Combining “Top-Down” and “Bottom-Up” Decision Making Behaviors to Simulate Urban Sprawl Using Multi-Agent System

Honghui Zhang,Xiaobin Jin,Liping Wang,Yinkang Zhou

DOI: https://doi.org/10.1109/geoinformatics.2013.6626022

2013-01-01

Abstract:In view of multi-agent system's ability to simulate complex spatial system, from the view of combining “top-down” and “bottom-up” decision-making behaviors, a spatiotemporal dynamical urban sprawl simulation model using multi-agent system was developed. In the model, “top-down” macro land use planning behaviors implemented by macro agents and “bottom-up” micro land use spatial appeal behaviors autonomously originated by micro agents interacted by two-dimensional spatial grid, and jointly promoted urbanization process of study area through joint decision-making. The simulation results show that combining “top-down” and “bottom-up” decision-making behaviors to simulate spatiotemporal dynamical urban sprawl model using multi-agent system can give full play to the potential of multi-agent system to understand the driving mechanism of urbanization and provide decision-making support for urban management.