Research Of Process Quality Control And Diagnosis Based On Multi-variety And Small-batch Production Model

The improvement of Socioeconomic level and the increase of personal income have promoted the transformation of product market from seller's market to buyer's market,and consumers have more voice in product market.Consumers' demand for diversification and individualization of products makes enterprises must turn to the pro-duction mode of multi-varieties and small batches.In view of the variety of products and small production batches in multi-variety and small-batch production modes,the traditional process quality control method is no longer applicable.Therefore,it is urgent to find a process quality control method suitable for multi-variety and small batch pro-duction modes.In order to improve the process quality control ability and enhance the process quality stability,this paper starts with the characteristics of multi-variety and small-batch production mode,introduces T and K statistics based on the analysis and sum-mary of the characteristics of multi-variety and small-batch production modes.Then Establish a T-K control chart for process quality analysis.In addition,the combination of MC(Monte Carlo)method and BP(Back Propagation)artificial neural network is used to identify the pattern of T-K control chart.On the one hand,it overcomes the shortcomings of limited sample size in multi-variety and small-batch production mode,and on the other hand,improves the speed and accuracy of control pattern recognition.Finally,on the basis of analyzing the influencing factors of the process quality fluctua-tion source,the process quality is decomposed according to the sequence of the two processes under the guidance of the two quality diagnosis theories.Based on this,the three-diagram diagnosis system is established to diagnose the process quality fluctua-tion source.Finally,an example is given to validate the above methods.The main results of this paper are as follows:(1)By introducing T and K statistics and establishing T-K control chart,the limi-tation of traditional statistical process quality control caused by the particularity of multi-variety and small-batch production mode is solved,so that the process quality of multi-variety and small-batch production mode can be controlled by using statistical process control method.(2)Using BP artificial neural network technology to recognize the established T-K control chart pattern not only improves the speed of pattern recognition of control chart,but also improves the accuracy of pattern recognition,which makes the quality management personnel's judgment of process quality more accurate.After recognition,the process quality is generally in a controlled state.(3)Based on two kinds of quality diagnosis theories,the accuracy of process qual-ity fluctuation source diagnosis is improved by establishing a three-graph diagnosis system.At the same time,the diagnostic basis of process quality fluctuation source is oriented,and the operability of diagnostic method is enhanced.Through empirical anal-ysis,the main reason is that there is an abnormal trend in the current process,that is,the latter one.