A probability chart for statistical process analysis, with applications

This paper reports an application of the Kolmogorov-Smirnov test for the purpose of detecting changes in the distribution of a sequence of measurements. The ʹprobability chartʹ of the title gave a value reflecting whether the distribution was constant in the neighbourhood of each observation. The chart had the advantage that the threshold for detection of a change was a dimensionless probability value. The non-parametric nature of the test made it suitable for measurements sampled from a non-Gaussian distribution.