Abstract—Crossover operator proved to be very important component for supporting diversity strategy, and producing new population. Many researches came out with different operators that are suitable to different types of problems. However, some of crossover techniques usually produce infeasible solution which implies the use of repair function to rebuild the solution. Moreover, some of existing crossover operators is not suitable for high constrained problems. Course timetabling of university is known to be highly constrained optimization problem, where lectures assigned to specific timeslots and rooms, with maintaining feasibility and satisfying soft constraints. In this paper, we present genetic algorithms for solving course timetabling problem. The algorithm consists of new optimized crossover utilizes Exponential Monte Carlo with Counter. The approach is tested and the result demonstrates that our approach is able to produce good quality solutions for course timetable problem.

Index Terms—Genetic algorithms, crossover, course timetabling.

Mufid Mohamed A. Musrati is with High Institute of Industrial Technology Engila, Tripoli (e-mail: mufid12@yahoo.com).
Hanan Ettaher Dagez is with Faculty of IT, University of Tripoli, Libya (e-mail: h.dagez@it.uot.edu.ly).

Cite:Mufid Mohamed A. Musrati and Hanan Ettaher Dagez, "New Optimized Crossover Utilizes Exponential Monte Carlo," International Journal of Computer and Communication Engineering vol. 3, no. 5, pp. 384-387, 2014.