Modelling Horizontal and Vertical Urban Development Using Parcel-Based Cellular Automata and Artificial Neural Networks

Document Type : Original Article

Authors

Faculty of Geodesy and Geomatics Engineering, K. N. Toosi University of Technology, GIS Department, Tehran, Iran.

10.22059/eoge.2025.386437.1163

Abstract

The uncontrolled horizontal sprawl of urban development and the lack of management in high-rise construction, both inside and outside cities, highlight the need for an integrated modeling approach to urban development that addresses both horizontal and vertical dimensions. However, existing models have struggled to predict horizontal and vertical developments simultaneously, resulting in unreliable planning outcomes. This research addresses this gap by developing a novel approach, introduced by combining weighted linear combination (WLC) and artificial neural networks (ANN) models. The model effectively predicts horizontal and vertical development likelihoods simultaneously, providing a more integrated understanding of urban growth patterns. Results indicated that the WLC model achieved 60% accuracy for horizontal development and 30% for vertical development. In contrast, the ANN model achieved 67% accuracy for horizontal development and 65% for vertical development, with an overall suitability accuracy of 66.3% for simultaneous modeling. This demonstrates this research contributes to the field by providing a robust and integrated model that addresses both horizontal and vertical urban development, optimizing land use and accommodating urban growth sustainably.

Keywords

Main Subjects

Files

Share

How to cite

Statistics