STATISTICAL INTERQUARTILE RANGE-BASED COMMODITY INVENTORY RISK EARLY WARNING METHOD AND SYSTEM, AND COMPUTER READABLE STORAGE MEDIUM

摘要

A statistical interquartile range-based commodity inventory risk early warning method and system, and a computer readable storage medium comprising the method. The method comprises: acquiring original commodity inventory data of all stores in a certain historical time period; calculating, according to the original commodity inventory data, to obtain inventory incremental data; calculating upper and lower quartiles of the inventory incremental data, and calculating an interquartile range and an abnormality detection threshold according to the upper and lower quartiles; and detecting whether a new inventory increment exceeds the abnormality detection threshold, if so, determining same as abnormal data and pushing same to a front end for early warning. The method solves problems in the prior art, such as missing determination of an abnormal value, and low timeliness in the case of large calculation overheads and a large amount of data.