A Study on Using the Moving Average Chart for Monitoring Time-between-Events

Among attributes control charts, the count of nonconformities is often used to monitor the quality of a manufacturing process. When the non-conformity rate in the process is very low, traditional control charts for nonconformities become ineffective. In such a situation, the process control may resort to monitoring the time between successive non-conformities (referred to as the time-between-events, TBE). Control schemes for monitoring TBE can range from simple to complicated, depending on the purpose of the monitoring. Previous studies have proposed a Shewhart-type chart with a runs rule (＂t＂-runs chart) to enhance control performance. This research considers a simple moving average chart (＂MA＂ chart) for the surveillance of TBE. An approximate analytical approach to compute the performance of the ＂MA＂ chart is first derived for exponential TBE data. The ＂MA＂ chart is then compared to the ＂t＂-runs chart in terms of the average run length and the standard deviation of the run length. The comparison results show that the ＂MA＂ chart has the better performance than the ＂t＂-runs chart in most cases. Because the MA chart is simple to interpret and implement, it is a good alternative to the ＂t＂-runs chart. The analytical approach is extended to obtain the design parameters for the ＂MA＂ chart for monitoring Weibull and Gamma TBE. The design parameters for the ＂MA＂ chart are provided for practical implementation.