Physical Sciences and Mathematics Commons^™

Detecting Trends And Patterns In Reliability Data Over Time Using Exponentially Weighted Moving-Averages, Harry F. Martz, Paul H. Kvam

Department of Math & Statistics Faculty Publications

A simple, easy-to-use graphical method is presented for use in determining if there is any statistically significant trend or pattern over time in an underlying Poisson event rate of occurrence or binomial failure on demand probability. The method is based on the combined use of both an exponentially weighted moving-average (EWMA) and a Shewhart chart. Two nuclear power plant examples are introduced and used to illustrate the method. The false alarm probability and power when using the combined procedure are also determined for both cases using Monte Carlo simulation. The results indicate that the combined procedure is quite effective in …