Agent-Based Simulation and Modeling of Retail Center Systems

Abstract

Central Place Theory has been proposed for a long time, but it is rarely applied in empirical research or planning practices on retail center systems in the real world primarily attributable to the complexities that are too difficult for the theory to capture and replicate in efficient ways. This paper proposes an agent-based simulation system which can derive spatial distributions of multilevel retail centers by simulating the interactions between retail centers and consumers. In the simulation, the retail centers can grow, recess, and adjust locations according to the revenues; the consumers generate different levels of shopping demands and choose retail centers according to spatial distance. By recursively simulating the interactions, equilibrium is reached with the distribution of emerging retail centers. The system is validated by examining the simulated retail center systems under five hypothetical but typical scenarios of consumer spatial distribution. It is also applied in modeling the retail center system of Shanghai in the 2000s. The results show a face validity of the simulation system as a potential tool for real-world planning practices.