A multi-agent-based negotiation approach for parallel machine scheduling with multi-objectives in an electro-etching process

Abstract

Electro-etching is a critical process in producing aluminium foil, and it always involves many parallel machines with three distinct voltages. Therefore, production scheduling involves these parallel machines and their eligibility. In this paper, an agent-based negotiation approach is proposed to develop a distributed parallel machine scheduling application in this manufacturing environment. The agent-based system consists of types of agent, i.e. job agents, machine agents and management agents, to represent jobs, machines and supervisors, respectively. To establish job allocations, job agents and machine agents have to bid interactively. Negotiation protocol and bid decision models are developed for negotiations between job agents and machine agents. The management agent plays a centralised controller and system coordinator role. The main function of the management agent is to manage the negotiation process between job agents and machine agents to ensure that the global objectives of the system, such as minimising total job tardiness and flow time, balancing machine loading and maximising total revenue, can be achieved. Experiments are conducted to evaluate the performance of the proposed system. The results show that the proposed agent-based scheduling mechanism will provide an integrated process plan and parallel eligible machine scheduling solutions with better global performance.

Keywords:

• scheduling
• software agent
• parallel machines
• preference of machines
• negotiation

Acknowledgements

This work was supported by funding from the National Science Council of the Republic of China under Grant NSC99-2221-E-167-021. Special thanks to all those who have contributed to this study.