Skripsi

PEMBANGUNAN MODEL PREDIKSI PERMINTAAN TRANSPORTASI UDARA DI INDONESIA BERDASARKAN FAKTOR EKONOMI DAN DEMOGRAFIS

---

Air transportation plays an essential role in Indonesia, the world’s largest archipelagic country, in supporting connectivity and national economic growth. This study aims to predict the number of air passengers using a machine learning model based on annual time series data (1970–2021) from the World Bank. In the preprocessing stage, a Fuzzy Logic method based on the Non-Stationary Fuzzy Time Series (NSFTS) approach with the Interquartile Range (IQR) technique was applied to improve feature quality and representation. The model was developed using 12 features covering key economic and demographic variables, including Gross Domestic Product (GDP), population growth, urbanization rate, household consumption, and an external factor represented categorically, namely the pandemic status. Evaluation using a time-based holdout approach showed that the Support Vector Regression (SVR) model achieved the best performance with an R² value of 97%, MAPE of 2.00%, MAE of 0.11, and MSE of 0.02 in predicting the 2021 data. The model also projected a recovery trend in 2022 with an estimated 55.7 million passengers. Therefore, the combination of Fuzzy Logic and SVR methods proved effective in capturing historical patterns and supporting national air transportation policy planning. Keywords: Indonesia Air Transport Demand, Machine Learning, Prediction, Fuzzy Logic, Socio-Economic Factors, Pandemic Impact.

---

Availability

Detail Information

Series Title

-

Call Number

T1854612025

Publisher

Indralaya : Prodi Sistem Komputer, Fakultas Ilmu Komputer Universitas Sriwijaya., 2025

Collation

xviii, 101 hlm.; ilus.; tab.; 29 cm.

Language

Indonesia

ISBN/ISSN

-

Classification

519.207

Content Type

Text

Media Type

unmediated

Carrier Type

other (computer)

Edition

-

Subject(s)

Prodi Sistem Komputer
Prediksi Jumlah Penumpang

Specific Detail Info

-

Statement of Responsibility

MI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment