エージェントベースモデルによる自己組織化的集積の再現：特別区におけるICT 事業所

　As a preliminary step to conducting a self-organizing simulation of the agglomeration and dispersion of the information and communications industry, we quantify the spatial agglomeration of the information and communications industry in Japan’s major cities. Using the town and district data from the Economic Census for Economic Activity, we attempted a spatial analysis of the information and communications industry in the Tokyo Special Wards as well as in Sapporo, Sendai, Hiroshima, and Fukuoka, which have regional central functions. As a result of detecting spatial autocorrelation in small areas within each city, hotspots indicating the agglomeration of information and communications industry offi ces were found in the city center of each city. At the same time, we were able to confi rm the impact of agglomeration economies, which are the premise of the self-organizing model, and confi rmed that the information and communications industry is an industry suitable for simulation of the self-organizing model. 　Krugman（1996）was the fi rst to model the self-organizing model in cities and clarifi ed the emergence principle of peripheral cities, but this was limited to numerical simulations. Later, Kumar et al.（2007）used actual data to show that Krugman’s self-organizing model could be applied to predict corporate agglomeration and dispersion. In this paper, we use an agent-based model to examine whether a self-organizing model is also effective in reproducing and predicting the concentration and dispersion of the information and communications industry in Japanese cities.

論説 (Articles)