Page 393 - Six Sigma Demystified
P. 393

Part 3  S i x   S i g m a  To o l S        373

                           Interpretation

                           The run tests apply statistical tests to determine if there are any patterns or
                           trends in the plotted points. Run tests 1, 2, 3, 5, and 6 provide an indication
                           that the process mean has shifted, whereas the others tell us something about
                           sampling errors, inconsistent with the base premise of rational subgrouping.

                                Run test 4: An indication of sampling from a multistream process
                                (subgroups alternate between two or more process levels)
                                Run test 7: An indication of stratification in sampling (multistream
                                sampling within a subgroup)
                                Run test 8: An indication of sampling from a mixture (multistream
                                sampling, subgroups on each side of the center from separate
                                distributions)

                             The statistical basis of the run tests is simply that if the subgroups are truly
                           from the stated distribution and independent of one another, then there will
                           not be any pattern to the points. The run tests increase the power of the control
                           chart (the likelihood that shifts in the process are detected with each subgroup)
                           but also provide an increased (yet minimal) false-alarm rate.
                             Keep in mind that the subgroup that first violates the run test does not
                           always indicate when the process shift occurs. For example, when run test 2 is
                           violated, the shift may have occurred nine points (or more or less) prior to the
                           point that violated the run test. As another example, a process may be in con-
                           trol and not in violation of the run tests for a period of time, say, 50 subgroups.
                           Then the process average shifts upward. As more and more subgroups are
                           added at the new level, subgroups in the original 50 subgroups will start violat-
                           ing run tests or control limits because these points now show an unnatural

                           pattern relative to the combined distributions of the two process levels.


                    Scatter Diagrams


                           A scatter diagram is an XY plot used to investigate the correlation of one vari-
                           able with another.

                           When to Use

                           Analyze Stage

                             •	 To investigate the correlation of one variable with another
   388   389   390   391   392   393   394   395   396   397   398