| Abstract | Scheduling exams and constructing a timetabling is in general a complex and difficult task. This decision problem can be approached as an optimization problem and the constraints can be categorized into two groups defined as soft and hard constraints. Finding a non overlapping exams schedule is considered a hard constraint while looking for an evenly distributed schedule and a short duration of the overall exams period can be regarded as soft constraints. To handle soft constraints under the hard constraints verification we adopted a multiobjective optimization approach and used Tabu Search to find a good solution. The tabu Search incorporates a Fuzzy Inference Ruled Based System to choose the tabu tenure of the elements in the tabu list. In addition, in each iteration the inspection of solutions in the neighborhood of a certain point is necessary and the election of an improved solution can be considered a multiple attribute decision problem. In order to rank the solutions in each neighborhood an aggregation method is proposed based on the Compromise Ratio (CR) methodology. However, we introduced a modification by considering weight functions instead of fixed weights which allows for a more flexible modeling of preferences. The chosen function should guarantee the monotonicity of the operator and we present a theoretical result regarding suficient conditions for achieving such property. |
