A Study on Merchandise Pricing and Replenishment Decision Based on Simulated Annealing Algorithm and Lasso Regression

This paper mainly focuses on the pricing strategy and purchase decision of vegetable commodities in the fresh supermarket. By using the historical data and data analysis method, the pricing strategy and purchase decision problem is transformed into a multidimensional and double-layer complex dynamic optimization problem, and the simulated annealing algorithm is used to solve it. Firstly, with profit rate and wholesale price as independent variables and total sales as dependent variables, a polynomial regression model is established for each category, and Lasso regression is used to optimize the model. Secondly, LSTM neural network model was used to get the daily replenishment of each category in the next week, and a single objective programming model was established with the profit margin as the decision variable and the profit maximization of the supermarket surplus as the objective function. The pricing strategy for vegetable categories was obtained by simulated annealing algorithm. Finally, considering the actual constraints such as the quantity and display quantity of supermarket items, a dynamic programming model was established to obtain the replenishment quantity and pricing strategy for vegetable items by using the simulated annealing algorithm. The research method and model in this paper provide a scientific basis for the supermarket to make the pricing strategy and purchase decision of vegetable commodities. The research combines the data science and technology with the operation of the supermarket, which provides an important reference for the supermarket to improve profits and reduce losses.