Study on an Artificial Society of Urban Safety Livability Change

Lihu Pan, Le Zhang, Shipeng Qin, Huimin Yan, Rui Peng, Fen LiISPRS Int. J. Geo Inf 2021 · OpenReview

TL;DR

This paper introduces an "Urban Livability Change Artificial Society" (ULC-AS) model to simulate and predict changes in urban safety livability. By modeling interactions between government, families, residents, and safety agencies, the ULC-AS reveals that population growth and government investment are key drivers of urban livability, offering a new approach for urban planning and resource allocation.

What Problem Does This Paper Address?

Urban livability is a critical goal for modern urban development, encompassing various factors that affect residents' quality of life. While existing research has explored urban livability through concept definition, indicator assessment, and spatial analysis, many models simplify complex human behaviors and interactions with the environment. This simplification limits their ability to accurately reflect the dynamic interplay of factors influencing urban livability. Specifically, the paper addresses the challenge of understanding how individual and collective decisions, influenced by environmental changes and policy interventions, shape urban safety livability over time. The authors aim to bridge this gap by developing a more comprehensive simulation model that captures these intricate dynamics.

Methods

The core of this research is the development and application of an "Urban Livability Change Artificial Society" (ULC-AS) model. This model employs an agent-based simulation approach, treating the city as a computer-simulated world. The ULC-AS operates on three hierarchical levels: facility, geography, and agent.

  • Environmental Model: This layer represents the physical infrastructure, including safety facilities (public security, fire, traffic, emergency, residential), transportation, and other essential services.
  • Geospatial Model: This layer establishes a coordinate system (cell, grid, street) to define the spatial relationships and positions of agents within the urban environment.
  • Agent Model: This is the most dynamic component, featuring four types of agents:
  • Government Agents: Responsible for policy-making and investment decisions.
  • Family Agents: Represent households making decisions about relocation and child-bearing.
  • Resident Agents: Individual actors with preferences, employment seeking, and consumption behaviors.
  • Safety Facility Management Agency Agents: Responsible for the construction and management of safety infrastructure.

The model simulates interactions between these agents and their environment, allowing for the analysis of emergent behaviors and changes in urban safety livability and resident satisfaction over time. The simulation is conducted annually, incorporating factors like birth/death rates, migration, economic changes, and policy implementations.

Datasets

The study utilizes real-world data for the Futian District in Shenzhen, China. Specific data sources mentioned include:

  • Infrastructure distribution data from the urban construction management department of Futian District.
  • Shenzhen Statistical Yearbook for birth rate, mortality rate, and foreign population growth rate.
  • China State Council data for retirement age.
  • Survey data for various parameters like safety facility weight, education level, and family member numbers.

The paper does not introduce new benchmark datasets but rather leverages existing geographical and demographic information to parameterize and validate its simulation model.

Key Insights and Ideas

  • Agent-Based Modeling for Urban Dynamics: The ULC-AS demonstrates the power of agent-based modeling to capture the complex, emergent behaviors that drive urban livability changes, moving beyond static analyses.
  • Interplay of Population and Infrastructure: The model highlights that while population growth (especially rapid migration) can strain resources and decrease livability, strategic government investment in safety facilities can effectively mitigate these negative impacts and improve overall satisfaction.
  • Policy Sensitivity: The ULC-AS can effectively simulate the impact of different policy interventions (e.g., fertility policies, investment strategies) on urban livability, providing a valuable tool for urban planners.
  • Spatial Heterogeneity Matters: By incorporating a hierarchical geospatial structure (cell, grid, street), the model accounts for spatial variations in livability, offering a more nuanced understanding than purely regional analyses.

Key Results

The simulation, conducted over 20 years for the Futian District, yielded several key findings:

  • Scenario 1 (Current Situation): Urban safety livability and satisfaction showed a stable trend, with government investment playing a crucial role in maintaining these levels despite demographic changes.
  • Scenario 2 (Increased Migration): Higher rates of migrant population growth led to a downward trend in urban safety livability and satisfaction, as existing infrastructure struggled to keep pace with increased demand.
  • Scenario 3 (Fertility Policy Liberalization): A rapid increase in population due to a liberalized birth policy caused a significant, albeit temporary, decline in livability, emphasizing the need for proactive infrastructure development.
  • Scenario 4 (Increased Government Investment): Increasing the coverage of safety facilities demonstrably improved urban safety livability and resident satisfaction, with the degree of improvement directly correlating with the extent of investment. The study found that government investment could effectively narrow the livability gap between different regions.

Limitations and Future Work

The authors acknowledge several limitations and suggest avenues for future research:

  • Data Availability: The model's accuracy is dependent on the availability and quality of real-world data. For instance, the model currently does not account for transportation, catering, or medical treatment, which are significant factors in overall livability.
  • Computational Constraints: Simulating a large number of agents (over 1.5 million residents) poses significant computational challenges, requiring optimization of simulation tools and processing methods.
  • Model Complexity: The model currently focuses primarily on safety livability. Future work should incorporate other crucial aspects of livability such as comfort, convenience, accessibility, and inclusiveness.
  • Dynamic Environments: The model's reliance on historical data for simulation might not fully capture future unpredictable changes in socioeconomic growth and human activities. Further optimization and validation with more dynamic data are needed.

Summary updated

Abstract

Improving the urban livability status has become the core goal of urban development, and reasonable assessment of the urban livability status and impact is crucial. By combining an objective environment with residents’ subjective cognition, an artificial society (urban livability change artificial society; ULC-AS) is constructed. The ULC-AS includes four types of agents, namely, government, family, resident and safety facility management agency agents, and recognizes dynamic interaction among various agents and between agents and the environment. Taking the Futian District of Shenzhen as an example, this paper examines factors such as migrants, birth policies, and government investment. We simulate the interactions among resident satisfaction changes, relocation decision-making behavior and urban safety livability and analyze the change processes and development trends of urban safety livability under multiple scenarios. Our main result indicates that population change and investment construction are important factors affecting urban safety livability. At present, the population of the Futian District is saturated. Therefore, the government must assess the urban safety livability and increase investment in high-demand areas. Through this method, the goals of urban resource allocation optimization and coordinated urban development can be achieved.
BibTeX
@article{pan2021study,
  author = {Lihu Pan and Le Zhang and Shipeng Qin and Huimin Yan and Rui Peng and Fen Li},
  title = {{Study on an Artificial Society of Urban Safety Livability Change}},
  journal = {ISPRS Int. J. Geo Inf.},
  year = {2021},
  url = {https://openreview.net/forum?id=rZLq6krmonW},
}
Figure