RESPONSE TO 'SHORT-RUN STATISTICAL PROCESS CONTROL: Q-CHART ENHANCEMENTS AND ALTERNATIVE METHODS'

摘要

Del Castillo and Montgomery have studied properties of the Shewhart-type Q-charts given by Quesenberry by simulating the mean, ARL, and standard deviation, SRL, of the distribution of run lengths for these charts, and some procedures they propose as possible alternatives. We show here that ARL and SRL are not useful criteria for comparing control procedures for the situations considered in those papers. Moreover, the simulation results reported are incorrect for some cases reported and suspect for all cases. We also point out a number of other errors and misleading unsupported assertions.