|
|
|
|
Six Sigma Quality and the extensive training provided in creating Black Belts and Green Belts have created unparalleled opportunities for exposing incredible numbers of engineers, managers, and manufacturing and service company employees and others to statistical methods, especially design of experiments. The teaching of Six Sigma has evolved considerably from our early efforts to expose large numbers of people to statistical process control and total quality management methods. In the early 1980s in Bell Laboratories we found engineers gave our design of experiments courses high ratings, but few bottom-line results came from these sessions. The DOE courses were redesigned to focus on creating actual experiments and meaningful results. Almost all Six Sigma courses are taught this way today. Well-planned projects are mandatory. Software supports every class. Class sessions are split to allow time for application of methods learned to real problems between sessions. Support is given to participants on their project both during classes and between class sessions. Our teaching of DOE has improved greatly, but there remain many challenges. Where should our focus be with such a wealth of material to cover in four or five days - or even less? How best to "unteach" what so many engineers have learned in labs where one factor at a time experiments are the norm? How much time should be spent in learning the basics of DOE versus learning by doing in labs and on projects? And most important, how do we expand the application of DOE to areas where it has rarely been used but where there is great promise?
|
|
|
|
|
|
|
|
|
|
|
|
|