| System dynamics | Systems science and computer simulation | Modeling process is simple and can be combined with an index system to identify system boundary and related variables | Difficult to reflect the characteristics of adaptive and spatial change in the system, and the feedbacks are in part regression relationships | Urban system change, urban sustainable development and urbanization and eco-environment element coupling |
| Artificial neural network | Artificial intelligence | A typical human brain model with three advantages: self-learning, associative storage and high-speed optimization | Defective in learning, causal explanation and other aspects, especially in dealing with system uncertainty | Urban land expansion, environmental change, and resources demand |
| Bayesian networks | Artificial intelligence, probability theory, statistics and graph theory | Good at causal and diagnostic reasoning, as empirical data can be incomplete | Difficult to deal with the large number of nodes and the learning ability is less than for ANN | Identification of urban ecological vulnerability and demand for resources |
| CLUE-s | LUCC, systems science and computer simulation | Good at dealing with different spatial scales based on empirical data | Focus on local equilibrium analysis | Land use allocation on multiple spatial scales |
| Cellular automata | LUCC, systems science and computer simulation | Simplifies complex problems by bottom-up modeling and can simulate complex discrete systems | Difficult to solve the problem of spatial heterogeneity and lacks explanation of the mechanism | Urban sprawl and land use change |
| Multi-agent system | Artificial intelligence and complexity science | Compensates for the neglect of policy factors and explains land use change processes | Research space is abstracted as homogeneous and model validation is difficult | Policy-driven urban sprawl and land use change |
