AN EMPIRICAL AGENT-BASED MODEL FOR RESIDENTIAL SEGREGATION, CASE STUDY: TEHRAN

Abstract. Residential segregation as a known consequence of rapid urbanization in developing countries is complicated socio-economic phenomenon. Quantified description, analysis, and prediction urban dynamics complex system have always been challenging issues. The recent intensive developments multi-agent simulations solution to these problems still lack enough real-world examples. Our purpose this research develop spatially explicit model residential space which accounts for specific city’s infrastructure. proposed agent-based simulation the Tehran over 20-year period between 1996 2016 based on GIS datasets provided by Municipality Statistical Center Iran.This first effort at presenting an Tehran’s segregation. We revised “Schelling” attempt customize it arrive acceptable fit Tehran. In addition Schelling’s parameters, comprise purely social factors, we identified several important spatial-environmental criteria, turn categorizing them AHP method. A certain number parameters like “neighborhood prestige” specifically belong are suggested time. expert-based was implemented Netlogo. Validation resulting pattern using Kappa indicator showed that simulated rate higher than 62%.