Optimization Algorithms: A Comparison Study for Scheduling Problem at UIN Raden Fatah's Sharia and Law Faculty

Mustakim, . and Tri Basuki, Kurniawan and Misinem, . and Edi Surya, Negara and Izman, Herdiansyah (2024) Optimization Algorithms: A Comparison Study for Scheduling Problem at UIN Raden Fatah's Sharia and Law Faculty. Journal of Data Science, 2024 (57). pp. 1-19. ISSN 2805-5160

	Text jods2024_57.pdf - Published Version Available under License Creative Commons Attribution. Download (305kB)
	Text 605 - Published Version Available under License Creative Commons Attribution. Download (25kB)

Abstract

The rapid advancement of information and communication technology significantly impacts various sectors, including education, by enhancing administrative and academic processes through sophisticated algorithms and systems. At Raden Fatah State Islamic University Palembang, specifically within the Faculty of Sharia and Law, technology is pivotal in managing complex course scheduling challenges due to increasing student numbers and curriculum intricacies. This study examines the effectiveness of optimization algorithms in improving the efficiency and quality of academic scheduling. We focus on two prominent optimization techniques, Genetic Algorithms (GA) and Ant Colony Optimization (ACO), chosen for their capability to address the complex optimization problems typical in academic settings. The research encompasses a systematic approach, beginning with a clear definition of constraints and objectives, followed by designing and implementing both algorithms to address the scheduling issues at the Faculty of Sharia and Law. Our experimental evaluation compares the performance of GA and ACO across multiple metrics, including execution time, memory usage, fitness, and adaptability to dynamic conditions. Results indicate that while GA generally offers faster solutions, it requires more memory and shows variability in achieving optimal fitness levels. Conversely, ACO, though occasionally slower, consistently produces higher quality solutions with greater memory efficiency, making it more suitable for resource-constrained environments. The best results from the experiments highlight that ACO outperformed GA in terms of overall solution quality and resource efficiency, with an execution time of 19.27 seconds and 14,218.14 KB. Specifically, ACO consistently achieved near-optimal fitness scores with significantly lower memory usage compared to GA. This demonstrates ACO's robustness and suitability for handling complex scheduling problems where resource conservation is crucial. The choice between GA and ACO should be influenced by specific situational requirements—GA is recommended where speed is critical, while ACO is preferable in settings requiring high-quality, resource-efficient solutions. Future research should explore refining these algorithms, possibly through hybrid approaches that leverage the strengths of both to enhance their effectiveness and adaptability in complex scheduling scenarios. This study not only informs the academic community about effective scheduling practices but also sets a benchmark for future technological implementations in educational institutions.

Item Type:	Article
Uncontrolled Keywords:	Genetic Algorithm, Ant Colony Optimization, Course Scheduling
Subjects:	Q Science > Q Science (General) Q Science > QA Mathematics Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software
Depositing User:	Unnamed user with email masilah.mansor@newinti.edu.my
Date Deposited:	28 Nov 2024 04:28
Last Modified:	28 Nov 2024 04:28
URI:	http://eprints.intimal.edu.my/id/eprint/2064

Actions (login required)

View Item