Research On Multivariate Process Capability For Multi-variety And Small Batch Production Modes

With the development of productivity and the improvement of living standards,driven by the customer-centered and market-oriented business philosophy,the production mode of traditional manufacturing has undergone a huge transformation,In the emerging manufacturing industry,multi-variety and small-batch production mode has become the mainstream mode.However,the traditional quality control theory is mostly used in mass production mode.It is difficult for the traditional quality control theory to work effectively when analyzing the process capability of multi-product and small-batch production mode with multi-process,multi-index and small sample size characteristics.Therefore,it is very important to carry out theoretical research on multi-process capability analysis for multi-variety and small-batch production models.In this paper,the product quality control problem of an engine shaft is taken as the research object,and the multi-process capability analysis for multi-variety small-batch production mode is carried out.The specific research contents are as follows:(1)The problem of fuzzy similarity between processes is studied.For the quality control of multi-variety and small-volume products,it is necessary to solve the problem of insufficient sample data,and the process similarity analysis has obvious advantages in solving this problem.Therefore,this paper uses the fuzzy matter-element theory to determine the process similar to the original process,and to establish similar parts family,in order to expand the sample size.Then the original data of different dimensions are transformed into standardized data of the same dimension by tolerance coefficient method.Normal distribution test and mean-variance consistency test are carried out for standardized data.(2)The weight calculation of key mass characteristics is studied.The objective weight reflects the relationship between the measured data and the ability of the indicators to influence the process state,while the subjective weight uses the experience of experts and the analysis of historical data to evaluate the importance of the indicators.(3)Study how to measure the multivariate process capability index and its meaning interpretation.Principal Component Analysis(PCA)reduces the dimension of the original process to obtain the key principal components,and then calculates the multi-process capability index of regional volume ratio.It can well interpret the meaning of process capability index.In addition,this paper also proposes a multivariate process capability index based on combined weights,and analyzes the advantages and disadvantages of the two methods.(4)Study the establishment of multi-quality control chart under multi-variety and small-batch production mode.Using the similar process identified by fuzzy matter-element theory,construct a"typical process"with large sample size,and establish a multi T~2 control chart,multi-variable EWMA control chart for"typical process",and combine with the mobile range chart analysis Process state,identification process variation.