Urban simulation using principal components analysis and cellular automata for land-use planning

Abstract

This paper discusses the integration of cellular automata (CA), principal components analysis, and GIS techniques in simulating alternative urban growth patterns for land-use planning. The simulation of actual cities usually involves multicriteria evaluation (MCE) in tackling the problems of complex spatial factors. Spatial factors often exhibit a high degree of correlation which is considered an undesirable property for MCE. It is difficult to determine the weights when many spatial variables are involved. This study uses principal components analysis (PCA) to remove data redundancy among a large set of spatial variables and determine the "ideal point" for land development. The simulation is based on transition rules that are related to the neighborhood function and similarity between cells and the "ideal point." Principal components analysis helps to deal with a large data set of spatial variables for the implementation of the CA model.