New Univariate Synthetic Double Sampling X And Multivariate Synthetic Double Sampling T2 Control Charts

Simple item page
dc.contributor.authorLee, How Chinh
dc.date.accessioned2019-08-01T02:36:55Z
dc.date.available2019-08-01T02:36:55Z
dc.date.issued2017-02
dc.description.abstractThis thesis proposes a univariate synthetic double sampling X control chart to monitor the mean of a univariate process and a multivariate synthetic double sampling 2 T control chart for monitoring the mean vector of a multivariate process. The proposed univariate chart integrates the conforming run length (CRL) chart and the double sampling (DS) X chart, while the proposed multivariate chart combines the CRL chart and the DS 2 T chart. Both the proposed univariate and multivariate charts are superior to their basic counterparts, namely the synthetic X and DS X charts for the univariate case, and the synthetic 2 T and DS 2 T charts for the multivariate case, in terms of the average run length (ARL) and average number of observations to signal (ANOS) performance criteria. ANOS is a more meaningful performance criterion compared to the ARL for the variable sample size charts, such as the proposed charts and the DS chart. This is because ANOS measures the actual number of observations required before an out-of-control signal is issued by the chart. Interpretation of a chart's performance based on ARL is complicated when a decision about a chart's performance is made based on either the first sample only or both the first and second samples. The proposed synthetic DS X chart is shown to outperform its basic counterparts, i.e. the synthetic X and DS X charts. By comparing the synthetic DS X chart with the exponentially weighted moving average (EWMA) chart, the former provides a better performance than the latter, for a larger range of mean shift sizes, compared with the synthetic X and DS X charts. The synthetic DS X chart has a better performance than the EWMA chart for detecting moderate and large mean shifts, while the latter overtakes the former for detecting small shifts.en_US
dc.identifier.urihttp://hdl.handle.net/123456789/8549
dc.language.isoenen_US
dc.publisherUniversiti Sains Malaysiaen_US
dc.subjectNew univariate synthetic double sampling xen_US
dc.subjectmultivariate synthetic double sampling T2 control chartsen_US
dc.titleNew Univariate Synthetic Double Sampling X And Multivariate Synthetic Double Sampling T2 Control Chartsen_US
dc.typeThesisen_US
Files
License bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
license.txt
Size:
1.71 KB
Format:
Item-specific license agreed upon to submission
Description:
Download
Collections
Pusat Pengajian Sains Matematik - Tesis