This project explores the integration of large-scale urban data analysis with small-scale spatial design generation through an Agent-Based Model (ABM) simulation. The research aims to bridge the gap between macroscopic urban characteristics and microscopic spatial experience in landscape design.

The methodology consists of two main parts: First, analyzing 102,817 POI data points within Yizhuang New City to evaluate 2,437 urban streets and identify 6 distinct street types connecting to the site. Second, developing a slime mold simulation model using NetLogo platform to generate self-organizing network patterns that respond to these urban characteristics.

The simulation model reproduces complex group behaviors exhibited by slime mold growth, with particles marked according to their origins from different street types. This enables the identification of spatial distribution patterns that inform functional layout, road planning, and detail design strategies.

As the project leader, I was responsible for the entire research framework design, urban data collection and analysis (662,197 POI data via Gaode Map API), Agent-Based Model development and simulation in NetLogo, pattern interpretation strategies, and the translation from analysis to final design scheme. Team members assisted with analytical diagram production.