Sulistyowati, Ratna (2018) Model Peramalan Hibrida untuk Prediksi Jumlah Penumpang Udara dan Volume Kargo di Indonesia. Masters thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
1316201706-Master_Thesis.pdf - Accepted Version Download (10MB) | Preview |
Abstract
Peramalan untuk penumpang udara dan kargo memiliki pengaruh besar terhadap master plan pengembangan infrastruktur bandara dan investasi oleh maskapai penerbangan sipil. Penelitian ini bertujuan untuk memperoleh nilai ramalan yang akurat dari data penumpang dan kargo di tiga bandara internasional Indonesia yaitu Bandara Internasional Soekarno Hatta, Bandara Internasional I Gusti Ngurah Rai, dan Bandara Internasional Juanda. Ketiga bandara internasional tersebut mempunyai kontribusi terbesar terhadap jumlah penumpang udara dan volume kargo di Indonesia. Penelitian ini menggunakan metode peramalan hibrida yang menggabungkan dua model linier dan nonlinier melalui dua tahapan pemodelan, sehingga melalui metode hibrida diharapkan dapat menghasilkan ramalan yang lebih akurat. Tahap pertama adalah melakukan pemodelan linier yaitu dengan model time series regression (TSR) dan Autoregressive Integrated Moving Average with Exogenous Factor (ARIMAX). Selanjutnya tahap kedua melakukan
pemodelan hibrida yang bekerja menggunakan input residual dari model linier yang dimodelkan dengan metode yang lebih kompleks seperti pendekatan machine learning yaitu Neural Network (NN) dan Support Vector Regression (SVR), untuk
menangkap pola nonlinier. Penelitian ini membandingkan akurasi hasil peramalan empat model peramalan hibrida antara lain hibrida TSR-NN, TSR-SVR, ARIMAX-NN, dan ARIMAX-SVR berdasarkan nilai Mean Absolute Percentage Error
(MAPE). Hasil penelitian menunjukkan bahwa pemodelan hibrida menggunakan NN (ARIMAX-NN dan TSR-NN) pada data testing menghasilkan peramalan yang lebih baik dibandingkan dengan model hibrida menggunakan SVR (TSR-SVR dan ARIMAX-SVR).
========================================================================================================
Forecasting of air passengers and cargo have a major influence on the
master plan of the airport infrastructure development and investment by the civil
airline. This research aims to obtain the most accurate predictive value of the air
passengers and cargo
at three international airports Indonesia, namely, Soekarno
Hatta Airport, I Gusti Ngurah Rai Airport, and Juanda Airport. Those international
airports are the three largest contributors to the number of air passengers and cargo
volumes in Indonesia. This
research uses a hybrid forecasting method that combines
linear models and nonlinear models. The combination of two models linear and
nonlinear is able to obtain accurate predictions.
The first
phase
is linear modeling
with time series regression model (TS
R) and Autoregressive Integrated Moving
Average with Exogenous Factor (ARIMAX).
In the second phase, hybrid modeling
works using
the
error
of linear models modeled by more complex machine learning
methods such as Neural Network (NN) and Support Vector Regr
ession (SVR)
,
to
capture
the
nonlinear patterns.
This paper compares the accuracy of forecasting the
four hybrid models include TSR
-
NN, TSR
-
SVR, ARIMAX
-
NN, and ARIMAX
-
SVR based on the Mean Absolute Percentage Error (MAPE). The results show that
modeling h
ybrid using NN, i.e. ARIMAX
-
NN and TSR
-
NN have better
performance prediction than modeling hybrid using SVR
, i.e. TSR-SVR and ARIMAX-SVR.
| Item Type: | Thesis (Masters) |
|---|---|
| Additional Information: | RTSt 519.55 Sul m-1 3100018075149 |
| Uncontrolled Keywords: | Penumpang udara, Kargo, Regresi time series, ARIMAX, Neural network, Support vector regression, Hibrida, Air passenger, Cargo, Time series regression, Hybrid |
| Subjects: | H Social Sciences > HA Statistics Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression |
| Divisions: | Faculty of Mathematics and Science > Statistics > 49101-(S2) Master Thesis |
| Depositing User: | Ratna Sulistyowati |
| Date Deposited: | 22 Feb 2018 08:25 |
| Last Modified: | 13 Apr 2020 01:13 |
| URI: | http://repository.its.ac.id/id/eprint/50977 |
Actions (login required)
 |
View Item |