Abstract

Ppm, Ppk, and Pp are performance measures that are more appropriate than capability measures when it is not possible to assume statistical control or when there are many process streams. Sampling for performance measures should produce large data sets, with a minimum of 1000 data values collected over at least one month. Ppm equals (USL-LSL)/(6((s2+(n/(n-1))(X-bar - T)2)); Ppk equals (min(USL - X-bar, X-bar, LSL))/3s; and Pp equals (USL-LSL)/6s, where: USL and LSL are the upper and lower specification limits; X is the observed value; X-bar is the sample mean; T is the target; s2 is the sample variance; and n is the number of observations. Additional measures include Pp(process stream) for processes with multiple process streams and Cp(potential), a measure of potential capability. Pp(process stream) equals (USL-LSL)/(6(swithin stream); and Cp(potential) equals (USL-LSL) divided by the product of 6 and the estimated potential variation. In Pp(process stream), swithin stream is the square root of the mean of the variances in the process stream.