: A case study of human feces distribution in San Francisco

Published in Sustainable Cities and Society, 2024

Urban sanitation is critical for public health, with the management of human feces presenting significant challenges in growing urban areas. While prior research has concentrated on the health impacts of fecal contaminants, the spatial distribution and determinants of open defecation in urban contexts have received less attention. To address these gaps, this study proposed an interpretable spatial machine learning framework integrating Geographically Weighted Random Forest (GW-RF) and SHapley Additive exPlanations (SHAP) analysis to reveal the complex spatial heterogeneity and factors influencing feces density in cities, taking San Francisco as a case study. Our findings highlight that homelessness, population density, and building density are critical drivers of feces distribution. Importantly, higher restroom density was linked to increased feces density, underscoring the need for urban planning to focus on improving restroom accessibility rather than merely increasing their number. Additionally, our research suggests that green spaces serve as a mitigating factor, indicating that enhancing urban greenery could be an effective strategy for addressing sanitation challenges. This study not only offers insights into San Francisco’s urban sanitation management but also provides practical implications for urban development strategies globally, advocating for targeted, evidence-based interventions to foster healthier and more sustainable cities.

Recommended citation: Yi, S., Li, X., Wang, R., Guo, Z., Dong, X., Liu, Y., & Xu, Q. (2024). Interpretable spatial machine learning insights into urban sanitation challenges: A case study of human feces distribution in San Francisco. Sustainable Cities and Society, 113, 105695.

AI Agent as Urban Planner: Steering Stakeholder Dynamics in Urban Planning via Consensus-based Multi-Agent Reinforcement Learning (https://arxiv.org/abs/2310.16772)

Published in arXiv, 2023

In urban planning, land use readjustment plays a pivotal role in aligning land use configurations with the current demands for sustainable urban development. However, present-day urban planning practices face two main issues. Firstly, land use decisions are predominantly dependent on human experts. Besides, while resident engagement in urban planning can promote urban sustainability and livability, it is challenging to reconcile the diverse interests of stakeholders. To address these challenges, we introduce a Consensus-based Multi-Agent Reinforcement Learning framework for real-world land use readjustment. This framework serves participatory urban planning, allowing diverse intelligent agents as stakeholder representatives to vote for preferred land use types. Within this framework, we propose a novel consensus mechanism in reward design to optimize land utilization through collective decision making. To abstract the structure of the complex urban system, the geographic information of cities is transformed into a spatial graph structure and then processed by graph neural networks. Comprehensive experiments on both traditional top-down planning and participatory planning methods from real-world communities indicate that our computational framework enhances global benefits and accommodates diverse interests, leading to improved satisfaction across different demographic groups. By integrating Multi-Agent Reinforcement Learning, our framework ensures that participatory urban planning decisions are more dynamic and adaptive to evolving community needs and provides a robust platform for automating complex real-world urban planning processes.

Recommended citation: Qian, Kejiang, Lingjun Mao, Xin Liang, Yimin Ding, Jin Gao, Xinran Wei, Ziyi Guo, and Jiajie Li. "AI Agent as Urban Planner: Steering Stakeholder Dynamics in Urban Planning via Consensus-based Multi-Agent Reinforcement Learning." arXiv preprint arXiv:2310.16772 (2023).