Assessing urban land-use development: Developing an agent-based model

We have developed an agent-based model of urban land-use sprawl for a case study area in Qazvin province of Iran that brings the risk-explicit attitudes of land-use developers into consideration. For this purpose, new methods for searching landscapes, for selecting parcels to develop, and for allowing competitions among agents have been considered and implemented. The agents are of five categories; they act as mobile developers, and have several land-related objectives. Our proposed model evaluates two major cases of regarding and disregarding risk for categorizing the agents. The model uses the data of the year 2005 to be calibrated and the results are evaluated with data of the year 2010. The results revealed that while the risk-disregarding case was slightly better than risk-regarding in predicting the location of developments, the later was a better method to produce contiguous patches of developments. The model achieved the value of 82.15% measured by Kappa index in predicting the occurred developments. Therefore, it is considered here that using the attitudes of people towards risk along with appropriate weights for criteria maps help the model to simulate land-use developments with acceptable accuracy and more contiguous parcels.