Abstract
When 100% inspection in the order of production is in progress, an alternative approach to the p-chart or the Poisson-based CUSUM chart is to monitor the lengths of runs of conforming items between successive nonconforming items. Two Shewhart-type charts and a CUSUM chart based on such run lengths are presented and comparisons are made with the performance of the p-chart and the Poisson CUSUM chart in detecting shifts in fraction nonconforming. The run-length CUSUM chart is found to be more efficient than the Poisson CUSUM chart, while the run-length Shewhart chart based on the sum of the two most recent run lengths (termed the RL2 chart) is more efficient than the p-chart for detecting large shifts in fraction nonconforming.
Additional information
Notes on contributors
Patrick D. Bourke
Dr. Bourke is a Lecturer in the Department of Statistics. He is a Member of ASQC.