000			02472nam a2200241 4500
001			UPMIN-00000010139
003			UPMIN
005			20221205103606.0
008			221205b |||||||| |||| 00| 0 eng d
040			_aDLC _cUPMin _dupmin
041			_aeng
090		0	_aLG993.5 2004 _bA64 C36
100	1		_aCapacillo, Annie C. _9317
245	0	0	_aUniversity examination timetabling algorithm based on genetic algorithm / _cAnnie C. Capacillo
260			_c2004
300			_a53 leaves
502			_aThesis (BS Applied Mathematics) -- University of the Philippines Mindanao, 2004
520	3		_aExamination timetabling deals with the scheduling of exams to the set of time slots subject to some constraints. Genetic algorithm was used as framework in making the examination timetabling algorithm. An algorithm for each genetic algorithm operators such as initialization, evaluation, selection, recombination, repair and mutation was developed to apply in the examination timetabling problem. The examination timetabling algorithm developed start with obtaining the initial population then evaluating each timetable and the population. After which, the parent timetables are selected and then recombined to produce a new set of populations. The recombined timetable is repaired to maintain feasibility and mutated afterwards. Whichever of the repaired and mutated population has the higher evaluation is compared to the previous population. If the produced population has the higher evaluation than the previous population, it will undergo the same process again, else the iteration stops and the previous population will be the set of the final timetables. The algorithm also end if the number of iteration predetermined is reached. Using the data of the University of the Philippines in Mindanao, the evaluation of the population after three iterations is higher by 0.093015262 compared with the initial population. Since this was done manually, it is recommended to encode the algorithm into a computer program. This study presented only one process of the possible combination of initialization, evaluation, selection, recombination, repair and mutation. It is thus recommended to experiment on the combination of these genetic algorithm operators in order to understand better and improve the performance of the genetic algorithm applied to the timetabling problem.
658			_aUndergraduate Thesis _cAMAT200
905			_aFi
905			_aUP
942			_2lcc _cTHESIS
999			_c442 _d442