An intelligent method for industrial location selection: Application to Markazi Province, Iran

Abstract

Decision-making and selection are important and sensitive aspects of planning. An important part of land-use planning is the location of human activities. Locating activities in the right places determines the future space of a region. Selection and definition of natural and human indices and criteria for location always face uncertainty. Thus, this study aimed to develop an intelligent method for industrial location. In this study a developmental-applied approach was used along with a descriptive-analytical method for data analysis. Through the review of related literature and a Delphi survey, 18 criteria were extracted and 6 main components were categorized. The data were analyzed and modeled by GIS, MATLAB software, and the Fuzzy Inference System (FIS) and Adaptive Neuro-Fuzzy Inference System (ANFIS) methods. For each modeling three industrial domains were extracted, i.e. weak, medium, and premium. A total of 42,968 hectares of premium industrial location with a score higher than 0.7 resulted from combining the produced maps. Other important findings were related to the architecture and methodology applied in the research based on computational intelligence and knowledge-based systems to analyze and understand the processes that influence the score of locations. The novelty of this method lies in the use of high computing power and information evaluation based on artificial intelligence (AI), making it possible to analyze and understand the processes influencing industrial location.