Resumen de A Weighted CUSUM Chart for Detecting Patterned Mean Shifts
Lianjie Shu, Wei Jiang, Kwok-Leung Tsui
In situations where the mean of the monitoring sequence has a dynamic shift pattern, conventional control charts may perform poorly because they do not consider the information known as forecast recovery that is contained in the patterned mean shift. A new weighted cumulative sum (WCUSUM) procedure is proposed for monitoring a sequence with a patterned mean shift that first estimates the dynamic mean of sequence, and then uses the estimates for weighting the incremental in the conventional CUSUM chart. Results show that the WCUSUM chart performs better than other charts for detecting small to moderate shifts when forecast recovery is present and performs competitively otherwise.
