Abstract:
The sales forecast is a crucial step in supply chain management as it enables organizations to mitigate the risks and uncertainties arising from customer demand. The purpose of this research is to compare novel approaches of sales forecasting which is a combination method that includes Artificial Neural Networks and Adaptive Neuro Fuzzy Inference Systems based on quantitative and qualitative input data with a single method. Data were gathered from a leading retail company in Thailand, comprising both quantitative and qualitative data. Qualitative factors were collected through literature review and selected by experts. Then, single methods are proceeded by Artificial Neural Network and Adaptive Fuzzy Network Inference System. The optimal model is then selected from the single method to conduct a combination forecast using the Simple Mean method, Mean Square Error Weights method, Least Square Weights method. The forecast results showed that the lowest forecast error is the proposed combination method with Least Square Weights based on quantitative and qualitative input data