An agent-based procedure with an embedded agent learning model for residential land growth simulation: The case study of Nanjing, China

The agent-based modelling (ABM) is commonly used to simulate urban land growth. A key challenge of ABM for the simulation of urban land-use dynamics in support of sustainable urban management is to understand and model how human individuals make and develop their location decisions that then shape urban land-use patterns. To investigate this issue, we focus on modelling the agent learning process in residential location decision making process, to represent individuals' personal and interpersonal experience learning during their decision making. We have constructed an extended reinforcement learning model to represent the human agents' learning when they make location decisions. Consequently, we propose and have developed a new agent-based procedure for residential land growth simulation that incorporates an agent learning model, an agent decision-making model, a land use conversion model, and the impacts of urban land zoning and the developers' desires. The proposed procedure was first tested by using hypothetical data. Then the model was used for a simulation of the urban residential land growth in the city of Nanjing, China. By validating the model against empirical data, the results showed that adding agent learning model contributed to the representation of the agent's adaptive location decision-making and the improvement of the model's simulation power to a certain extent. The agent based procedure with the agent learning model embedded is applicable to studying the formulation of urban development policies and testing the responses of individuals to these policies.