Identification of deprivation degrees using two models of fuzzy-clustering and fuzzy logic based on regional indices: A case study of Fars province

摘要

Increases in deprivation and inequality within urban areas will result in unwanted negative impacts, such as depopulation (immigration), suburbanization, and increases in crime. Hence, the mitigation of deprivation should be a primary consideration for policy makers when promoting sustainable development. A robust deprivation model is needed to analyze the effects of deprivation indices and related parameters. It is thus important to identify the significant deprivation parameters first. Subsequently, this paper attempts to derive proper deprivation indices and also proposes a new model to determine the degrees of deprivation of different cities in a province (called region). Eight deprivation indices (e.g. educational, cultural, health, welfare, housing, transportation, and service) are considered and to completely capture each index, four new parameters are designated. Next, a new hybrid model is proposed based on two techniques: fuzzy-clustering and fuzzy logic. Using the fuzzy-clustering method, cities are first classified into two groups of deprived and fully developed. To determine the degree of deprivation, we then develop a new system using fuzzy logic. The proposed fuzzy logic system feeds in the outputs from the fuzzy-clustering system and the deprivation of each city (for each index) is finally obtained. As a case study, 29 cities in the Fars province (Iran) were considered and the degree of deprivation for each city was identified. Results (deprivation degree) for each city and for each individual index were presented both quantitatively and qualitatively. The proposed model, unlike dassical methods, has a non-binary view to deprivation, assigns a degree to deprivation for mitigating its negative effects, can be used for proper future planning, and is generic so that it can be easily applied to other cases as well. (C) 2016 Elsevier Ltd. All rights reserved.