PERAMALAN KUNJUNGAN WISATAWAN MANCANEGARA MENGGUNAKAN MODEL SARIMA-XGBOOST DENGAN OPTIMASI OPTUNA

Malva, Maisie Yunita (2025) PERAMALAN KUNJUNGAN WISATAWAN MANCANEGARA MENGGUNAKAN MODEL SARIMA-XGBOOST DENGAN OPTIMASI OPTUNA. Undergraduate thesis, Universitas Pembangunan Nasional "Veteran" Jawa Timur.

	Text 21081010038-cover.pdf Download (1MB)
	Text 21081010038-bab1.pdf Download (260kB)
	Text 21081010038-bab2.pdf Restricted to Repository staff only until 5 December 2028. Download (1MB)
	Text 21081010038-bab3.pdf Restricted to Repository staff only until 5 December 2028. Download (860kB)
	Text 21081010038-bab4.pdf Restricted to Repository staff only until 5 December 2028. Download (6MB)
	Text 21081010038-bab5.pdf Download (242kB)
	Text 21081010038-daftarpustaka.pdf Download (212kB)
	Text 21081010038-lampiran.pdf Restricted to Repository staff only until 5 December 2028. Download (1MB)

Abstract

The Indonesian tourism sector has experienced significant changes following the COVID-19 pandemic, highlighting the need for accurate forecasting systems to support data-driven policy decisions. This study aims to predict the number of international tourist arrivals through six international airports in Indonesia using a combination of SARIMA and XGBoost models optimized with Optuna. The data used are monthly observations from January 2008 to December 2024. In this combined approach, SARIMA captures long-term trends and seasonal patterns, while XGBoost models the residuals to correct prediction errors. Hyperparameter optimization using Optuna was conducted to obtain the best-performing model configuration. The results indicate that the optimized SARIMA-XGBoost combination achieves the highest prediction accuracy across all airports, with MAPE values of 3.08% for Ngurah Rai, 3.73% for Soekarno-Hatta, 6.36% for Juanda, 6.82% for Kualanamu, 3.07% for Sam Ratulangi, and 7.84% for Minangkabau. Hyperparameter optimization consistently improved prediction accuracy compared to models without optimization. Therefore, the SARIMA-XGBoost-Optuna combination represents an effective and reliable approach to support planning and decision-making in the Indonesian tourism sector.

Item Type:	Thesis (Undergraduate)
Contributors:	Contribution Contributors NIDN/NIDK Email Thesis advisor Sari, Anggraini Puspita NIDN0716088605 anggraini.puspita.if@upnjatim.ac.id Thesis advisor Puspaningrum, Eva Yulia NIDN0005078908 evapuspaningrum.if@upnjatim.ac.id
Subjects:	T Technology > T Technology (General)
Depositing User:	Maisie Yunita Malva
Date Deposited:	08 Dec 2025 01:28
Last Modified:	08 Dec 2025 01:28
URI:	https://repository.upnjatim.ac.id/id/eprint/48105

Actions (login required)

View Item