One of the peculiarities of university timetabling problems lies in their huge complexity and the easy transition between complex constrained system and an over-constrained one. The Faculty of Informatics timetabling problem represents very complex scheduling and resource allocation problem as individual timetable for every student has to be scheduled with respect to heterogeneous choices of the students for their courses. The problem is solved via constraint logic programming approach, with variables' annotations for computing variable ordering and with special objective function aimed to minimize students' conflicts, both helping to solve over-constrained part of the problem as well as to control exploration of the search space.