Solving multicriteria university timetabling problem by a self-adaptive genetic algorithm with minimal perturbation

The most of methods for solving the university timetabling problem deal with the problem statically, which makes great difficulties when requirements are changed, because a new timetable is usually very different from the previous one. To overcome it the minimal perturbation problem is incorporated into the timetabling model. For solving the model a self-adaptive genetic algorithm was proposed. Comparing to previous approaches we designed the encoding for self-adaptation not only one parameter or several ones but for all possible parameters of genetic algorithms at the same time. Also the multicriteria system for timetable evaluation is presented.