JLS

Prediction Interval is used to predict the next single response where Confidence Interval is used to predict the average of multiple responses. Due to this the Prediction Interval will have a larger range than the Confidence Interval.

When using historical data, the inputs are not controlled but merely observed... this allows the identification of correlation and the ability to make inferences and predictions but lacks the ability to identify true cause-and-effect relationships. When using experimental data, the inputs are controlled... this allows the identification of true cause-and-effect relationships and optimization of the output(s) of interest.

A Sigma Level (Z-score) can be determined regardless of data type (continuous / discrete) where other process performance measures do not have this ability.

Prediction of warranty costs

When qualitative data is used for making decisions.

When the cost of changing X outweighs the potential benefits gained in Y.