A meta-model for the analysis and design of organizations in multi-agent systems

TLDR

The paper proposes a generic meta-model of multi‑agent systems grounded in organizational concepts such as groups, roles, and structures, and introduces organizational reflection for system‑level tasks. The authors develop the AALAADIN meta‑model, which describes coordination and negotiation schemes and supports building diverse organizational forms, and implement it in the MADKIT platform built on a minimal agent kernel with platform‑level services as agents, groups, and roles. The study demonstrates that the AALAADIN meta‑model enables agent heterogeneity across languages, applications, and architectures.

Abstract

This paper presents a generic meta-model of multi-agent systems based on organizational concepts such as groups, roles and structures. This model, called AALAADIN, defines a very simple description of coordination and negotiation schemes through multi-agent systems. Aalaadin is a meta-model of artificial organization by which one can build multi-agent systems with different forms of organizations such as market-like and hierarchical organizations. We show that this meta-model allows for agent heterogeneity in languages, applications and architectures. We also introduce the concept of organizational reflection which uses the same conceptual model to describe system level tasks such as remote communication and migration of agents. Finally, we briefly describe a platform, called MADKIT, based on this model. It relies on a minimal agent kernel with platform-level services implemented as agents, groups and roles.