基于协商的智能排课系统的研究

Research of Intellective Class Scheduling System Based on Negotiation

多Agent系统作为近年来人工智能领域中的研究热点已在多个应用领域中取得了许多研究成果。多Agent系统便于建模和实现复杂、庞大的应用系统,是一种新的软件开发范例。本文提出了一种基于协商的智能排课系统框架就是以高校排课为应用背景来研究多Agent系统中的理论和技术实现问题。研究通过在多Agent之间的协商使得排课系统能够兼顾教师的意愿,提高教师满意度。本文首先介绍一下排课的基本问题;其次介绍研究提出的基于协商的排课系统的构架,说明多Agent系统中不同Agent的任务;然后给出系统的流程设计;讨论协商在系统中的应用;最后是系统的试验以及结论。

Multi-Agent Systems has become the focus of artificial intelligence in recent years. It generates lots of excitements because of the promise as a new paradigm for modeling and implementing complex, largescale systems. On the purpose of research theories and technologies in Multi-Agent Systems, this research proposes the frame of intellective class scheduling system. Based on the negotiation techniques of Multi-Agent Systems, the study aims to concern about majority teachers＇ expectations, intending to solve teacher＇s satisfaction problem of course timetable. First, we present some basic problems about class scheduling. Second, we present a multi-agent model of class scheduling system where heterogeneous agents have different tasks. Third, we define the process of the class scheduling system. Follow ing that, we discuss the application about negotiation for the multi agent scheduling system. In the next part, some experiments for class scheduling are introduced. Finally, we conclude on the perspectives of future researches and developments for the new generation of our systems.