Penerapan Fuzzy Time Series Dalam Peramalan Produksi Crude Palm Oil (CPO) Di Indonesia

Pristka, Asna Norma (2023) Penerapan Fuzzy Time Series Dalam Peramalan Produksi Crude Palm Oil (CPO) Di Indonesia. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 10611910000002-Undergraduate_Thesis for D4 Program.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (3MB) | Request a copy

Abstract

Crude Palm Oil (CPO) menjadi salah satu komoditas cukup penting bagi perekonomian Indonesia, karena sebagai penghasil minyak nabati yang sangat dibutuhkan dan pasarnya telah mendunia. Kontribusi Crude Palm Oil (CPO) terhadap PDB sebesar 13,28%. Indonesia menjadi negara pengekspor CPO terbesar, dengan negara Tiongkok dan India menjadi negara tujuan ekspor terbesar. Dua tahun terakhir produksi CPO mengalami penurunan padahal 3 tahun sebelumnya produksi CPO selalu mengalami peningkatan. Penurunan ini disebabkan karena gangguan cuaca, keterbatasan pupuk, dan kelangkaan tenaga kerja. Sebagai komoditas cukup penting sangatlah penting bagi Indonesia untuk menjamin ketersediaan produksi minyak kelapa sawit di masa mendatang. Diperlukan peramalan untuk mempersiapkan hal-hal yang perlu dalam proses produksi. Metode baru dalam peramalan produksi CPO menggunakan fuzzy time series chen dan fuzzy time series cheng. Data yang digunakan data bulanan dari Januari 2011 – Desember 2021. Hasil analisis dan pembahasan statistika deskriptif sebagai gambaran umum data produksi CPO mengalami peningkatan setiap tahunnya akan tetapi tahun 2021 mengalami penurunan, jika dilihat dari akumulasi perbulan produksi CPO paling tinggi pada bulan Oktober. Berdasarkan analisis fuzzy time series chen dan cheng terbentuk 8 kelas yang digunakan dengan nilai tengah masing-masing kelas dan Panjang interval sama rata, himpunan fuzzy yang terbentuk sebanyak 8, dengan fuzzyfication data setiap periode menyesuaikan dengan interval yang sudah terbentuk, dari fuzzyfication yang terbentuk dapat diketahui hubungan relasi fuzzy dan FLRG untuk menghitung deffuzification ramalan fuzzy time series chen. Peramalan fuzzy time series cheng sama seperti langkah-langkah fuzzy time series chen, namun setelah pembentukan FLRG terdapat matriks pembobot dari FLR yang memiliki current state yang sama untuk menghitung defuzzfication ramalan. Model terbaik yang dipilih untuk meramalakan periode 2022 berdasarkan nilai MAPE, MAD, RMSE adalah fuzzy time series cheng karena memiliki akurasi lebih kecil dari fuzzy time series chen dengan nilai MAPE 6,035%, MAD 0,259, RMSE 0,322.
================================================================================================================================
Crude Palm Oil (CPO) is one of the most important commodities for the Indonesian economy, because it is a producer of vegetable oil that is needed and the market has gone global. Thecontribution of Crude Palm Oil (CPO) to GDP was 13.28%. Indonesia is the largest CPO exporting country, with China and India being the largest export destination countries. The last two years of CPO production have decreased even though the previous 3 years CPO production has always increased. This decline was due to weather disturbances, fertilizer limitations, and labor scarcity. As an important commodity, it is very important for Indonesia to ensure the availability of palm oil production in the future. Forecasting is needed to prepare the necessary things in the produksi process. The new method of forecasting CPO production uses fuzzy time series chen and fuzzy time series cheng. Data used monthly data from January 2011 – December 2021. The results of the analysis and discussion of descriptive statistics as ageneral increase in CPO production data have increased every year, but in 2021 it has decreased, when viewed from the monthly accumulation of CPO production the highest in October. Based on the fuzzy analysis of the time series chen and cheng formed 8 classes used with the middle value of each class and the length of the interval equal, the set of fuzzy formed as much as 8, with the fuzzyfication of data each period adjusting to the already formed interval, from the formed fuzzyfication can known fuzzy and FLRG relation relationships to calculate deffuzification of fuzzy time series predictions. Fuzzy time series forecasting is the same as chen's fuzzy time series steps, but after the formation of FLRG there ist weighting matrix of FLR which has the same current state to calculate the defuzzfication of forecasts. The best model chosen to forecast the 2022 period based on MAPE, MAD, RMSE values is fuzzy time series cheng because it has less accuracy than chen's fuzzy time series with MAPE value of 6.035%, MAD 0.259, RMSE 0.322.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Crude Palm Oil (CPO), Fuzzy Time Series Chen, Fuzzy Time Series Cheng
Subjects:	H Social Sciences > HA Statistics > HA29 Theory and method of social science statistics Q Science > QA Mathematics > QA276 Mathematical statistics. Time-series analysis. Failure time data analysis. Survival analysis (Biometry) Q Science > QA Mathematics > QA9.64 Fuzzy logic
Divisions:	Faculty of Vocational > 49501-Business Statistics
Depositing User:	Asna Norma Pristika
Date Deposited:	13 Jun 2023 02:58
Last Modified:	13 Jun 2023 02:58
URI:	http://repository.its.ac.id/id/eprint/98092

Actions (login required)

View Item