Analysis and prediction model based on ARMI-GARCH and single objective optimization

What is it about?

This model first analyzes the distribution law of the sales volume of each vegetable category and single item. Since there are many single varieties of each vegetable category, the Contingency table analysis model is considered, and the cross-table of the sales volume and month of each vegetable category and single item is established respectively, and the test of independence is conducted. Then, the relationship between the total sales volume and the cost plus pricing of each vegetable category is analyzed. The partial least squares fitting model is used to analyze the relationship between the average sales price and the total sales volume of different vegetable categories. The ARMI-GARCH model was used to predict the average daily wholesale price of each vegetable in the next week, and the error analysis was made with the data. Then we select 7 indexes, such as display volume, loss rate, selling price and profit, and explain these indexes and give the solution formula。 Finally, under the premise of satisfying the market demand for various vegetable commodities as much as possible, a single objective optimization model is established with the objective function of maximizing the profit obtained by the supermarket selling vegetable commodities. We use genetic algorithm to solve the model and get the replenishment quantity and pricing strategy of each commodity the next day. Finally, the reliability of the model is judged by sensitivity analysis..

Featured Image

Why is it important?

We analyze the distribution law of the sales volume of each vegetable category and single item. Since there are many single varieties of each vegetable category, the Contingency table analysis model is considered, and the cross-table of the sales volume and month of each vegetable category and single item is established respectively, and the test of independence is conducted.