Agent-based Model (ABM) is a simulation technique that consists of autonomous agents interacting with each other and the environment they are set in. It is a quickly growing method of modeling complex phenomena, especially those that arise from the individual and collective behavior of the agents. Agent-based models are often applied in computer science and used to model social phenomena, such as the emergence of complex organizations and markets, individual decision making and resource consumption, social networks, and epidemic spreading.
Agent-based models are constructed using a number of related processes. These include creating an environment in which agents may interact, defining parameters and variables for agents and the environment, and implementing rules that govern how agents interact. Agents in an ABM can come in a variety of shapes and sizes and can possess different levels of heterogeneous or homogeneous capabilities. Such agents interact with their environment and also with other agents, either directly or through the environment. Different rules can be applied to the agents to make them behave in this way.
Agent-based models are commonly used to better understand the behaviors of real world phenomena, perform what-if analysis on processes, aid in the design of complex systems, and assist in decision-making. In addition, ABMs are increasingly used to compare theories of social systems to the behavior of empirical systems.
The model scenarios created by agent-based models have made them increasingly popular for applied research. ABMs are used in a wide variety of applications, including business, economics, health sciences, artificial intelligence, ecology, geography, and climate modeling.
