Level: Intermediate José Ramírez, Chief Statistician, Amgen Jon Weisz, JMP Senior Vice President, SAS Shewhart charts are a direct plot of the data and are good at exposing different types of deviations from statistical stability. However, the Shewhart chart is not as sensitive to small changes in the process average; it is in this type of situation where we need a more sensitive chart. Cumulative sum (CUSUM) charts plot the cumulative differences between a process variable and a target, which makes them particularly sensitive to small deviations in the process mean from target. In this talk, we introduce the concepts behind CUSUM charts, and show examples of how easy it is to develop these charts using the new CUSUM Control Chart platform in JMP 14. This new platform emphasizes the use of decision limits rather than the traditional V-Mask, facilitating the development and interpretation of the charts. We will also discuss the use of cumulative score (CuScore) charts to detect changes in the parameters of a given model.
... View more
José G. Ramírez, PhD, Chief Statistician, Quality Excellence Group, Amgen
This paper was voted one of the three finalists for Best Contributed Paper.
After the analysis of a designed experiment with multiple responses, one may end up with a large number of reduced models that are a function of several experimental factors. How do we determine which of these models are good for prediction? How do we assess which experimental factors are the most important ones across all of the responses? The Box-Wetz criterion is a useful goodness-of-prediction measure for statistical models that compares the average change in a response to the average estimation error. The Sobol decomposition provides sensitivity indices that are very helpful in assessing the importance of the experimental factors across multiple responses. Using examples from the biotechnology industry, we will demo a JMP script that calculates the Box-Wetz criterion for a given model and discuss its interpretation. We will also show how the sensitivity indices generated by the “Prediction Profiler > Assess Variable Importance” can be used to prioritize the importance of experimental factors across multiple responses.
... View more