Widya, Prasasti Arika (2023) Peramalan Volume Bongkar Muat Petikemas di Terminal Teluk Lamong Menggunakan Metode ARIMA, Variasi Kalender, dan Support Vector Regression. Other thesis, Institut Teknologi Sepuluh Nopember.
Text
06211940000150-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (4MB) | Request a copy |
Abstract
Terminal Teluk Lamong (TTL) merupakan salah satu terminal petikemas yang berlokasi di Jawa Timur. TTL menyediakan layanan operasi bongkar muat petikemas domestik dan internasional. Peramalan volume bongkar muat petikemas domestik akan dilakukan menggunakan ARIMAX dengan efek variasi kalender dikarenakan adanya efek variasi kalender Hari Raya Idul Fitri, sedangkan peramalan volume bongkar muat petikemas internasional dilakukan menggunakan ARIMA dikarenakan tidak ada efek variasi kalender. Selanjutnya kedua metode tersebut akan dibandingkan dengan SVR dikarenakan pola data petikemas internasional yang bersifat nonlinear. SVR merupakan metode machine learning yang dilengkapi dengan beberapa fungsi kernel untuk data linear maupun data nonlinear. Berdasarkan hasil analisis yang telah dilakukan, volume bongkar muat petikemas domestik baik diramalkan menggunakan SVR kernel RBF (C = 6; σ = 0,2; ε = 0,1) dengan RMSE dan MAPE out sample sebesar 6.060 dan 13,53%. Hasil peramalan menggunakan input data aktual terbaru selama satu semester menunjukkan bahwa terdapat penurunan volume bongkar muat petikemas domestik sebesar 5,3% dibandingkan dengan satu tahun sebelumnya. Namun apabila dibandingkan dengan semester satu pada tahun sebelumnya, terdapat kenaikan volume petikemas domestik sebesar 6,3%. Sedangkan volume bongkar muat petikemas internasional baik diramalkan menggunakan ARIMA (0,1,[1,12]) dengan RMSE dan MAPE out sample sebesar 3.717 dan 9,43%. Hasil peramalan menunjukkan bahwa terdapat kenaikan volume petikemas internasional sebesar 19,5%. Selama 12 bulan mendatang, TTL mampu melayani operasi bongkar muat petikemas domestik dan internasional dengan fasilitas yang dimiliki.
===============================================================================================================================
Terminal Teluk Lamong (TTL) is a container terminal located in East Java. The future operations of TTL can be optimzed by considering the growth of supply chain. TTL provides domestic and international container loading and unloading operations services. The ARIMAX model with calendar variation will be used for forecasting domestic container throughput, considering the effect of Hari Raya Idul Fitri. Meanwhile, for international container throughput, the ARIMA model will be used as there is no calendar variation effect. Furthermore, these two methods will be compared with SVR due to nonlinear international container data. SVR is a machine learning method that is equipped with several kernel functions for linear data and nonlinear data. The analysis results indicate that the domestic container throughput is well predicted using the SVR RBF kernel (C = 6; σ = 0,2; ε = 0,1) with RMSE and MAPE out samples of 6.060 and 13,53%. Forecasting results using the latest actual input data for one semester shows a 5,3% decrease in volume compared to the previous year. However, when compared to the first half of the previous year, there was an increase in domestic container volume of 6,3%. Meanwhile, international container throughput is well forecasted using ARIMA (0,1,[1,12]) with RMSE and MAPE out samples of 3.717 and 9,43%. Forecasting results show an increase in international container volume of 19,5%. Over the next 12 months, TTL is able to serve domestic and international container throughput with its own facilities.
Item Type: | Thesis (Other) |
---|---|
Uncontrolled Keywords: | ARIMA, Calendar Variation, Container Throughput, Forecasting, SVR, Bongkar Muat Petikemas, Peramalan, Variasi Kalender. |
Subjects: | H Social Sciences > HA Statistics > HA30.3 Time-series analysis |
Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis |
Depositing User: | Prasasti Arika Widya |
Date Deposited: | 30 Aug 2023 04:45 |
Last Modified: | 30 Aug 2023 04:45 |
URI: | http://repository.its.ac.id/id/eprint/104645 |
Actions (login required)
View Item |