Search

Predicting patterns of urban growth will be a major challenge for policy makers and environmental scientists in the 21st century. How cities grow --their shape and size -- will have enormous implications for environmental sustainability and infrastructure needs. This paper presents a spatiotemporal ART-MMAP neural method to simulate and predict urban growth. Factors that affect urban growth -- that is, transportation routes, land use, and topography -- were directly used as inputs to the neural network model for model calibration. The calibrated network was then applied to a study site -- St Louis, Missouri -- to predict future urban growth and to examine future land development scenarios. This paper also introduces an effective and straightforward method for model validation and accuracy assessment, the prediction error matrix, which has been used in the pattern recognition field for several decades. In order to assess the performance of the neural network model, an in-depth accuracy assessment was conducted in which the model results were compared against a null model, an alternative naïve model, and two random models. The neural network model consistently outperformed the naïve model and two random models, and produced similar or better results than the null model. Furthermore, we evaluated the models' performance at different spatial resolutions. The prediction accuracy increases when spatial resolution becomes coarser. One particularly interesting result is that when the results are aggregated to 1 km spatial resolution, there is 100% accuracy of urban growth predicted by the neural network model versus actual urban growth. [ABSTRACT FROM AUTHOR]

This paper develops and applies neural network (NN) models to forecast regional employment patterns in Germany. Computer-aided optimization tools that imitate natural biological evolution to find the solution that best fits the given case (namely, genetic algorithms, GAs) are also used to detect the best NN structure. GA techniques are compared with more `traditional' techniques which require the supervision of experienced analysts. We test the performance of these techniques on a panel of 439 districts in West and East Germany. Since the West and East datasets have different time spans, the models are estimated separately for West and East Germany. The results show that the West and East NN models perform with different degrees of precision, mainly because of the different time spans of the two datasets. Automatic techniques for the choice of the NN architecture do not seem to outperform selection procedures based on the supervision of expert analysts. [ABSTRACT FROM AUTHOR]

The objective of this paper is to present a combined, two-step spatial analysis approach for the definition of metropolitan regions. The proposed approach, which constitutes an option to avoid the endless confrontations that may be derived from the essentially subjective political criteria, explores two branches of spatial analysis: spatial statistics and spatial modelling. Spatial statistics tools are used to identify the characteristics of local association and are combined with a neural network in order to build prediction models. The analyses conducted with exploratory spatial data analysis tools and census data give a clear indication of clusters of zones with similar characteristics, which can be seen as uniform regions. Spatial models can then be used to foresee the global behaviour of regions in terms of growth, albeit the basis of local (and historical) relationships among zones. The proposed approach is tested in a case study carried out in Portugal, where this is a timely issue. [ABSTRACT FROM AUTHOR]