Improvement of the Statistical Process Control Certainty in an Automotive Manufacturing Unit

Abstract

To control a process means to make adjustments in order to improve the performance, identify and fix anomalies. The statistical process control (SPC) is a solution developed to easily collect and analyze data, allowing performance monitoring as well as achieving sustainable improvements in quality which in turn allows increasing the profitability. The SPC makes it possible to monitor the process, identifying special causes of variation and defining the corresponding corrective actions. The SPC enables the monitoring of the characteristics of interest, ensuring that they will remain within pre-established limits and indicating when corrective and improvement actions should be taken. The focus of this study is to analyze the SPC control chart of an industrial unit operating in the automotive industry. The normality test used at this manufacturing unit is Kolmogorov-Smirnov (K-S). This test shows that if the data follows a normal distribution then the SPC is valid. However, by increasing the accuracy of the normality test a starkly different result could be obtained. Thus, in this paper a comparison between two normality tests is made and the results and the consequences of the Anderson-Darling test are analyzed and discussed.