Abstract
Several control charting schemes have been developed for monitoring the stability of manufacturing processes. The most popular of these charts are the Shewhart, cumulative sum, and exponentially weighted moving average charts. To compare these schemes, we look at the distribution of the run length, that is, the length of time it takes to detect a change in the process. Simulation has frequently been used to analyze the run length distribution of a control chart when other methods were not available. Two methods other than simulation that have proved useful are the Markov chain approach and the integral equation approach. In this article we review how simulation, Markov chains, and integral equations are used to analyze the run length distribution of quality control charts. Further, we will compare and contrast the three approaches. We would recommend the integral equation approach first and next the Markov chain approach. Simulation should be used for checking the results of the first two approaches or when neither approach can readily be used.
| Original language | English |
|---|---|
| Pages | 617-622 |
| Number of pages | 6 |
| State | Published - 1990 |
| Event | Proceedings of the Twenty-First Annual Pittsburgh Conference Part 4 (of 5) - Pittsburgh, PA, USA Duration: May 3 1990 → May 4 1990 |
Conference
| Conference | Proceedings of the Twenty-First Annual Pittsburgh Conference Part 4 (of 5) |
|---|---|
| City | Pittsburgh, PA, USA |
| Period | 05/3/90 → 05/4/90 |