An urban growth simulation model based on integration of local weights and decision risk values

Monitoring and managing the rapid growth of cities, especially in developing countries, highlights the need for appropriate spatio-temporal models to predict urban growth. The parameters affecting the spatio-temporal analysis of urban growth play a key role in the prediction results. This study proposes an urban growth simulation model by integrating local subjective-objective weights and decision risk values (ORness) into a CA-Markov model. This involves the use of ordered weighting averaging as one of the multicriteria decision analysis methods to combine the weights and degree of risk for generating a variety of risk-averse and/or risk-taking urban growth prediction scenarios. The proposed model has been applied to predict the physical growth of Babol city, located in Mazandaran, Iran. The results indicate that the degrees of ORness = 0.3 and ORness = 0.9 yield better prediction results in the case of using the local and global weighting strategies, respectively. Furthermore, the overall accuracy of local and global weighting strategies at different degrees of risk was 87.6 and 86.8, respectively. This implies that the use of local subjective-objective weights leads to more accurate results than the global weights for simulating urban growth.