Journal of Quality Technology · 1997 · 225 citations · 11 references
A multivariate exponentially weighted moving average control chart improves detection of small shifts in multivariate statistical process control. The paper provides recommendations for selecting parameters of a multivariate EWMA control chart. The authors derive recommendations from Markov chain analyses of performance distributions and average run lengths across zero‑state, steady‑state, and worst‑state scenarios.
AbstractA multivariate exponentially weighted moving average control chart can be used to improve the detection of small shifts in multivariate statistical process control. Recommendations are provided for the selection of parameters for such a chart. The recommendations are based on performance distributions and average run lengths for zero-state, steady-state, and worst-state cases that are obtained from a Markov chain analysis of the scheme.Average Run LengthChi-Squared Control ChartsExponentially Weighted Moving Average Control ChartsMultivariate Control ChartsStatistical Process Control Additional informationNotes on contributorsSharad S. PrabhuDr. Runger is an Associate Professor in the Department of Industrial and Management Systems Engineering. He is a Member of ASQC.George C. RungerDr. Prabhu is a Member of the Technical Staff in the Statistical Quality Research and Development Department. He is a Member of ASQC.
11
E. S. Page · Biometrika · 1954 · 4.5K citations
Reliability Engineering, Engineering, Quality Inspection +11
Control Chart Tests Based on Geometric Moving Averages
S. W. Roberts · Technometrics · 1959 · 1.4K citations
Techniques of Statistical Analysis.
Edwin G. Olds, Statistical Research Group, Churc̀hill Eisenhart et al. · American Mathematical Monthly · 1950 · 834 citations