SUMMARYAlthough information technology is expanding in the university environment, there are still problems, such as the difficulty of performing registration accurately due to the extremely complex rules in a registration system. In order to make such determinations with a high degree of precision, databases with a variety of information and programs that precisely describe the rules are needed, resulting in an expensive system. As a solution to this problem the authors introduce a framework which allows for submitting registration requests even when there is an error in the registration. As a result, a screening mechanism that can efficiently narrow down and clarify the various problems present in the system can be created while at the same time academic affairs staff can check the content of the submitted registrations. As is the case with paper-based registration, staff can take a look at problematic registrations submitted by students, and then can give precise guidance to each person based on the student's individual information. In addition, the authors provide specific operational data such as the access count for when registration was allowed from outside the school, the number of submissions during the registration period, and the load averages for the server when access peaked while running the system for about 3200 students during 2002 and 2003.