Research on accurate prediction of vegetable commodity pricing strategy and dynamic replenishment quantity based on optimization model

Based on the characteristics of timeliness, fast price change and easy loss, this paper deeply studies the automatic pricing and replenishment strategy of fresh suppliers, in order to maximize the profit of the supermarket. We propose and solve the number of related problems, and fit the linear and nonlinear models, and the results have limited effect. Therefore, this study translated the replenishment problem into sales volume prediction, using LSTM and predicting future sales by random forest model. At the same time, based on the periodicity of vegetable commodity sales, this study proposes the single product replenishment volume and pricing strategy. Finally, in view of the poor fitting effect of a single factor in this study, the improvement  method  of  collecting  data  from  multiple  dimensions  is  proposed  to  deeply  explore  the relationship  between  replenishment  and  pricing,  so  as  to  improve  the  accuracy  of  prediction   and supermarket revenue.