Investigation the Capabilities of Geo-Simulation Based Approaches to Urban Development Modeling

Modeling of geospatial phenomena such as urban growth and development is one of the most effective approaches in the field of urban policy, decision making, and natural resource management. Because the changes have been happened due to the interaction between human being and the environment. It affects the ecosystem and will be a threatens for human vital resources as well. Several studies have been accomplished in urban development modeling. Spatial simulation models are a simulation tool for solving real problems in the geographic framework. Two main spatial simulation models are cellular automata and agent-based modeling, which also have been used in many research and decision making as efficient tools. The aim of this study is to examine the performance of these two methods in the field of urban forecasting and physical development. For this purpose, the simulation of the above mentioned models have performed for Zanjan city between 2005 and 2015. Kappa index and overall accuracy has been considered as measuring the accuracy of the proposed models, calculated 71.44% and 96.5% for cellular automata and 74.09% and 97.01% for agent-based model. The results indicated that the models used in this study have ability to predict the development of the city with satisfactory precision. Moreover, it is concluded that the accuracy of the modeling will increase and the predictions will be closer to reality with considering social relational effects of the people in the society by adding PSO algorithm in agent base model and logical regression in cellular automata model.