Urban growth processes joining cellular automata and multiagent systems. Part 2: computer simulations

The mathematical model of urban dynamics introduced in an earlier paper is applied to a case study in a small region in the southern part of Switzerland. The model mixes the point of view of cellular automata (cellular decomposition of the space, neighbourhood relations among cells, dynamics based on local evolution rules) with the approach to multiagent systems: the dynamics of the urban system are described in terms of decision processes of agents formalized using fuzzy-decision-theory methods. The region chosen for the case study evolves under the pressure of the nearby city of Lugano, owing to several factors: the accessibility of the area from Lugano, the availability of spatial resources for new development, the high level of quality of space, and the high level of congestion of Lugano. Local changes in the master plan from agricultural to residential use are taken into account, by introducing suitable memory variables that keep track of the demand of spatial resources for building in the recent past. Computer simulations showing the time evolution of the spatial distribution of the population and the real estate value are presented. The use of measures of attractiveness based on a fuzzy logic approach, and of Poisson-distributed events, enables us to introduce new effective approaches to parameter calibration and model validation.