Data Mining In Six Sigma Quality Management Applications

As a manufacturer of general-purpose automation components and systems,the customer industry is widely distributed.If traditional quality management theories and methods are used to manage product quality,it will not be able to meet the growing special needs of customers in different industries.With the continuous participation of domestic enterprises in the competition of high-end products in the world,the international market has put forward higher and higher requirements for the quality,cost and development cycle of products.This thesis integrates and constructs the application method of data mining in Six Sigma quality management,and carries out empirical application in the improvement of product quality management.Based on Six Sigma research methods and systems,this paper establishes a corresponding recognition model based on advanced statistical techniques and tools such as principal component analysis and Adaptive-Lasso variable selection method for key quality indicators of customer demand.Neural network technology and logistic regression have been used for data mining and analysis of the relationship between manufacturers' quality indicators and customer quality indicators.This paper also utilized grey prediction theory and ARIMA time series method to establish a product performance and quality characteristics change prediction model.It also demonstrates the application of EWMA and MEWMA control charts in enterprises.The focus of this paper is to use a variety of data model combinations to improve product quality,which has been well applied in the actual case of the enterprise.Finally,combined with project management theory,organizational reengineering and Six Sigma related technical tools,the overall theory and method have been comprehensively applied in some products of the enterprise,and have achieved relatively good results.This management method and ideas can also be promoted and applied to multiple companies on the road to quality improvement.