A very common problem when using traditional P and U control charts is false alarms due to overdispersion. David Laney devised P' and U' charts to overcome this problem.
False alarms can be caused by too much variation in process data, or “overdispersion,” especially if you collected your data using large subgroups. The larger your subgroups, the narrower your control limits will be on a traditional P or U chart. This situation creates artificially tight control limits, causing points on a traditional P chart to look like they’re out of control even when they are not.
Alternatively, too little variation, or “underdispersion,” can also be problematic. When underdispersion exists, control limits on a traditional P chart or U chart may be too wide, so data points (or “processes”) you should be concerned about appear to be in control.
Whether your data exhibit overdispersion or underdispersion, a P' or U' chart will help you distinguish true common cause versus special cause variation.