The study of a forecasting sales model for fresh food

Convenience stores (CVS) are an integral part of the retail industry. Merchandises are circulated from suppliers and CVSs to consumers. Therefore, if the goods which consumers desire are always out of stock, no matter how good the service, customer satisfaction will be hard to be improved. The point of sale (POS) system provides the information analysis ability and can be used to analyze consumers’ purchasing behavior as well as forecast needs. Therefore, managers of CVSs improve their revenues based on these data. No matter whether in the urban area, suburb or mountainous area, own the different characteristics of business circles. Therefore, business circles are important influencing factors. As Franchise Chains grow new operators increasingly join the business. However, it is difficult to train ordering operators in the short time. Therefore, some CVSs run out of stock and some may always order too large quantities. Only control each order precisely can meet customers’ need. Therefore, how to control the order and stock of CVSs has become one of the important issues in the management of CVSs. The purpose of this research is to discuss and develop a mechanism for controlling the order and managing the stock for CVSs. The theory of the buyer and seller exchanging system proposed by Kotler was modified to include the theory of consolidating loop structure. The “Ordinary day and holiday moving average method” and “back-propagation neural network” were proposed and tested based on the operating characteristics of business circle and sale forecasting. The method can be used to improve the ordering and discarding rate for achieving the goal of ordering the right items in the right amount.