Multi-Agent Based Scheduling System For Chemical Batch Production Process

Batch process can catch up the rapid changes of customers' demands due to its natural flexibility. Therefore, utilizing batch process becomes increasingly important in the chemical industry. In batch plant, different kinds of product share the resources, which make the production and economic benefit deeply depend on planning and scheduling. Hence, building a perfect batch process scheduling system is of great value both in theory and application. Recently, the most batch process scheduling systems have problem on calculate speed and dynamic performance. Multi-agent technology has characteristics of autonomy, distributing and coordination, and has advantage in calculate speed and robustness. This technology offers a new way to solve the scheduling for batch process.According to the structured and integrated scheduling system theory, a prototype multi-agent scheduling system for batch process is proposed. Agents are created for each task and resource. Tasks are assigned through negotiates base on Contract-Net Protocol, while the management agent settles the conflicts between tasks. To reduce the load of management agent and increase the robustness of system, a scheduling strategy based on distributing rules is established. Agents work concurrently, and have respective rules. The architecture of this system is hybrid, in which high-level agents can supervise low-level agents to improve the global perspective of the result. At the same time, the autonomous of agent is regarded to increase the robustness and flexibility of the system. The agents in this system have a reaction structure or hybrid structure according to practice needs.To solve the scheduling for a cosmetic plant, a model system is implemented with Java on IBM Aglets platform. The test of the system shows that it can offer a reasonable scheduling plan, decrease the burden of management people and improve the efficiency of production and resources.