MeetingSchedulingProblem (MSP) arranges meetings between a number of participants. Reaching consensus in arranging a meeting is very diffuclt and time-consuming when the number of participants is large. One efficient approach for overcoming this problem is the use of multi-agent systems. In a multi-agent system, agents are deciding on behalf of their users. They must be able to elicite their users' preferences in an effective way. This paper focuses on the elicitation of users' preferences. Analytical hierarchy process (AHP) - which is known for its ability to determine preferences - is used in this research. Specifically, an adaptive preference modeling technique based on AHP is developed and implemented in a system and the initial validation results are encouraging. Copyright © 2011, IGI Global.