On the Adoption of Multi-Agent Systems for the Development of Industrial Control Networks

TL;DR

This paper demonstrates how multi-agent systems can enhance industrial control networks by adding intelligence, flexibility, and cost-effectiveness to legacy systems through a layered, cooperative agent approach tested in simulation.

Contribution

It introduces a multi-layered agent-based industrial control network that integrates with legacy systems to improve supervision, control, and safety in industrial processes.

Findings

Agents effectively compensate for legacy system limitations

The approach is cost-effective and scalable

Simulation results show improved supervision and safety

Abstract

Multi-Agent Systems (MAS) are adopted and tested with many complex and critical industrial applications, which are required to be adaptive, scalable, context-aware, and include real-time constraints. Industrial Control Networks (ICN) are examples of these applications. An ICN is considered a system that contains a variety of interconnected industrial equipments, such as physical control processes, control systems, computers, and communication networks. It is built to supervise and control industrial processes. This paper presents a development case study on building a multi-layered agent-based ICN in which agents cooperate to provide an effective supervision and control of a set of control processes, basically controlled by a set of legacy control systems with limited computing capabilities. The proposed ICN is designed to add an intelligent layer on top of legacy control systems to compensate their limited capabilities using a cost-effective agent-based approach, and also to provide global synchronization and safety plans. It is tested and evaluated within a simulation environment. The main conclusion of this research is that agents and MAS can provide an effective, flexible, and cost-effective solution to handle the emerged limitations of legacy control systems if they are properly integrated with these systems.