| Adaptive control schemes have gained in importance over the last decade, due to their effectiveness in process monitoring. Very little work has been done in adaptation of control limit in a cumulative sum control chart for attribute data. Although it has been suggested that the CUSUM be restarted after the signal of an out of control limit, there is no procedure given to improve the control limits of processes with decreasing defects per million levels. The research follows the Time-Between-Event CUSUM chart proposed by Lucas. The design of the TBE CUSUM chart is discussed. The comparison of the TBE CUSUM control chart with other attribute charts is given to show the effectiveness of this chart at low DPM levels. The adaptation of the control parameters of the TBE CUSUM chart, based on the average run length is proposed. These average run lengths are calculated by converting the movement of the test statistic into a Markov Chain. The application and its results are illustrated using simulated data. |
