Factory product one-time verification qualified rate prediction method and device and medium

Abstract

The invention provides a factory product one-time verification qualified rate prediction method and device, and a medium, and particularly relates to the technical field of quality and prediction, and the method comprises the steps: collecting factory production process historical data, and sending the data to a cloud server; the cloud server carries out training on the received factory production process historical data based on a prediction and early warning model of a long short-term memory (LSTM) network; judging whether the accuracy rate of the model is lower than a preset threshold value or not, and if yes, continuing to collect factory production process historical data for training; if not, obtaining a prediction and early warning model of a trained long short-term memory (LSTM) network; the edge computing device in the factory collects real-time field data related to the production process and sends the real-time field data to the cloud server; the cloud server receives the real-time field data related to the production process acquired by the edge computing device in the factory, and performs data fusion and storage; and the cloud server runs the trained prediction and early warning model of the long short-term memory network LSTM, and predicts the one-time delivery inspection qualified rate of the product.

Description

technical field [0001] The invention relates to the technical field of quality and prediction, in particular to a method, a device and a medium for predicting the pass rate of factory products in one-time verification. Background technique [0002] At present, the one-time inspection pass rate of products and their important components is an important indicator that factories are concerned about. The existing technology cannot predict the pass rate of a product and its important components in advance, and can only perform statistical analysis at a certain time. Once a problem occurs, it cannot be remedied in time, which will bring huge costs and economic losses. [0003] If it is possible to predict and pre-warn the pass rate of the finished product and its important components for one-time inspection, it will be of great help to the control of the production and manufacturing process of the product, and the occurrence of quality problems can also be predicted in advance, so...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the deficiencies of the prior art, the present invention provides a method and device for predicting the pass rate of factory product verification, which solves the problem that the prior art proposed in the above-mentioned background art cannot predict in advance the passing inspection of the product and its important components at one time. rate problem

Images