Development of an Agent-Based System for Collaborative Multi-Project Planning and Scheduling

Due to the widespread availability of the Internet, large scale distributed projects in manufacturing are becoming popular. In order to address the dynamic requirements of multi-project production, this paper develops a collaborative multi-project planning and scheduling system based on the multi-agent technology. The hybrid framework of system is presented and discussed with detail. Six types of intelligent agents are proposed in this system. These agents are dynamically deployed at each location where a project is executed. Further more, a negotiation mechanism for multi-project planning and scheduling is developed to coordinate these distributed agents. Task allocation strategy and critical resource leveling strategy are also presented . Through which a predictable and nearly optimal scheduling with agility and adaptability can be realized in distributed multiple projects manufacturing environment. Finally, a prototype system was implemented upon this framework, and was validated in an aerospace company.