Demand Forecasting and Inventory Prediction for Apparel Product using the ARIMA and Fuzzy EPQ Model

Abstract

Demand forecasting is essential for almost companies, especially in the garment industry. Accurate forecasting will provide data input for efficient production, inventory, and distribution planning, thereby helping businesses reduce operating costs and improve supply efficiency. This paper will forecast demand for long-sleeved shirts at Tay Do Garment Company with ARIMA (Autoregressive Integrated Moving Average) model. Firstly, monthly data from January 2016 to December 2020 build models based on the Arima model. The results indicated better efficiency after comparing ARIMA accuracy with Holt's regression model. Besides, the ARIMA model is predicted demand for the period from January 2021 to December 2021. Moreover, the forecast data will be used to survey the "Fuzzy economic production quantity" (Fuzzy EPQ) inventory model. Besides, using Arena software is to simulate the production speed of the line. Finally, the optimal batch size per production lot and the number of production times per year to minimize costs are suggested.