基于分数阶灰色模型的生鲜电商产品销量预测研究

新型冠状病毒的爆发,使供应链受到了不良影响,蔬菜水果等生鲜商品线下批发零售受阻,生鲜电商却迎来了新一轮的发展.从生鲜消费者角度出发,通过构建分数阶灰色预测模型对生鲜产品的需求量进行预测,并引入甘福园生鲜电商销售数据进行验证,得到柠檬、车厘子、苹果及火龙果4种水果拟合值的平均绝对百分比误差(MAPE)分别为8.61％、7.15％、7.16％、6.81％,其值均小于灰色预测模型和一次指数平滑法,且由该模型得到预测值的平均绝对百分比误差均小于10％,结果表明该模型适用于生鲜电商产品销量预测.

Research on Forecasting Sales of Fresh E-commerce Products Based on Fractional Order Grey Model

The outbreak of novel coronavirus has affected the supply chain badly. The offline wholesale and retail of fresh commodities, such as vegetables and fruits, have been severely held back. However, the online fresh produce business has embraced a new round of development. The main purpose of the research is to forecast the sales volume of fresh e-commerce through the fractional order grey prediction model so as to resolve the imbalance between supply and demand of fresh products during the epidemic. In this paper, a fractional order grey forecasting model was built to forecast the demand for fresh products, which was verified by introducing Ganfuyuan sales data from the perspective of fresh produce consumers. The mean absolute percentage error (MAPE) fitting values of four fruits (lemon, cherry, apple and dragon fruit) were 8.61%, 7.15%, 7.16% and 6.81% respectively. Those practical values were less than the predicted values using grey prediction model or single exponential smoothing (SES) method. The mean absolute percentage error of predicted values was less than 10% on average. The results showed that the model was suitable for e-commerce fresh produce suppliers to forecast the sales volume of fresh products.