TY  - BOOK
AU  - Leonor, Daniellle Ann Marie de Leon.
TI  - Genetic algorithm with shuffled frog leaping algorithm for the University course timetabling problem
PY  - 2009///
KW  - College of Science and Mathematics, University of the Philippines Mindanao
KW  - Davao City
KW  - Philippines
KW  - Genetic algorithm
KW  - Hybrid methods
KW  - SFLA (Shuffled frog leaping algorithm)
KW  - University course timetabling problem
KW  - Weights assignment
KW  - Darwin's Theory of Evolution
KW  - Survival of the fittest
KW  - Timetabling
KW  - Chromosome representation
KW  - Mutation operators
KW  - Maximazation algorithm
KW  - Selection operators
KW  - Recombination operators
KW  - Solution representation
KW  - Fitness evaluation operators
KW  - Undergraduate Thesis
KW  - AMAT200
N1  - Thesis (BS Applied Mathematics) -- University of the Philippines Mindanao, 2009
N2  - University course timetabling problem is arrangement of lecture classes of the teachers with blocks of students on certain timeslots and classrooms, satisfying several kinds of constraints. Genetic algorithm (GA), inspired from Darwin's theory of evolution, is usually used as research algorithm in timetabling problems. Though GA is a promising method in solving the university course timetabling problem of the University of the Philippines Mindanao-College of Science and Mathematics, it converges rather slowly. A newly developed algorithm, the shuffled frog leaping algorithm (SFLA), has a deep local search strategy that would help improve GA method. Thus, in this study, the concepts of shuffling and partitioning from the shuffled frog leaping algorithm were inserted in the general genetic process to aid the convergence of pure GA, named as GA-SFLA. Small, medium and large population sizes were explored and optimal sets of parameter values per population size were determined for GA-SFLA. Also, penalty weights were developed to enhance the quality of the generated timetables. Results showed that GA-SFLA outperformed pure GA on all population sizes. However, the different parameters set that used GA-SFLA were not significantly different from each other. These results were further verified by statistical analysis. Although present results showed a good performance for GA-SFLA, further studies are still needed to explore other potentials of GA-SFLA
ER  -