Forecasting Technique for the Production Plan of the Local Earthenware Industries of Thailand

The purpose of this work is to compare the forecasting of time series models between two different models. One is the classical model and another is the Box-Jenkins model. The data are calculated using the circulation of Angbuaand Ahongwhich are the local earthenware products from Ratchaburi province, Thailand. Results show that the mean absolute percentage error (MAPE) of Angbua and Ahong are 17.80, 36.12 and 16.38,17.21 respectively. Also,prediction using the Box-Jenkins Model by ARIMA form of both products are (1, 0, 0) and (1, 1, 1). From this work the Box-Jenkins Model shows more appropriate method than the classical model considered by the less error.