Research On Quality Control Of Enterprise Production Process Based On Nonparametric Method

Online control of the quality of an enterprise's production process is an effective way to improve the quality of an enterprise's products.Statistical process control(SPC)is an important theoretical basis for online control of the production process quality of an enterprise.In statistical process control theory,when the product quality characteristic data of an enterprise can be described by a functional relationship between a dependent variable and multiple independent variables,such quality characteristic data is called profile data.Controlling the quality characteristic data that shows the profile type in the production process of the enterprise is the research hotspot in enterprise product quality management.This paper focuses on the actual problems in the production process of the enterprise,and mainly studies the quality control of the production process of the enterprise when there is correlation in the profile data,the position of the measurement point in the profile changes,and the profile data has both parameter errors and shape errors.When the profile data describing product quality characteristics has an internal correlation,a density-based phase I profile control chart is established to identify the in-control profile set in the quality characteristic profile data of an enterprise's production process.For the profile data with internal correlation,it is modeled using a linear mixed model;then the initial in-control profile is selected based on the density,and the first stage profile control chart is established by~2 statistics and the iterative method.At the same time,the influence of the density parameter and the number of initial profiles on the proposed method is analyzed.Three typical phase I profile control methods were selected and compared with the proposed density-based method.Simulation results show that,compared with other methods,the density-based method can identify in-control profile sets more effectively.When the positions of measurement points in different profiles change and there are both parameter errors and shape errors,a non-parametric EWMA(F-P)joint control chart based on the difference metric is constructed.Aiming at the situation that the parametric error and shape error appear in the profile data at the same time,it is proposed to construct a statistical parameter based on line profile error that can describe the parameter error and shape error respectively;then establish a non-parametric EWMA(F-P)joint control chart based on the difference metric.Four methods based on difference measurement are selected and compared with the proposed method.The simulation results show that the non-parametric EWMA(F-P)joint control chart based on the difference metric can effectively overcome the problem that the parameters of the profile model are difficult to determine.At the same time,it can promptly alert when the shape deviation and parameter deviation occur simultaneously in the production process of the enterprise Small offsets are more sensitive.With the continuous advancement of production technology,enterprise process quality control technology based on non-parametric methods is becoming more and more popular in the quality control of enterprise manufacturing processes,and it is also increasingly concerned by relevant scientific researchers.Based on the non-parametric method,this paper studies the actual problems that occur in the production process of the enterprise,which can help the relevant quality management personnel of the enterprise to find the abnormalities in the production process of the enterprise in time,so as to take the necessary measures to eliminate the abnormalities,improve the quality of the enterprise's products,and reduce The loss caused by the product quality of the enterprise will ultimately help the enterprise achieve better development.