Forecasting Demand for MSME Products Using the ARIMA Method (Case Study of Berlian Salak Processing in Sleman, Yogyakarta)

Maharani, Meylany Putri (2025) Forecasting Demand for MSME Products Using the ARIMA Method (Case Study of Berlian Salak Processing in Sleman, Yogyakarta). Tugas Akhir thesis, University of Technology Yogyakarta.

Text Abstrak_Meylany Putri Maharani_5201811016_Sains Data.docx Download (20kB)

Abstract

ABSTRACT This study discusses the forecasting of demand for processed salak products at the "Prolahan Salak Berlian" MSME using the ARIMA method. The main problem faced is the uncertain demand for products, so that business actors have difficulty in predicting the stock that must be provided in the coming months. The purpose of this study is to provide a benchmark for business actors in determining the amount of production so that there is no shortage or excess stock. The data used is the demand data for the Olahan Salak Berlian product from January 2022 to February 2024. The method used is pure ARIMA, which is implemented using the Python programming language on Google Collaboratory. The prediction results show that there is no increase or decrease in sales for Jenang Salak Sesame, Dodol Salak, Geplak Salak, and Processed Salak products in the next 12 months. The model implementation stage begins by finding the best parameters for the forecasting, modeling, and error evaluation processes. The best parameters obtained for modeling Jenang Salak Sejen are the ARIMA model with parameters (0, 1, 2), Dodol Salak with the ARIMA model with parameters (0, 0, 1), Geplak Salak with the ARIMA model with parameters (1, 0, 2), and Processed Salak with the ARIMA model parameters (2, 1, 2). Based on the results of the residual test, the ARIMA model is quite accurate for several products such as Jenang Salak Sejen and Processed Salak. However, in Dodol Salak, problems in statistical calculations need to be fixed to obtain more accurate conclusions. However, the results of the model evaluation show that the error value is quite high so that the model cannot be said to be good. This can be an indication that the ARIMA model has not been able to accurately predict the sales data for this Olahan Salak Berlian product. Keywords: Forecasting, Product Demand, ARIMA, Processed Salak Berlian.

Item Type:	Thesis (Skripsi, Tugas Akhir or Kerja Praktek) (Tugas Akhir)
Subjects:	T Technology > T Technology (General)
Divisions:	Fakultas Sains Dan Teknologi > Data Science
Depositing User:	Sains Data
Date Deposited:	07 May 2025 03:40
Last Modified:	08 May 2025 01:39
URI:	http://eprints.uty.ac.id/id/eprint/17894

Actions (login required)

View Item