Control charts for high-quality processes: MAX or CUMAX?

Willem Albers

    Research output: Contribution to journalArticleAcademicpeer-review

    2 Citations (Scopus)


    For attribute data with (very) small failure rates control charts were introduced which are based on subsequent groups of r failure times, for some r >= 1. Within this family, it was shown to be attractive to stop once the maximum of such a group is sufficiently small, because this choice allows a very satisfactory nonparametric adaptation. The question we address here is whether a cumulative approach offers even further improvement. Thus instead of fixed groups, we shall use the first sequence of r consecutive sufficiently small failure times to produce a signal. A further reason for considering this type of chart is the fact that it forms the nonparametric counterpart of the well-known sets method.
    Original languageEnglish
    Pages (from-to)1924-1932
    Number of pages14
    JournalJournal of statistical planning and inference
    Issue number7
    Publication statusPublished - 2012


    • MSC-62P10
    • Statistical process control
    • Order statistics
    • Average run length
    • Geometric charts
    • Health care monitoring
    • Estimated parameters
    • Sets method


    Dive into the research topics of 'Control charts for high-quality processes: MAX or CUMAX?'. Together they form a unique fingerprint.

    Cite this