Optimal Single-Item Replenishment Scheme and Pricing Strategy for Vegetable Category Based on LSTM Time-Series Prediction Mixed Integer Programming Model

The replenishment and pricing of superstore vegetables is an important part of daily life, and the merchants' prediction and optimisation scheme for the future has gradually become an important issue in production and life. This paper focuses on the problem of total replenishment and pricing strategy of superstores, predicts and analyses the results based on LSTM time-series prediction, and finally gives the optimal total replenishment and pricing strategy of superstores in different scenarios. Firstly, the data are preprocessed to filter out the types of products available for sale from 24-30 June 2023 Meanwhile, based on the LSTM time-series prediction model, the basic relationship between short-term sales quantity and pricing is predicted. Then, under the premise of trying to meet the market demand for each category of vegetable goods, a mixed integer programming model aimed at maximising the benefits of the business dynasty is established, the 0-1 variable is introduced to constrain the selection of single products, and multiple constraints are established on the number of single products, the order quantity of single products, and the demand for each type of goods, which is solved to produce a profit of 1098.360 yuan on 1 July. Finally, based on a simple model perturbation, it is judged that the scheme is more reasonable and the result is superior, and finally gives the optimal single-item replenishment scheme and pricing strategy.