A demand forecasting system for retail industry based on neural network and VBA
Yuefang Gao, Yongsheng Liang, Fei Tang, Zhi-Wei Ou, Shaobin Zhan · 2010
To provide retailers with market and trend analysis, and lower inventory cost from large amounts of data accumulated in the sales process, this paper presents a neural-network-based demand forecasting system implemented in the VBA environment. Through the use of Excel built-in VBA, this demand forecasting system can easily handle the data exchange between the raw data tables, and can achieve forecasting process and results visualization according to users' requirements. Based on the neural network algorithm, this demand forecasting system does not depend on the accuracy of mathematical models, and its model parameters can be auto-adjusted according to the learning of the forecast errors. The experimental results show that the speed and accuracy of forecasts have been greatly improved through the use of this system.