Agent-based simulation has become an important and matured research method in many domains over the past two decades. Multi-agent systems - as the technological basis - in combination with appropriate modelling languages allow simulation of complex systems, i.e. systems composed of many entities in interaction with themselves and with an external environment.

Such systems are used in many domains in which classical mathematical models are not available for various reasons, among others:

• the dynamics are far from equilibrium, or

• the systems are open (creation and deletion of entities in the course of simulation), or

• emergent phenomena have to be explored, or

• the entities are heterogeneous (for example, the eco-socio-systems which combine ecological and social dynamics), or

• simply because they provide a more intuitive understanding.

• The track is intended to offer a forum for people interested in agent-based simulation to present and discuss methodologies, techniques and applications.

We encourage contributions which cover not only case studies but also critical analysis on the pertinence of multi-agent models and their usefulness for understanding or for decision making, design and validation methodologies, comparative studies in the modelling sciences.

In particular, the subjects include but are not limited to:

• Design, implementation and validation methodologies, including participative protocols

• Modelling languages and platforms

• Agent architectures for different fields of simulation

• Interaction protocols including with humans

• Multi-level simulation and emergence of global phenomena

• Time and space representation including multiple scales

• Comparison with centralised models, relationship between individual-based versus aggregative models

• Conceptual and epistemological analysis