THE DEMAND FORECASTING FOR OF CONTRACTOR BADGE PRINTING

Abstract

This paper objective is the pilot project to find the optimize method for inventory forecasting and to improve inventory of contractor badge printing. The forecasting methods are Moving Average, Holt’s Exponential Smoothing Method, Winters’ Additive, Winters’ Multiplicative, Double Exponential Smoothing and Decomposition. The data collections are from two hard disk drive manufacturers and are forecast for the next three months of contractor badge inventory. We select the forecasting methods by measuring the forecasting errors and accuracy from MAD, MSE, MAE and MAPE. The results found that the best forecasting method for the contractor badge inventory for the next three months is exponential smoothing, which has the minimum average error, MAE = 35.43, MAD = 0.91, MSE = 7.94 and MAPE = 1.28% for forecasting the next three months. In conclusion, the result of contractor badge forecasting is helpful for preparing the appropriate number of raw materials for the contractor badge.