What is the proper method to use for calculating Cpk for a one-sided spec using variable data? For example, one-sided spec parameters such as minimum tensile strength or maximum roundness TIR? The standard Cpk calculation & analysis requires a bi-lateral tolerance. The distribution for this type of variable is often non- normal and skewed out from the min or max spec. But the out-liers are usually unimportant. We use Mini-Tab. Any recommendations?