Agents in a multiagent system are not typically entirely self-sufficient; instead, they frequently need to enlist other agents to perform tasks for them or to exchange goods or services with them. This creates a problem: how can an agent efficiently locate other agents to work or trade with? As the number of agents grows, the cost of this computation can become prohibitively large. One solution to this is for the system to self-organize into smaller groups of agents. In this paper, we apply the idea of congregating to a model of an information economy. We illustrate how participants in this economy can self-organize into a set of markets such that agents are able to find suitable partnets while retaining low computational costs. We show how congregating can help allocation problems scale to large populations by allowing agents to interact locally.

Abstract not found

The Emerald Research Register for this journal is available at The current issue and full text archive of this journal is available at

We operationalize the theory associated with an economic model of network formation and conduct experiments to determine the feasibility of endogenous, dynamic network formation in multi-agent organizations. We develop a learningbased, decentralized network formation strategy that allows the agents to make network adaptation decisions based on past performance. We compare our method with a dynamic network formation process proposed in the economics literature that relies on a global computation over the entire network structure. Our findings demonstrate that local decisions based solely on prior experience perform as well as local decisions based on perfect knowledge and a global computation.

We analyze the problem of modeling and developing multiagent systems from the organizational theory point of view. In particular, we focus on the critical issue of adapting multiagent systems organizations whenever changes in the structure of the multiagent system are required. We survey different relevant agent-oriented methodologies and discuss their suitability in dealing with adaptation in multiagent organizations. Finally, we present some critical considerations about the analyzed methodologies together with some open issues related to the process of modeling organizations for facilitating their adaptations.

Multi-agent systems (MAS) offer promising solutions to complex problems in distributed, open environments. Each MAS agent uses limited resources, incomplete knowledge, and limited capabilities to solve complex problems via agent interactions. For this reason, how to perform efficient searches for compatible agents is currently an important research topic. The author describes a proposed decentralized service lookup (DSL) mechanism that constructs service directories that are distributed among agents, who use keywords to find resources, capabilities, and knowledge held by other agents. The proposed DSL avoids three common problems in this area: broadcast queries, single point failures, and performance bottlenecks. ii Acknowledgement I am deeply indepted to my advisor, Dr. Chuen-Tasi Sun. All my life, I will remember my first research results from his considerate instruction. I am also very grateful to my dear firends, Hsun-Cheng Lin, Wen-Heng Chung,

Multi-agent systems (MAS) offer promising solutions to complex problems in distributed, open environments. Each MAS agent uses limited resources, incomplete knowledge, and limited capabilities to contribute to the solution of complex problems via agent interactions. For this reason, how to perform efficient searches for compatible agents is currently an important research topic. The author describes a proposed decentralized service lookup (DSL) mechanism that constructs service directories that are distributed among agents, who use keywords to find resources, capabilities, and knowledge held by other agents. The proposed DSL avoids three common problems in this area: broadcast queries, single point failures, and performance bottlenecks. 1.