FORECASTING BAHAN BAKU RAW SUGAR DENGAN METODE TIME SERIES & USULAN PERENCANAAN SAFETY STOCK PT MEDAN SUGAR INDUSTRY

Abstract

Abstrak

Proses pembuatan gula rafinasi dilakukan melalui pengolahan bahan baku berupa raw sugar, dan memiliki kualitas kemurnian yang sangat tinggi. Terkait persediaan bahan baku yang harus disiapkan, PT Medan Sugar Industry membutuhkan perkiraan jumlah pemakaian bahan baku ke depannya. Estimasi jumlah pemakaian bahan baku dapat diperkirakan melalui teknik forecasting (peramalan). Tujuan dari penelitian ini adalah untuk merancang usulan metode peramalan pemakaian bahan baku raw sugar serta memberikan usulan jumlah safety stock yang dapat diterapkan oleh perusahaan. Penelitian ini menggunakan Metode Time Series untuk melakukan peramalan pemakaian bahan baku raw sugar. Dalam hal ini, Metode Time Series yang digunakan ialah Double Moving Average, Double Exponential Smoothing, dan Holt-Winters Multiplicative. Setelah dilakukan pemilihan metode peramalan yang menghasilkan error terkecil dengan menggunakan metode Mean Square Error (MSE), Mean Absolute Percentage Error (MAPE), dan Mean Absolute Deviation (MAD) untuk menghitung nilai error, kemudian dilakukan perhitungan jumlah safety stock yang dapat diterapkan. Pada kasus ini, diketahui bahwa Holt-Winters Multiplicative memiliki nilai error terendah. Usulan safety stock yang dapat diterapkan oleh perusahaan disesuaikan dengan tingkat pelayanan yang diinginkan, dalam penelitian ini tersedia tingkat pelayanan 90% dan 95%.

Kata kunci: peramalan; time series; nilai error; safety stock

Abstract

[Forecasting Raw Sugar Materials Using Time Series Methods & Designed Safety Stock Planning PT Medan Sugar Industry] The process of making refined sugar is carried out through the processing of raw materials in the form of raw sugar, and has a very high quality of purity. Related with the supply of raw materials that must be prepared, PT Medan Sugar Industry requires the estimated amount of raw material used in the future. This estimated amount of raw material usage can be predicted through forecasting technique. The objective of this study is to design a proposed method of forecasting the use of raw sugar material and to propose regarding the amount of safety stock that can be applied by the company. This study uses the Time Series Method to forecast the use of raw sugar material. In this case, the Time Series Method used is Double Moving Average, Double Exponential Smoothing, and HoltWinters Multiplicative. After the selection of forecasting method that resulted in the smallest error by using the Mean Square Error (MSE), Mean Absolute Percentage Error (MAPE), and Mean Absolute Deviation (MAD) methods to calculate the error value, then calculated the number of safety stock that can be applied. In this case, it is known that Holt-Winters Multiplicative has the lowest error value. Proposed safety stock that can be applied by the company adjusted with the desired level of service, the number of safety stocks proposed in this study is available from the service level of 90% and 95%. Keywords: forecasting; time series; error value; safety stock