Automated Timetabling Using Stochastic Free-Context Grammar Based on Influence-Mapping

Abstract

This paper presents a new system that solves the problem of finding suitable class schedule using strongly-typed heuristic search technique. The system is called Automated Timetabling Solver (ATTSolver). The system uses Stochastic Context-Free Grammar rules to build schedule and make use of influence maps to assign the fittest slot (place & time) for each lecture in the timetable. This system is very useful in cases of the need to find valid, diverse, suitable and on-the-fly timetable which takes into account the soft constraints that has been imposed by the user of the system. The performance of the proposed system is compared with the aSc system for the number of tested schedules and the execution time. The results show that the number of tested schedules in the proposed system is always less than that in aSc system. Moreover, the execution time of the proposed system is much better than aSc system in all cases of the sequential runs.