SPC for Autocorrelated Data using Automated Time Series Forecasting
Statistical process control for autocorrelated processes have been addressed using the EWMA (Exponentially Weighted Moving Average) one-step-ahead forecast or simple ARIMA (Auto-Regressive Integrated Moving Average) models. The time series model forecasts the motion in the mean and an individuals control chart is plotted of the residuals to detect assignable causes. Failure to account for the autocorrelation will produce limits that are too narrow, resulting in excessive false alarms, or limits that are too wide resulting in misses.
The challenge with this approach is that the user needs an advanced level of knowledge in forecasting methods to pick the correct model, especially when there is negative autocorrelation or seasonality in the data and ARIMA models are needed.
This session introduces recent developments in time series forecasting that use modern techniques with automatic model selection to accurately pick the simplest time series model that produce minimum forecast error.
An accurate forecast for your time series means the residuals will most often have the right properties to correctly apply a control chart, thus leading to an improved control chart with reduced false alarms and misses.