عنوان مقاله [English]
نویسندگان [English]چکیده [English]
Nowadays, urban expansion in peripheral lands is one of the most important issues for urban planners in various levels of decision making and decision taking. For recognizing this phenomenon, planners need to understand the current and future state of the urban expansion system. Therefore they require an accurate and appropriate tool for predicting the future. Moreover, plans and programs which are prepared for promoting the present state are failed because of the lack of proper tools for prognosticating the future state of urban expansion. Thus, this study aims to provide a CA-based model through combination of basic rules of CA model and AHP for predicting the future urban expansion in peripheral lands.
For the first time the main idea of CA model, has represented by Turing in 1930. After that, in 1940, the concept of CA has been evolved in computer science by Ulam and Newman. Then Conway developed a CA model which is known âgame of lifeâ, in University of Cambridge. The application of CA model in the field of urban planning begins with the Toblerâs efforts in Michigan University that is called cellular geography in 1970, and up to now a lot of activities have been done for improvement and making the CA models better for modeling the real world. CA model includes four components: state, lattice, neighborhood and transition rules. Conventional CA model has some limitation. For this reason, in this paper, basic rules of CA model are combined with the AHP and probabilistic approaches to enhance the efficiency of the model. The conventional CA models are unresponsive to the different weights of influential factors in urban expansion process. For this reason and because of the capabilities of AHP model, in this research CA model is combined with the AHP.
The urban expansion in peripheral lands depends on factors that influence the probability of transforming the vacant areas to the occupant areas. Therefore, first, the influential factors are extracted from reviewing the literature which is related to the research topic. After that the weaknesses of conventional CA model and the methods for ameliorate them have been determined. Then the AHP method and probabilistic approaches have been chosen for calibrating CA model and attain the transition rules. After the calibration methods the conceptual framework of the model is designed and converted to computational codes through MATLAB programming language. Then the transformation rate was calculated via regression model, and it is imported to the designated model with the maps that were prepared, overlaid and rasterized in GIS. Finally the Kappa coefficient is calculated for examination the coincidence of the results of the model and real expansion.
The results show that the combination of CA model with AHP and probabilistic aprroaches provide an appropriate tool for pridicting the future urban expansion in pripherial lands.