Peramalan Permintaan Semen di PT. XYZ Menggunakan Metode Time Series Regression dan ARIMA

Wirdyacahya, Berliana Salsa (2021) Peramalan Permintaan Semen di PT. XYZ Menggunakan Metode Time Series Regression dan ARIMA. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

Text 10611710000073-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2023. Download (2MB) | Request a copy

Abstract

Tahun 2020, infrastruktur pembangunan di Indonesia sedikit terhambat karena adanya penyesuaian anggaran APBN akibat pandemi Covid-19. Terhambatnya pembangunan infrastukur menyebabkan permintaan semen nasional ikut menurun dan berimbas pada jumlah permintaan semen di PT. XYZ. Penurunan jumlah permintaan semen di PT. XYZ juga terjadi pada waktu bulan Ramadhan hingga Hari Raya Idul Fitri setiap tahunnya. Hingga saat ini PT. XYZ masih menggunakan metode winter’s exponential smoothing, dekomposisi, dan Time Series Regression (TSR) untuk meramalkan jumlah permintaan semen periode yang akan datang, namun metode tersebut menghasilkan nilai kesalahan yang besar. Pada penelitian ini akan dilakukan peramalan jumlah permintaan semen di PT. XYZ menggunakan data bulanan sejak Januari 2015 hingga Desember 2020. Metode peramalan yang digunakan adalah TSR dengan menambahkan efek intervensi adanya Covid-19, efek variasi kalender waktu terjadinya bulan Ramadhan hingga Hari Raya Idul Fitri, serta efek musiman. Metode ini dibandingkan dengan metode peramalan lainnya yaitu ARIMA. Hasil penelitian yang diperoleh menunjukkan bahwa model peramalan terbaik adalah model TSR dengan variabel dummy berupa efek intervensi adanya Covid-19, efek variasi kalender waktu terjadinya bulan Ramadhan hingga Hari Raya Idul Fitri, serta efek musiman periode bulan 2,8,9,10, dan 11. Hasil ramalannya menunjukkan terjadi kenaikan jumlah permintaan semen pada tahun 2021.
================================================================================================
In 2020, infrastructure development in Indonesia was slightly hampered due to the pandemic Covid-19, resulting in adjustments of APBN. Obstracted of infrastructure development caused the national cement demand to decrease and resulted in the amount of cement demand in PT. XYZ. In addition, the decrease in the number of cement demand in PT. XYZ also occurs during Ramadan until Eid Fitr every year. Until now, PT. XYZ still uses winter's exponential smoothing, decomposition, and Time Series Regression (TSR) methods to forecast the number of cement demand. However, these methods generate large error values. So, in this study will be conducted forecasting the number of cement demand in PT. XYZ uses monthly data from January 2015 to December 2020. The forecasting method used is TSR by adding the intervention effect of Covid-19, the effect of variations calendar during Ramadan until Eid Fitr, as well as seasonal effects. This method compared to other forecasting methods namely ARIMA. The results of the research obtained that the best forecasting model is the TSR model with dummy variables in the form of intervention effects of Covid-19, the effect of variations calendar during Ramadan until Eid Fitr, as well as seasonal effects with monthly periods 2,8,9,10, and 11. Forecasting results shows an increase in the number of cement demand in 2021.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	ARIMA, Covid-19, Intervensi, Peramalan, Permintaan Semen, Time Series Regression, Variasi Kalender, Musiman, Calender Variation, Cement Demands, Forecasting, Interventions, Seasonal
Subjects:	H Social Sciences > HA Statistics H Social Sciences > HA Statistics > HA30.3 Time-series analysis H Social Sciences > HD Industries. Land use. Labor Q Science Q Science > QA Mathematics > QA280 Box-Jenkins forecasting
Divisions:	Faculty of Vocational > 49501-Business Statistics
Depositing User:	Berliana Salsa Wirdyacahya
Date Deposited:	14 Aug 2021 17:35
Last Modified:	14 Aug 2021 17:35
URI:	http://repository.its.ac.id/id/eprint/86534

Actions (login required)

View Item