Adaptive Control Charts For Monitoring The Univariate And Multivariate Coefficient Of Variation

• Main Author: Khaw, Khai Wah
• Format: Thesis
• Language: English
• Published: 2017
• Subjects: QA1 Mathematics (General)
• Online Access: http://eprints.usm.my/47524/1/KHAW%20KHAI%20WAH.pdf%20cut.pdf; http://eprints.usm.my/47524/
• Institution: Universiti Sains Malaysia

The first objective of this research is to propose a univariate adaptive CV chart using the variable sample size and sampling interval (VSSI) approach, called the VSSI CV chart, to monitor the process CV. The VSSI CV chart will be optimally designed, where two parameters, namely the sample size and sampling intervals are allowed to vary. In real life scenarios, there are many situations in which a simultaneous monitoring of two or more correlated quality characteristics is necessary. Erroneous conclusions will occur if quality practitioners use univariate control charts to monitor a multivariate process. The second objective of this study is to propose CV charts for monitoring the multivariate process CV (MCV) by adopting the adaptive procedures. Three new charts for monitoring the MCV, namely the variable sampling interval MCV (by varying the sampling interval), variable sample size MCV (by varying the sample size) and VSSI MCV (by varying both the sample size and sampling interval) charts are proposed to improve the performance of the existing MCV chart. All the charts proposed for monitoring the univariate and multivariate CVs are designed using the Markov chain approach. The implementation procedures and optimization designs of these proposed charts are enumerated in this xxiv thesis.