Agency systems : standardization and architecture

Abstract EN

In this paper, we describe a principled methodology of agent-based systems architecture and standardization.

We illustrate the use of this methodology on an example of a supply chain management system.

Using the methodology, it is possible to assess architectural decisions on software qualities such as performance, modifiability, and security.

The focus for the methodology is risk identification, evaluation, and mitigation in the early stages of system design.

We describe a test bed to analyze and evaluate agent-based systems in manufacturing enterprises that is based in part on this methodology.

This paper provides an overview of the standardization effort in the area of specifying standards for heterogeneous, interpreting agent-based systems.

We describe also the types of software agents, which are of interest.