Multi-Agent Simulation (MASim) forms an innovative paradigm for modeling and simulation. It basically means to use the concepts of Multi-Agent System for designing a simulation model. The basic metaphor of a model is therefore an "agent" -- an autonomous actor situated in an environment. Thus, the basic elements of a multi-agent model are the models of the (different) agents, a model of their interactions and a model of the simulated environment, the agents are living in.

Multi-Agent Simulation has many advantages and offers great opportunities mainly due to

• Intuitive modeling: an actor in the original system can be directly modeled as an agent
• High-level design: the gap between concepts to describe the original systems structure and behavior and the model structure and behavior are lower than in other micro-simulation paradims as multi-agent system (simulation) technology provides appropriate high-level abstractions.
• Freedom of design: There are no inherent constraint/restriction in model design. Structural dynamics, heterogeneity of model elements (actors, space...) , even adaptivity and evolutionary dynamics can be integrated in a natural way.

Therefore, multi-agent simulation is apt for a variety of applications from the simulation of human decision making in dynamic contexts or detailed biological systems to emergence of social networks/organizations, etc... Typical domains are landuse simulation, traffic and transportation, social science, economic system analysis, hybrid human-machine systems, gaming,...