Abstract:
Plastic is vital to Thailand's economy due to its widespread use in various merchandise industries. According to the statistics of plastic consumption in Thailand, the Polypropylene (PP) has the highest consumption quantity compared to other plastic types which makes it becomes the most important plastic to the plastic industry in Thailand. This research is conducted to present and compare the forecasting models of PP granule price in Thailand. The objective of this research is to determine the most accurate forecasting model of PP granule price by using mean absolute percentage error (MAPE) as a criterion to compare the efficiency of the forecasting models. The forecasting models in this research are Holt - Winters exponential smoothing model, SARIMA model, multiple regression model, Support Vector Regression Model, Decision Tree Model, Random Forest Model, XGBoost Model, Artificial Neural Networks Model and Hybrid Forecasting Model. This research uses the data of PP granule price from January 2011 to December 2022. The results show that SARIMA ANN Model have the highest accuracy with the value of MAPE equal to 5.54% at CV Data Set and when evaluating with the Testing Data Set, the MAPE is equal to 6.92%.