SIMULASI PERAMALAN PERMINTAAN DENGAN REGRESI LINEAR, ARIMA, DAN HOLT-WINTERS PADA INDUSTRI KEMASAN FLEKSIBEL

    Fahmi Arif Fajar, - and Mumu Komaro, - and Hanissa Okitasari, - (2025) SIMULASI PERAMALAN PERMINTAAN DENGAN REGRESI LINEAR, ARIMA, DAN HOLT-WINTERS PADA INDUSTRI KEMASAN FLEKSIBEL. S1 thesis, Universitas Pendidikan Indonesia.

    Abstract

    Industri kemasan fleksibel mengalami pertumbuhan pesat secara global yang didorong oleh perubahan pola konsumsi, urbanisasi, dan keunggulan biaya yang kompetitif. PT. XYZ, sebagai perusahaan multinasional di sektor ini, menghadapi tantangan signifikan berupa kesalahan dalam peramalan permintaan dengan nilai eror 27,4% pada periode 2023–2024, yang memicu kerugian akibat overstock bahan baku. Penelitian ini bertujuan untuk simulasi peramalan permintaan produk kemasan fleksibel pada PT. XYZ. Penelitian ini menggunakan tiga pendekatan statistik yaitu Regresi Linear, Autoregressive Integrated Moving Average (ARIMA), dan Holt-Winters model Additive dan Multiplicative dengan pemrograman R melalui perangkat lunak RStudio untuk melakukan pemodelan, simulasi, dan validasi, termasuk uji diagnostik residual dan pengukuran kesalahan dengan MAPE dan U-Theil. Hasil dari penelitian ini berupa perbandingan tingkat akurasi ketiga metode tersebut dan menentukan metode terpilih dengan nilai eror terkecil. Metode Holt-Winters Additive merupakan metode dengan performa terbaik dengan nilai MAPE terkecil yaitu 13,28% sehingga meningkatkan akurasi peramalan permintaan pada PT. XYZ secara signifikan. Peningkatan akurasi peramalan disebabkan oleh penerapan metode statistik berbasis data historis dan pemilihan model yang sesuai dengan karakteristik data. The flexible packaging industry is experiencing rapid growth globally driven by changing consumption patterns, urbanization, and competitive cost advantages. PT. XYZ, as a multinational company in this sector, faces significant challenges in the form of errors in demand forecasting with an error value of 27,4% in the 2023–2024 period, which triggers losses due to overstocking of raw materials. This study aims to simulate demand forecasting for flexible packaging products at PT. XYZ. This study uses three statistical approaches, namely Linear Regression, Autoregressive Integrated Moving Average (ARIMA), and Holt-Winters Additive and Multiplicative models with R programming through RStudio software to perform modelling, simulation, and validation, including residual diagnostic tests and error measurements with MAPE and U-Theil. The results of this study are a comparison of the accuracy levels of the three methods and determine the selected method with the smallest error value. The Holt-Winters Additive method is the method with the best performance with the smallest MAPE value of 13.28%, thus significantly increasing the accuracy of demand forecasting at PT. XYZ. The increase in forecasting accuracy is due to the application of statistical methods based on historical data and the selection of models that suit the characteristics of the data.

    Baca Full Text klik disini
    [thumbnail of S_TL_2101911_Title.pdf] Text
    S_TL_2101911_Title.pdf
    Download (830kB)
    [thumbnail of S_TL_2101911_Chapter1.pdf] Text
    S_TL_2101911_Chapter1.pdf
    Download (421kB)
    [thumbnail of S_TL_2101911_Chapter2.pdf] Text
    S_TL_2101911_Chapter2.pdf
    Restricted to Staf Perpustakaan
    Download (1MB)
    [thumbnail of S_TL_2101911_Chapter3.pdf] Text
    S_TL_2101911_Chapter3.pdf
    Download (415kB)
    [thumbnail of S_TL_2101911_Chapter4.pdf] Text
    S_TL_2101911_Chapter4.pdf
    Restricted to Staf Perpustakaan
    Download (1MB)
    [thumbnail of S_TL_2101911_Chapter5.pdf] Text
    S_TL_2101911_Chapter5.pdf
    Download (277kB)
    [thumbnail of S_TL_2101911_Appendix.pdf] Text
    S_TL_2101911_Appendix.pdf
    Restricted to Staf Perpustakaan
    Download (3MB)
    Official URL: https://repository.upi.edu/
    Item Type: Thesis (S1)
    Additional Information: https://scholar.google.com/citations?view_op=list_works&hl=en&user=SQ2wzjIAAAAJ ID SINTA Dosen Pembimbing Mumu Komaro: 5993878 Hanissa Okitasari: 6753821
    Uncontrolled Keywords: ARIMA, Holt-Winters Additive, Kemasan Fleksibel, MAPE, Peramalan Permintaan ARIMA, Demand Forecasting, Flexible Packaging, Holt-Winters Additive, MAPE
    Subjects: T Technology > T Technology (General)
    T Technology > TS Manufactures
    Divisions: Fakultas Pendidikan Teknik dan Industri > Teknik Logistik - S1
    Depositing User: Fahmi Arif Fajar
    Date Deposited: 22 Sep 2025 04:31
    Last Modified: 23 Sep 2025 10:05
    URI: http://repository.upi.edu/id/eprint/140040

    Actions (login required)

    View Item View Item