Research On Supermarket Management Strategies Based on Statistical Inference and Genetic Algorithm Technology

Abstract

As one of the indispensable foods in people's daily life, reasonable pricing and replenishment decisions of vegetables are crucial for merchants to achieve the business objectives of lower cost and profit maximisation. Based on this, this paper adopts the commodity information and water flow detail data of a supermarket in Qingdao City from 2020 to 2023 to study the optimal pricing strategy of merchants. The research results show that: through correlation analysis and K-means clustering analysis, there is a certain connection between the six categories of vegetables and 251 single products; through HP filter analysis and visual observation, the sales of vegetable categories and single product sales show a cyclical pattern of change in time; from the seven categories of vegetables obtained from the clustering of each of the seven categories of vegetables, a single product is selected, and through the use of logarithmic nonlinear regression model, it is found that with the increase of the cost of cost plus Pricing increases, the total number of sales decreased logarithmically; finally, through the ARIMA time series prediction model, the sales volume of the next seven days to make a prediction, and then through the genetic algorithm optimisation model, to give the superstore's optimal pricing strategy for the next week. In this paper, through a more scientific and systematic mathematical method to study and analyse the vegetable market, a dynamic and flexible pricing of vegetable commodities as well as replenishment strategies are formulated, which will help the business to continue to revitalize and achieve higher profits.