期刊名称：International Journal of Hybrid Information Technology
印刷版ISSN：1738-9968
出版年度：2015
卷号：8
期号：8
页码：233-244
DOI：10.14257/ijhit.2015.8.8.24
出版社：SERSC
摘要：In order to meet the increasing daily demands of customers and reduce the unnecessary cost in retail stores as far as possible, the inventory management of retail stores becoming more and more important. However, because of various characteristics of demand in retail stores, the traditional demand forecasting technologies can't work well. In this paper, we use the modified K-means clustering analysis to help determine the groups with different characteristics of demand. In addition, a demand forecasting model integrated BP neural networks and grey model is proposed to make the prediction more intelligent and general. The example illustrates that the proposed method for forecast is feasible by the comparative analysis between the predicted values and the actual values.
关键词：Demand forecasting; clustering analysis; retail stores; BP neural network; ; grey theory
