A multi-agent system for course timetabling

This paper proposes a multi-agent system for solving the university course timetabling problem. The solution of the university course timetabling problem requires the development of an intelligent decision-making system. This work attempts to show how agent-technology can be harnessed in the development of such an intelligent decision-making system. Course timetabling is a dynamically distributed problem and as such requires a decision-making system which can partition itself to the characteristics of the problem instance as required. In this agent-based solution, agent autonomy and a flexible communication methodology are used to create the back-bone of the intelligent decision-making system. Course Agents, representing each course in the problem, communicate and negotiate with other Course Agents through a Signboard Agent to find a mutually acceptable timetable. The Signboard Agent, is the mechanism that is used to identify course agents which need to negotiate with each other in order to resolve conflicts. It is also the mechanism through which the evolving timetable is made available to the user. A key strength of the agent-based approach is the use of the fundamental attribute of agent autonomy to represent all aspects of the fundamental unit in the problem - the course. By mapping the problem domain exactly into a fundamental attribute of the agent paradigm, we believe powerful and effective decision-making system is developed. Experimental results show that this intelligent decision system for course timetabling leads to an effective and flexible solution. Through negotiation and cooperation of the mobile and stationary agents in the system, the timetabling problem can be solved in a dynamic and distributed way.

[1]  Hyacinth S. Nwana,et al.  Software agents: an overview , 1996, The Knowledge Engineering Review.

[2]  Hoay Beng Gooi,et al.  Co-evolutionary algorithm approach to a university timetable system , 2002, Proceedings of the 2002 Congress on Evolutionary Computation. CEC'02 (Cat. No.02TH8600).

[3]  Alon Itai,et al.  On the complexity of time table and multi-commodity flow problems , 1975, 16th Annual Symposium on Foundations of Computer Science (sfcs 1975).

[4]  L. Benedicenti,et al.  Mobile agents for economic market simulation , 2002, IEEE CCECE2002. Canadian Conference on Electrical and Computer Engineering. Conference Proceedings (Cat. No.02CH37373).

[5]  Raman Paranjape,et al.  Mobile multi-agent system for medical image retrieval , 2001, Canadian Conference on Electrical and Computer Engineering 2001. Conference Proceedings (Cat. No.01TH8555).

[6]  Nicholas R. Jennings,et al.  Intelligent agents: theory and practice , 1995, The Knowledge Engineering Review.

[7]  Christian Blum,et al.  Metaheuristics in combinatorial optimization: Overview and conceptual comparison , 2003, CSUR.

[8]  Ramón Alvarez-Valdés,et al.  Design and implementation of a course scheduling system using Tabu Search , 2002, Eur. J. Oper. Res..

[9]  Gerhard Weiss,et al.  Multiagent Systems , 1999 .

[10]  Yen-Zen Wang,et al.  Using genetic algorithm methods to solve course scheduling problems , 2003, Expert Syst. Appl..

[11]  Rebecca Y. M. Wong,et al.  Optimizing agent-based meeting scheduling through preference estimation , 2003 .

[12]  Geoffrey C. Fox,et al.  A Comparison of Annealing Techniques for Academic Course Scheduling , 1997, PATAT.

[13]  Rong Xie,et al.  Scheduling multi-task multi-agent systems , 2001, AGENTS '01.

[14]  D. de Werra,et al.  An introduction to timetabling , 1985 .

[15]  Katia P. Sycara,et al.  Distributed Intelligent Agents , 1996, IEEE Expert.

[16]  Min-Jung Yoo.  An industrial application of agents for dynamic planning and scheduling , 2002, AAMAS '02.

[17]  Luigi Benedicenti,et al.  Mobile agent network management system performance study in frame relay network , 2002, IEEE CCECE2002. Canadian Conference on Electrical and Computer Engineering. Conference Proceedings (Cat. No.02CH37373).