View Item 
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - School of Data Science, Mathematic and Informatics
      • UT - Statistics and Data Sciences
      • View Item
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - School of Data Science, Mathematic and Informatics
      • UT - Statistics and Data Sciences
      • View Item
      JavaScript is disabled for your browser. Some features of this site may not work without it.

      Peramalan Produksi Crude Palm Oil (CPO) di Pulau Sumatera Menggunakan Neural Network Autoregressive (NNAR)

      Thumbnail
      View/Open
      Cover (759.2Kb)
      Fulltext (3.307Mb)
      Lampiran (1.553Mb)
      Date
      2025
      Author
      Harahap, Farhan Abdillah
      Anisa, Rahma
      Djuraidah, Anik
      Metadata
      Show full item record
      Abstract
      Sektor kelapa sawit memiliki peran strategis dalam perekonomian Indonesia. Pada 2023, ekspor CPO (Crude Palm Oil) mencapai USD 25,61 miliar atau 9,90% dari total ekspor nasional. Pulau Sumatera menjadi wilayah penghasil utama, menghasilkan 25,69 juta ton (54,56% dari produksi nasional) dari 8,74 juta hektare (54,86% dari luas perkebunan nasional). Dominasi produksi CPO di Pulau Sumatera menunjukkan perlunya dukungan proyeksi yang sistematis dan terukur. Penelitian ini memanfaatkan data produksi bulanan CPO dan luas areal dari sepuluh provinsi di Sumatera selama Januari 2012 hingga Desember 2023 untuk membangun model peramalan tahun 2024. Metode Neural Network Autoregressive (NNAR) dipilih karena kemampuannya menangkap pola non-linear dalam data deret waktu. Model dikombinasikan dengan kerangka expanding window, penentuan lag optimal menggunakan Partial Autocorrelation Function (PACF), dan hyperparameter tuning dengan grid search terhadap jumlah neuron, learning rate decay, dan maxit. Evaluasi menggunakan Mean Absolute Percentage Error (MAPE) menunjukkan kinerja baik, dengan 9 dari 10 provinsi mencatat MAPE di bawah 20% dan MAPE tingkat pulau sebesar 7,49%. Produksi CPO tahun 2024 diproyeksikan mencapai 25,32 juta ton, dengan Provinsi Riau sebagai kontributor terbesar.
       
      The palm oil sector played a strategic role in Indonesia’s economy. In 2023, CPO (Crude Palm Oil) exports reached USD 25.61 billion or 9.90% of total national exports. Sumatra Island was the main production area, producing 25.69 million tons (54.56% of national production) from 8.74 million hectares (54.86% of national plantation area). The dominance of CPO production in Sumatra Island indicated the need for systematic and measurable projection support. This study used monthly CPO production and plantation area data from ten provinces in Sumatra from January 2012 to December 2023 to develop a forecasting model for 2024. The Neural Network Autoregressive (NNAR) method was chosen for its ability to capture non-linear patterns in time series data. The model was combined with an expanding window framework, optimal lag selection using the Partial Autocorrelation Function (PACF), and hyperparameter tuning using grid search for the number of neurons, learning rate decay, and maxit. Evaluation using Mean Absolute Percentage Error (MAPE) showed good performance, with 9 out of 10 provinces recording MAPE below 20% and island-level MAPE of 7.49%. CPO production in 2024 was projected to reach 25.32 million tons, with Riau Province as the largest contributor.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/165954
      Collections
      • UT - Statistics and Data Sciences [82]

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository
        

       

      Browse

      All of IPB RepositoryCollectionsBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

      My Account

      Login

      Application

      google store

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository