期刊名称：Journal of Artificial Societies and Social Simulation
印刷版ISSN：1460-7425
出版年度：2012
卷号：15
期号：4
页码：1-20
DOI：10.18564/jasss.2100
出版社：University of Surrey, Department of Sociology
摘要：This paper describes a simulation model of the spatial development of economic activities over time. The key principle addressed is how spatial patterns of economic activity emerge from decisions of individual firms, which are in turn influenced by the existing spatial configuration. A stylized simulation is presented, in which two types of firms grow at different rates, giving rise to split offs and spatial relocations. The influence of the spatial pattern on individual firms' decisions is implemented in various ways, related to well-known effects such as Jacobs and Marshall externalities described in the economic literature and congestion effects. We demonstrate that different assumptions about the spatial scale of these externalities lead to different spatial configurations. Function concentration (Marshall effects) is more likely to lead to the emergence of subcentres with a specific specialisation. However, the spatial scale of the market and agglomeration effects matters. In particular, if Marshall advantages stretch out over a longer distance, more subcentres emerge. Somewhat surprisingly, congestion seems to have a minor impact on the emerging patterns. The simulation outcomes are intuitively plausible, suggesting that micro-simulation is a promising tool for developing forecasting models to support spatial and economic policies. However, they also articulate the need for validation of the behavioural decision rules, in particular by investigating how growth rates and the spatial scale of externalities differs between different industrial sectors.
