The Role of Organization in Agent Systems

The Role of Organization in Agent Systems

Virgina Dignum (Utrecht University, The Netherlands)
DOI: 10.4018/978-1-60566-256-5.ch001
OnDemand PDF Download:
$37.50

Abstract

Agent Organization can be understood from two perspectives: organization as a process and organization as an entity. That is, organization is considered both as the process of organizing a set of individuals, or as an entity in itself, with its own requirements and objectives. In fact, agent organizations demand the integration of both perspectives and rely for a great extent on the notion of openness and heterogeneity of MAS. Practical applications of agents to organizational modeling are being widely developed, however, formal theories are needed to describe interaction and organizational structure. Furthermore, it is necessary to get a closer look at the applicability of insights and theories from organization sciences to the development of agent organizations. In this chapter, current approaches to agent organizations are discussed. Agent Organizations bring concepts and solutions from sociology and organizational theory into agent research, integrating organizational and individual perspectives, and aim at the dynamic adaptation of models to organizational and environmental changes, both of which are impacted by the notion of openness and heterogeneity. The chapter also presents a promising application of agent organizations to the area of human-agent teams.
Chapter Preview
Top

Abstract

Agent Organization can be understood from two perspectives: organization as a process and organization as an entity. That is, organization is considered both as the process of organizing a set of individuals, or as an entity in itself, with its own requirements and objectives. In fact, agent organizations demand the integration of both perspectives and rely for a great extent on the notion of openness and heterogeneity of MAS. Practical applications of agents to organizational modeling are being widely developed, however, formal theories are needed to describe interaction and organizational structure. Furthermore, it is necessary to get a closer look at the applicability of insights and theories from organization sciences to the development of agent organizations. In this chapter, current approaches to agent organizations are discussed. Agent Organizations bring concepts and solutions from sociology and organizational theory into agent research, integrating organizational and individual perspectives, and aim at the dynamic adaptation of models to organizational and environmental changes, both of which are impacted by the notion of openness and heterogeneity. The chapter also presents a promising application of agent organizations to the area of human-agent teams.

Top

Introduction

The main topic of this book is Agent Organization. Since their coming out in the 80’s, Multi-Agent Systems (MAS) have often been defined as organizations or societies of agents, i.e. as a set of agents that interact together to coordinate their behavior and often cooperate to achieve some collective goal (Ferber et al., 2003). The term agent organization has become common-place within the MAS community, but is used to mean different, often incompatible, issues. In short, one class of views takes organization as the process of organizing a set of individuals, whereas the other sees organization as an entity in itself, with its own requirements and objectives. These differences are for a great part due to the diverse world views and backgrounds of different research fields, namely Sociology and Organization Theory (OT) on the one hand, and distributed Artificial Intelligence (AI) on the other hand. From a sociologic perspective, agent organization is specified independently from its participants and relates the structure of a system to its externally observable global behavior. The artificial intelligence view on MAS is mostly directed to the study of the mental state of individual agents and their relation to the overall behavior. As it is often the case in such situations, such differences can lead to a fragile sense of understanding causing interpretation and integration problems when trying to compare, merge or analyze different models and frameworks. Without attempting to merge different views into one general whole, this book presents a comprehensive overview of the different perspectives, such that the reader will be able to better understand and judge the differences.

The stance in this chapter, and in the rest of this book is that agent organizations demand the integration of organizational and individual perspectives, the dynamic adaptation of models to organizational and environmental changes, and rely for a great extent on the notion of openness and heterogeneity of MAS. Practical applications of agents to organizational modeling are being widely developed but formal theories are needed to describe interaction and organizational structure. Furthermore, it is necessary to get a closer look at the applicability of insights and theories from organization sciences to the development of agent organizations. All these issues are becoming increasingly recognized and researched within the MAS community, as can be seen by the growing number of publications and workshops on the subject, of which COIN1 (International Workshop Series on Coordination, Organization, Institutions and Norms in MAS) is the most known example.

Key Terms in this Chapter

Organization Design: (viewed as entity) The complete specification of strategy, structure, processes, actors (people or agents), coordination and control components of the organization.

Performance: Comprises the actual output or results of an organization as measured against its intended outputs (that is, its goals and objectives). Performance depends on how well the organization fits with its environment.

Environment: The forces outside the organization that can impact it. Include marketplace, regulatory and legal situation, opportunities and threats of the context in which the organization operates.

Human-Agent Team: A network where both intelligent systems (agents) and people cooperate effectively within one group, creating synergy by reinforcing each other’s strong points and anticipate on each other’s weak points.

Agent Organization: Comparable to human organizations, agent organizations are characterized by global goals and formalized social structures representing the stakeholder desired results. Can be seen as the structural setting in which agent interactions occur.

Organizational Structure: Represents the relations between entities of an agent organization that are taken to be invariant through time. The main constructs found in it are roles, groups, and relationships between them.

Organization Design: (viewed as process) Is the process of determining and implementing goals, roles and rules that define interaction. It involves both the partition of global organization goals into specific tasks to be executed by smaller units, and the coordination of such smaller units so that they fit together and enable efficient realization of the larger goal.

Complete Chapter List

Search this Book:
Reset