Agent-based modelling is a powerful approach for simulating the dynamics of geographical systems by breaking them down into individual components or agents, each with its own characteristics, properties, rules and behavior. Unlike traditional models that treat geographical components as homogeneous entities, agent-based modelling allows for the simulation of diverse agents, such as people, cities, or abstract representations, interacting with each other and their environment at various spatial and temporal scales. This bottom-up approach makes it possible to observe how individual decisions lead to complex system behaviors over time, providing deeper insights into urban problems like urban sprawl, congestion, and segregation, as well as to model natural and social phenomena such as animal behavior, pedestrian behavior, social insects and biological cells. Therefore, the macro-level behavior of the system arises from the interaction of individual agents and the environment over time.
Agent-based modelling development stems from automata-based models, which use rule-based mechanisms to process information and evolve over time. Two prominent automata-based approaches—cellular automata and agent-based modelling —have been widely adopted in geographic modelling. Agent-based modelling's advantage lies in its ability to model heterogeneous agents and dynamic interactions, which traditional models, focused on aggregate behaviors, cannot capture as effectively. While agent-based modelling offers unique insights into geographical systems, it also poses challenges, such as the complexity of simulating realistic agent behaviors. Nonetheless, agent-based modelling continues to grow in popularity for its ability to represent dynamic spatial changes in a more detailed and realistic manner.