Method and system of predicting dynamic commodity sales

Abstract

The invention relates to a method and a system of predicting dynamic commodity sales. According to the method, the commodity sales are dynamically predicted on the basis of a gray model (GM) and a catastrophe model; a large amount of historical data are not required, and the problem of insufficient data of small / medium-sized enterprises is solved; and very high adaptability is achieved for fluctuating data, time at which outliers occur can be effectively predicted, and valuable data reference is provided for enterprise marketing decision. The method includes: collecting commodity sales data, and preprocessing the commodity sales data, wherein preprocessing includes detecting catastrophe data in the commodity sales data, and eliminating the catastrophe data; completing vacancy numerical-values, which correspond to the eliminated catastrophe data in the commodity sales data, to generate a commodity sales data sequence; and using the commodity sales data sequence to construct the unbiasedgray prediction model, and using the unbiased gray prediction model for commodity sales prediction.

Description

technical field [0001] The invention relates to a method and system for predicting the dynamic sales volume of commodities. Background technique [0002] As an important link in the marketing decision-making and inventory management of e-commerce retail enterprises, sales forecast has always been a hot issue that people pay attention to. High-quality, high-precision sales forecasts provide reliable data support for enterprise decision-making, thereby improving the economic benefits of enterprises. However, the environment of the e-commerce retail market is complex and changeable. Factors such as one after another promotional activities, shifts in consumer preferences, and seasonal cycle changes in commodities lead to the diversity and variability of commodity sales characteristics, which makes the forecast of market demand more and more difficult for enterprises. more difficult. In order to ensure customer demand and maintain stable sales volume, many retail companies have...

AI Technical Summary

Problems solved by technology

[0007] In the process of realizing the present invention, the inventor found that there are at least the following problems in the prior art: for the weighted average forecasting method, the calculation model of the forecasting method is relatively simple, the weight is not easy to determine, and there is no fixed rule to follow, and the calculation model of this forecasting method is relatively simple, and there is no fixed rule to follow. It is too strong, and it is difficult to deal with the sales data with mutation characteristics, so it is difficult to meet the requirements in accuracy; for the ARIMA model prediction method, the prediction method has a better fitting effect on the data with regular characteristics

Therefore, the use of a single static forecasting method will inevitably lead to an increase in the forecast deviation, leading to product out-of-stock or product backlog

Images