| Scheduling arrangement is an important task in the school teaching work, and the actual situation that there is relatively few teaching resource makes the course arrangement not simple. Now most universities have used the computer aided scheduling, but have not used the computer to complete the whole process of course arrangement. Most of these systems have solved the problem of resource conflict, but have not solved the problem of the optimization of resource allocation. This paper uses a genetic algorithm with self-organizing, self-adapting and self-learning ability to solve the problem, and to make the system more adaptable to the school actual situation.The paper has done the detailed demand analysis for arrangement system. Three kinds of users including the system administrator, teaching staff, teachers and students have been determined.The overall architecture and the functional modules of the systemt have been designed according to the different needs of these three kinds of users. Using the E-R diagram and data table form, it completes the system database design. In the system modeling stage, it respectively uses UML modeling language class diagram and sequence diagram to establish static structure and make dynamic behavior analysis. In the detailed design of genetic algorithm, it uses three-dimensional coding to make scheduling five elements relative to three-dimensional coordinates; the fitness function is designed by considering the special course arrangement, the distribution of courses which need more times, the class hour daily distribution uniformity, the expected value of teachers’ class distribution, the problem of acrossing teaching area and so on; in the genetic operation, it selects the roulette wheel selection method and the optimal preservation for two optimized preservation strategy; in order to make the system suitable for different data, it uses the crossover probability and mutation probability with self-adaptive. Finally, it realizes the course timetabling operation according to the working process of genetic algorithm. On the system implementation, it uses C#programming language for program design, uses SQL Server2008database management for background. By actual test, the system generates the ideal schedule shortly and improves the efficiency of the process. |