Facilitating Scheduling Courses at USU FMIPA with Genetic Algorithms

Oct 13, 2025 by magicalceylontours.com 112 views

Facilitating Scheduling Courses at USU FMIPA with Genetic Algorithms

Scheduling courses is a crucial task that must be carried out in every tertiary institution. In the University of North Sumatra (USU), especially in the Faculty of Mathematics and Natural Sciences (FMIPA), the scheduling process is still carried out manually by involving human labor. This often takes a long time and has the potential to cause schedule conflicts, such as teaching lecturers at the same time or classrooms that are not available. The manual scheduling process can lead to inefficiencies, errors, and frustration among lecturers and students.

To overcome this problem, this research carries a genetic algorithm approach. Genetic algorithms are computational methods inspired by the biological evolution process. In this study, the genetic algorithm process was implemented in a computer program to produce an optimal course schedule. The use of genetic algorithms in scheduling courses can provide significant benefits in improving the efficiency and effectiveness of the scheduling process.

The Genetic Algorithm Process in Scheduling Courses

The genetic algorithm process is a computational method that mimics the process of natural selection and genetics. The process involves several stages, including population initialization, selection, crossover, mutation, and evaluation.

Population Initialization

The program begins with raising ten initial populations that contain various possible course schedules. Each population is a representation of one potential solution. The initial populations are generated randomly, and each population is a unique combination of course schedules.

Selection

The program then uses the Roulette Wheel Selection method to choose the two best populations from the first ten initial populations. This method provides a greater opportunity for the population with a higher fitness score to be selected. The fitness score is calculated based on the quality of the course schedule, including the minimization of schedule conflicts and the maximization of classroom use.

Crossover

After being selected, the two best populations are mated to use the one-point crossover method. In this method, the program will choose one intersection in the population and exchange genetic information (course schedule) in the part after the intersection point. This process helps to create new populations with improved course schedules.

Mutation

After the crossover process, the program will carry out the mutation process using the exchange mutation method. This method will exchange randomly some course schedules in the population of crossover results to increase genetic diversity and avoid solutions trapped in local optimal points.

Evaluation and Iteration

The process of selection, crossover, and mutation will be repeated repeatedly (iteration) until the amount of iteration specified or until the solution is obtained that meets optimal criteria. The program will evaluate the fitness score of each population and select the best population to proceed to the next iteration.

Research Results

This study succeeded in producing a computer program that implemented a genetic algorithm for scheduling courses at FMIPA USU. This program is able to produce an optimal schedule that is more effective and efficient compared to the manual method that has been used so far.

Advantages of Scheduling with Genetic Algorithms

The use of genetic algorithms in scheduling courses at USU FMIPA provides several advantages, including:

Short Scheduling Time

Genetic algorithms are able to produce an optimal schedule in a shorter time than the manual method. This is because the program can process a large number of possible course schedules in a short time, reducing the time required for scheduling.

Ease of Preparation of Schedules

This program simplifies the process of preparing schedules and reducing the possibility of human error. The program can automatically generate course schedules based on the input data, reducing the need for manual intervention.

Optimal Solution

Genetic algorithms are able to find more optimal solutions by minimizing schedule conflicts and maximizing classroom use. This is because the program can evaluate a large number of possible course schedules and select the best one based on the fitness score.

Flexible

Genetic algorithms can be easily adapted for diverse scheduling needs. The program can be modified to accommodate different scheduling requirements, such as changing the number of courses or the availability of classrooms.

Conclusion

The application of genetic algorithms in scheduling courses at USU FMIPA can provide significant benefits in improving the efficiency and effectiveness of the scheduling process. This program can not only help lecturers and students in getting a more optimal schedule, but can also improve the quality of learning and overall lecture processes. The use of genetic algorithms in scheduling courses can be a valuable tool for tertiary institutions, providing a more efficient and effective way of scheduling courses.

Future Research Directions

Future research can focus on improving the genetic algorithm process, such as:

Developing a more efficient selection method
Improving the crossover and mutation processes
Incorporating additional constraints, such as student preferences or lecturer availability
Developing a web-based interface for the program

By improving the genetic algorithm process, the program can provide even more optimal course schedules, reducing the time and effort required for scheduling and improving the overall learning experience.

Q: What is a genetic algorithm, and how does it work in scheduling courses?

A: A genetic algorithm is a computational method inspired by the biological evolution process. In the context of scheduling courses, a genetic algorithm works by generating a population of possible course schedules, evaluating their fitness, and selecting the best ones to reproduce and evolve into new generations. This process is repeated until an optimal course schedule is obtained.

Q: What are the advantages of using genetic algorithms in scheduling courses?

A: The advantages of using genetic algorithms in scheduling courses include:

Short scheduling time: Genetic algorithms can produce an optimal schedule in a shorter time than the manual method.
Ease of preparation of schedules: The program can automatically generate course schedules based on the input data, reducing the need for manual intervention.
Optimal solution: Genetic algorithms can find more optimal solutions by minimizing schedule conflicts and maximizing classroom use.
Flexible: Genetic algorithms can be easily adapted for diverse scheduling needs.

Q: How does the genetic algorithm process work in scheduling courses?

A: The genetic algorithm process in scheduling courses involves the following stages:

Population initialization: The program generates a population of possible course schedules.
Selection: The program selects the best course schedules based on their fitness.
Crossover: The program combines the selected course schedules to create new ones.
Mutation: The program introduces random changes to the new course schedules to increase genetic diversity.
Evaluation and iteration: The program evaluates the fitness of the new course schedules and selects the best ones to reproduce and evolve into new generations.

Q: What are the constraints of the genetic algorithm process in scheduling courses?

A: The constraints of the genetic algorithm process in scheduling courses include:

Limited number of iterations: The program has a limited number of iterations to find an optimal course schedule.
Limited population size: The program has a limited population size to generate possible course schedules.
Limited computational resources: The program has limited computational resources to process the large number of possible course schedules.

Q: How can the genetic algorithm process be improved in scheduling courses?

A: The genetic algorithm process can be improved in scheduling courses by:

Developing a more efficient selection method
Improving the crossover and mutation processes
Incorporating additional constraints, such as student preferences or lecturer availability
Developing a web-based interface for the program

Q: What are the potential applications of the genetic algorithm process in scheduling courses?

A: The potential applications of the genetic algorithm process in scheduling courses include:

Scheduling courses in different departments or faculties
Scheduling courses in different semesters or academic years
Scheduling courses for different student populations, such as undergraduate or graduate students
Scheduling courses for different lecturer populations, such as full-time or part-time lecturers

Q: What are the potential limitations of the genetic algorithm process in scheduling courses?

A: The potential limitations of the genetic algorithm process in scheduling courses include:

Computational complexity: The program may require significant computational resources to process the large number of possible course schedules.
Limited generalizability: The program may not be generalizable to different scheduling contexts or constraints.
Limited interpretability: The program may not provide clear insights into the scheduling process or the optimal course schedule.

Q: How can the genetic algorithm process be validated in scheduling courses?

A: The genetic algorithm process can be validated in scheduling courses by:

Comparing the optimal course schedule obtained by the program with the manual schedule
Evaluating the fitness of the optimal course schedule using different metrics, such as schedule conflicts or classroom use
Conducting sensitivity analysis to test the robustness of the program to different scheduling constraints or parameters.