GIS-Based SWARA and Its Ensemble by RBF and ICA Data-Mining Techniques for Determining Suitability of Existing Schools and Site Selection of New School Buildings

Abstract Nowadays, educational centers are regarded as one of the most important buildings in cities; therefore, in the present study, the geographical suitability of the existing schools in two districts (6, 12) of Tehran Metropolitan, Iran, is studied using five standpoints including urban facilities, accessibility to urban road networks, population density, city services, and cultural centers. For this purpose, a stepwise weight assessment ratio analysis (SWARA) algorithm was applied to determine the importance of each subfactor. By use of “R” programming language, a combination of SWARA, radial basic function (RBF), and imperial competitive algorithm (ICA) was applied to eliminate SWARA gaps. Finally, five maps for factors, subfactors, and classes of each subfactor were developed. And, based on the location of existing schools and the final map, the site selection map for new schools in the two studied districts is presented. Also, the results showed that the percentage of schools in districts 6 and 12 in “good” areas was 24.30% and 40.80%, respectively. Therefore, the condition of schools located in district 12 is better than district 6. Additionally, the combination of the SWARA–RBF–ICA method rectified the SWARA method limitations and led to 5% more optimized results.