A Process Capability Analysis Method Using Adjusted Modified Sample Entropy

The evolution of sensors and data storage possibilities has created possibilities for more precise data collection in processes. However, process capability analysis has become more difficult. Traditional methods, such as process capability ratios, cannot handle large volumes of process data over time because these methods assume normal process distribution that is not changing. Entropy methods have been proposed for process capability studies because entropy is not dependent on distribution and can therefore provide accurate readings in changing distribution environments. The goal of this paper is to explore the use of entropy-based methods, specifically modified Sample Entropy to identify process variations over time. A study based on simulated data sets showed that the proposed method provides process capability information.