Modeling the physical development of Rasht city with methods based on soft computing

golamreza miri ¹ (Assistant Professor, Department of Geography and Urban Planning, Zahedan Branch, Islamic Azad University, Zahedan, Iran)
parviz rezaei ² (Associate Professor of Department of Geography, Rasht Branch, Islamic Azad University, Rasht, Iran)
reza zarei ³ (Assistant Professor, Department of Statistics, Faculty of Mathematics, Gilan University, Gilan, Iran)
tala abedi ⁴ (PhD student in Geography and Urban Planning, Astara Branch, Islamic Azad University, Astara, Iran)

Submited date : 2023-10-13 accepted date : 2024-01-13

Keywords: Modeling, Rasht City, urban physical development, Soft Computing,

Abstract :

The rapid growth of population and urbanization is an undeniable phenomenon. When cities grow in terms of size and population, coordination between the physical development of the city and population growth is very important . With the help of methods based on soft computing, including artificial neural networks and support vector machines, it is possible to predict the directions of urban development in the coming years. The purpose of this research is to model the development of Rasht city in the last twenty years and predict the directions of development of this city until 2032. By using ETM+ Landsat 7 and 8 satellite images of 2002, 2012 and 2021 of Rasht city and with GIS software, images with suitable band composition are prepared and then using two methods of artificial neural networks and support vector machine for floor images. are grouped The indicators considered for the neighborhood model of urban areas are distance from urban points, distance to central areas of the city, and distance to main streets and roads. Rasht city, the capital of Gilan province, is located at 49 degrees 35 minutes 45 seconds east longitude and 37 degrees 16 minutes 30 seconds north latitude from the Greenwich meridian, and its area is about 10,240 hectares. In this model, in the first stage of the model, an activation function was achieved by applying four input indicators to the images of 2002 and comparing them with the images of 2012, and in the second stage, the network test was performed with the input of images of 2012 and the output of 2021 of Rasht city. In the last stage, the prediction function has provided 2032 images of Rasht city. The artificial neural network model has a correct estimate of 95.9% in 2012 and 93.8% for 2021, so these numbers can be acceptable. The support vector machine model has been able to predict the development of Rasht city in 2012 by 96.4% and for 2021 by 95.3%, which has provided more accurate results than the artificial neural network model.