Optimization Decision Making of Vegetable Replenishment and Pricing Based on ARIMA Model and Greedy Strategy

Fresh products become the important part in people's daily life. At present, the market of fresh products in China reaches a scale of 5.3 trillion, and the number of weekly purchases of fresh products by ordinary residents reaches about 3 times. Due to their short freshness period, superstores face the challenge of daily replenishment pricing. In order to develop a reasonable replenishment and pricing plan, this paper determines the cost margin in the markup cost pricing method by formula conversion using four indicators: sales volume, wastage rate, cost unit price, and replenishment volume, and further derives a functional relationship between sales volume corresponding to the six categories and cost markup pricing. The ARIMA model is used to forecast the cost margin, cost unit price, and sales volume for the next seven days, and a greedy strategy is used to equate the value of the daily replenishment volume to the predicted value of the sales volume before the loss on that day in order to achieve the goal of reducing the cost and meeting the goal of maximizing the revenue per day with certainty of the sales volume to arrive at the optimal pricing decision and the daily replenishment volume for the next seven days. Finally, the relationship between sales volume and pricing was directly fitted using the least squares method to reformulate the pricing decision, while the model was tested and evaluated. The integrated predictive pricing decision model proposed in this model can be a better solution to the commodity pricing problems related to time series, like supply chain product management system and online vegetable commodity delivery platform. It can also be used in models that contain interactions between time-dimension and space-dimension variables, such as the prediction of traffic flow and people flow.