Integrating Constrained Cellular Automata Models, GIS and Decision Support Tools for Urban Planning and Policy Making

In this paper we present a Decision Support System developed to assist urban planners and policy makers to simulate and analyse alternative urban layouts, land uses, and growth patterns. The core of the system is a modelling and simulation shell allowing the user to specify Constrained Cellular Automata models of urban and regional systems. Unlike conventional Cellular Automata, these are defined with a relatively large neighbourhood and a large number of states, representing both human and natural land-uses. They forecast changes in land-use for small parcels on the basis of both the activities present in the local neighbourhood and the specific characteristics of the parcels themselves. Since each parcel affects every other within its neighbourhood, complex dynamics emerge. More traditional dynamic models --ideally spatial interaction based models-control the overall dynamics of the cellular automata. In the DSS, a custom-built GIS is integrated in the modelling shell. It stores the detailed geographical characteristics and feeds them into the model at the cellular level as required. Thus the approach permits the straightforward integration of physical, environmental, technical and institutional factors into a single dynamic socio-economic model at a high level of spatial detail. By way of illustration, we report applications of the models to both intra-urban and inter-urban cases, and examine specifically the case where the cellular model is forced by means of a regionalised macro-model.