Time Series Forecasting for Determining Economic Order Quantity of Silicon Rubber Manufacturer

Abstract

The objectives of this research were 1) to study the pattern of time series forecasting for inventory and consumption. 2) To propose guidelines for inventory management with economic order quantity and safety stock. The research started by collecting sales by Product K between 2017 and 2019 of silicone rubber manufacturers. The data collection was done by an interviewing the three salespersons, reviewing of sale documents and observing selling. The consumption data were analyzed using ABC Analysis and selected consumption A to investigate. Then, a forecasting model was creating by adopting the concept of time series. Subsequently, the variance of the dataset was measured and the EOQ was calculated. Finally, the safety stock was explored by using the result of the sales forecast, economic order quantity (EOQ), and raw materials required for production. The findings indicated that 1) exponential smoothing time series forecast with α = 0.1 showed the most accurate result. This way identified the lowest error (MAD = 55,745.43 and MAPE = 17.83%). 2) Using Safety Stock and Economic Order Quantity can reduce the cost of raw material inventory. If the silicone rubber factory uses the research results, it will reduce the cost of raw material inventory enough to support production and sales in a competitive environment that tends to increase steadily.