Peramalan Volume Impor Palm Oil Di Indonesia Menggunakan ARIMA

Adinata, Aditya (2026) Peramalan Volume Impor Palm Oil Di Indonesia Menggunakan ARIMA. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 2043221039-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (2MB)

Abstract

Indonesia merupakan produsen dan eksportir Crude Palm Oil (CPO) terbesar di dunia,namun peningkatan konsumsi CPO domestik akibat program mandatori biodiesel menimbulkanalokasi pasokan bagi industri penyulingan (refinery) lokal menjadi terbatas. Meskipunpemanfaatan biodiesel mampu menghemat biaya impor bahan bakar fosil dan mengurangi emisigas rumah kaca, kondisi ini memicu kebutuhan impor insidental atas komoditas palm oil besertaproduk turunannya (HS 151110000, HS 151190000, dan HS 151329000) guna menjagastabilitas pasokan serta harga pasar nasional. Penelitian ini bertujuan untuk mendeskripsikankarakteristik data, membangun model time series terbaik, dan meramalkan volume impor palmoil di Indonesia. Data yang digunakan merupakan data sekunder bulanan periode 2008–2023sebanyak 192 observasi dari Badan Pusat Statistik (BPS), yang dibagi menjadi data training(168 bulan) dan data testing (24 bulan). Metode peramalan yang digunakan adalah ARIMABox-Jenkins. Hasil analisis menunjukkan bahwa data volume impor tidak memiliki polamusiman dan sangat fluktuatif dengan lonjakan tajam pada tahun 2010, 2013, dan 2019. Modelterbaik yang terbentuk adalah ARIMA (0,1,1) dengan nilai estimasi parameter sebesar 0,8103.Residual model terbukti memenuhi asumsi white noise dan secara visual mendekati distribusinormal. Evaluasi akurasi pemodelan pada data testing menghasilkan nilai MAPE sebesar 52%,MAD sebesar 29,656 Ton, serta RMSE sebesar 43,528 Ton. Hasil peramalan untuk 24 periodeke depan (Januari 2024 hingga Desember 2025) setelah dilakukan back transformationmenunjukkan proyeksi yang konstan sebesar 42,639 Ton per bulan, dengan interval batas atasyang mengalami peningkatan eksponensial hingga 178.287,817 Ton pada Desember 2025akibat akumulasi ketidakpastian jangka panjang. Hasil ini memberikan gambaran penting bagipemerintah dan pelaku industri dalam perencanaan logistik serta kebijakan ketahanan pangandan energi nasional.
===================================================================================================================================
Indonesia is the world's largest producer and exporter of Crude Palm Oil (CPO); however,the increase in domestic CPO consumption driven by the mandatory biodiesel program hasconstrained supply allocation for local refineries. Although the utilization of biodiesel is ableto save fossil fuel import costs and reduce greenhouse gas emissions, this condition triggers theneed for incidental imports of palm oil commodities and their derivatives (HS 151110000, HS151190000, and HS 151329000) to maintain national supply stability and market prices. Thisstudy aims to describe data characteristics, build the best time series model, and forecast palmoil import volume in Indonesia. The data used is monthly secondary data from 2008 to 2023,comprising 192 observations from BPS-Statistics Indonesia (Badan Pusat Statistik), which isdivided into training data (168 months) and testing data (24 months). The forecasting methodutilized is the Box-Jenkins ARIMA. The results indicate that the import volume data exhibitsno seasonal pattern and is highly fluctuating, with sharp spikes occurring in 2010, 2013, and2019. The optimal model identified is ARIMA(0,1,1) with an estimated parameter value of0.8103. The model residuals are proven to satisfy the white noise assumption and visuallyapproximate a normal distribution. Accuracy evaluation on the testing data yields a MAPE of52%, a MAD of 29.656 tons, and an RMSE of 43.528 tons. The forecast results for the next 24periods (January 2024 to December 2025) following back-transformation project a constantvolume of 42.639 tons per month, with an upper bound interval exponentially increasing to178,287.817 tons by December 2025 due to the accumulation of long-term uncertainty. Thesefindings provide critical insights for the government and industry stakeholders in logisticsplanning as well as national food and energy security policies.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	ARIMA, Palm Oil, Peramalan, Volume Impor. ARIMA, Forecast, Import Volume, Palm Oil.
Subjects:	Q Science > QA Mathematics > QA280 Box-Jenkins forecasting
Divisions:	Faculty of Vocational > 49501-Business Statistics
Depositing User:	Aditya Adinata
Date Deposited:	28 Aug 2026 03:35
Last Modified:	28 Aug 2026 03:35
URI:	http://repository.its.ac.id/id/eprint/144394

Actions (login required)

View Item