Decentralized approaches for self-adaptation in agent organizations

Abstract

Self-organising multi-agent systems provide a suitable paradigm for developing autonomic computing systems that manage themselves. Towards this goal, we demonstrate a robust, decentralised approach for structural adaptation in explicitly modelled problem solving agent organisations. Based on self-organisation principles, our method enables the autonomous agents to modify their structural relations to achieve a better allocation of tasks in a simulated task-solving environment. Specifically, the agents reason about when and how to adapt using only their history of interactions as guidance. We empirically show that, in a wide range of closed, open, static and dynamic scenarios, the performance of organisations using our method is close (70 − 90%) to that of an idealised centralised allocation method and is considerably better (10 − 60%) than the current state of the art decentralised approaches.

Advances in infrastructures and tools for multiagent systems

“…Adaptation of an organizational system is related to the reconfiguration of this system without explicit external control (Campos et al 2011;Kota et al 2012). Thus, techniques from the field of coordination models and languages are particularly promising in this context (Ossowski and Menezes 2006).…”

Utility-Based Mechanism for Structural Self-Organization in Service-Oriented MAS

“…Starting from entities that are structured in a sub-optimal organization or that are not organized at all, a self-organizing system is able to form a specific organization to pursue a well-defined goal [Kota et al 2012]. The main issue in self-organization is to determine which is the best mechanism to reorganize the current structure through the execution of local actions to achieve the desirable behavior with a high degree of uncertainty in the system.…”

“…There are other approaches focused in cooperative problem solving in organizations and how these organizations can be rearranged in order to improve its performance as the environmental conditions and the organizational goals change [Kota et al 2012;Kamboj and Decker 2007]. Many of these approaches rely on hierarchical structures where agents change their relations in order to distribute their workload to subordinates.…”

“…2007; Gaston and desJardins desJardins desJardins 2005; Kota et al 2012]. Therefore, agents should check their structural relations with a frequency based on the number of changes in the environment conditions.…”

“…The success of the service discovery process relies on the collaboration of other agents in the system [Del Val et al 2012b] and the selforganization of the structural relations between agents [Abdallah and Lesser. 2007; Gaston and desJardins desJardins desJardins 2005; Kota et al 2012].…”