A Neural Network Model for China B2C E-Commerce Sales Forecast Based on Promotional Factors and Historical Data

Abstract

Sales forecast is the basis and premise of supply chain management. It affects enterprise inventory level and supply chain operational efficiency. With the rapid development of online shopping, B2C e-commerce enterprises have an increasingly strong demand for accurate and efficient sales forecasting. Influenced by various promotional activities, B2C e-commerce sales in China are characterized by big fluctuations, fast pace and difficult prediction. The problem of sales forecast considering the influence of promotion is an important problem to be solved. This paper combines the characteristics of B2C e-commerce promotion in China and a quantitative model of the influence of promotion on B2C e-commerce sales is established. Based on promotion factors and historical data, we constructed a BP neural network model for B2C e-commerce sales forecast to solve the problem that e-commerce sales are difficult to predict accurately under promotional activities. Through the verification of Alibaba's actual case, it is concluded that the GA-BP algorithm model has a good adaptability to sales forecasting and it achieved 94 percent accuracy. The B2C e-commerce sales forecast model based on sales promotion and historical data established in this paper has high application value.