such processes is to define the source of the data as a distribution where one or more of its parameters are a function of time, sequence, or other system state. Consider an activity for the installation of valves, where the time required for each of the first 200 installations was collected. The histogram of the data along with the distribution fitted to it assuming that the data are IID by Stat::Fit Benneyan 1998 are shown in Fig. 5. Since the fit is very good the ChiSquare, Kolmogorov-Smirnov, and Anderson-Darling tests do not reject the hypothesis at the 0.15 level of significance, it may seem a good idea to use the fitted distribution in a model. However, further inspection by using a sequentially ordered plot Fig. 6, a scatter diagram Fig. 7, and a correlation plot