Effect of Measurement Errors on the Multivariate CUSUM CoDa Control Chart for the Manufacturing Process (2201.12318v2)

Abstract: Control charts, one of the main tools in Statistical Process Control (SPC), have been widely adopted in manufacturing sectors as an effective strategy for malfunction detection throughout the previous decades. Measurement errors (M.E's) are involved in the quality characteristic of interest. The authors explored the impact of a linear covariate error model on the multivariate cumulative sum (CUSUM) control charts for a specific kind of data known as compositional data(CoDa). The average run length ARL is used to assess the performance of the proposed chart. The results indicate that M.E's significantly affects the multivariate CUSUM-CoDa control charts. The authors have used the Markov chain method to study the impact of different involved parameters using four different cases for the variance-covariance matrix (i.e. uncorrelated with equal variances, negatively correlated with equal variances, uncorrelated with unequal variances, positively correlated with unequal variances). The authors concluded that the ARL of the multivariate CUSUM-CoDa chart increase with an increase in the value of error variance-covariance matrix, while the ARL decreases with an increase in the subgroup size m or the constant powering b. For the implementation of the proposal, two illustrated examples have been reported for multivariate CUSUM-CoDa control charts in the presence of M.E's. One deals with the manufacturing process of uncoated aspirin tablets, and the other is based on monitoring machines in the muesli manufacturing process.