Demand forecasting for a food condiment manufacturer in Thailand

Abstract

The food industry is one of the most important industries in Thailand. The case study company is a condiment manufacturer that needs to manage and plan for their business efficiently. One of the most important issues is demand forecasting. The company should precisely forecast their product demands, which will be used for operation planning. This study proposes forecasting models for both short-term and long-term planning for three main condiment products. The data used are monthly condiment product demands from January 2013 to December 2020. The forecasting methods explored in the study are time series, machine learning, combined forecasting, and hybrid forecasting models. The accuracy of those models are measured by mean absolute percentage error (MAPE) where the results are also compared to the method currently used in the case-study company. The results show that artificial neural network (ANN) model provides the lowest overall MAPE for both short-term and long-term forecast. ANN model’s MAPE from short-term forecast is 4.44, while MAPE from long-term forecast (12 months in advance) is 4.64. When comparing with the company’s existing MAPE of 18.97, the proposed model can increase forecast accuracy effectively.