Judging from the stoicism with which many manufacturing companies endure X-bar/R charting, you might think the technique was handed down to us by Moses. Regardless of whether X-bar/R provides too much or not enough control for our particular processes, we carry on while suppressing our doubts of this method's applicability, even when the expense of control charting fails to bring about any measurable quality improvement or prevention of quality losses.
Sure, X-bar/R charting was invented by a very bright man (Walter Shewhart ),), and it works well for some processes. But as R.W. Hoyer and W.C. Ellis point out in their article "A Graphical Exploration of SPC" (Quality Progress, May 1996), quality technology hasn't changed, in principle, since 1930. Problems tackled by manufacturing since the 1930s have inspired numerous new manufacturing processes. These processes, in turn, should give rise to new process control techniques. And they have everywhere but in quality assurance.
Why has quality technology seen so few new techniques integrated into standard practice? Has absolutely no new thinking occurred in this field? On the contrary; quality engineering and statistics magazines regularly publish new research results. But very little of this new research makes it into mainstream use.
The explanation for such scarcity lies in the politics and economics of quality assurance. Quality control can claim the dubious distinction of being, possibly, the only technical discipline in the United States today where technical issues are decided on the basis of political correctness, and where political considerations supersede engineering reasoning.
What forces this political correctness? Quite simply, the weight of quality assurance bureaucracies in major manufacturing companies and the comfort of existing systems in professional quality circles. Also, the big-business consequences (read: canceled contracts) of deviating from SPC methods dictated by valuable customers are enough to silence even the most forward-thinking professional. Would you risk the wrath of an auditor surprised by a chart he or she doesn't understand when your revenue is on the line? Creating SPC charts, regardless of whether they contribute to product quality, has become a cost of doing business.
Having largely ignored W. Edwards Deming's teachings on the applications of statistical methods for quality control during the manufacturing heydays of the 1950s, U.S. manufacturers later took to SPC with the fanatical zeal of new converts. Although we learned the importance of quality, we overlooked the root cause of our original mind-set. While we focused on whether to use SPC, whether to use X-bar/R or another technique, we failed to grasp that our openness to new ideas and the resulting changes will make today's best practices obsolete. Rather than adapting the principles of Deming's teachings, we merely substituted staunch orthodoxy for staunch ignorance.
A number of new SQC methodologies, ready and waiting for mainstream use, can control not just processes but also product quality. Techniques exist to control nonremovable special causes (for example, process drift due to tool wear). Cross-pollination with other, nonquality process control methods -- such as heuristic process control -- would result in a wealth of new quality control ideas. Our only criteria for adopting new methodologies should be scientific soundness and the ability to economically improve product, reduce costs or increase productivity.
We can earn a much better return on investment in SPC charting if we take an honest and critical look at the effectiveness of methods currently in use. SPC ought to stand for statistical process control in its broadest, modern sense, and not for "statistical political correctness." The new tools are out there. The question is: Do we have the will to break out of the status quo, or will we wait for a foreign competitor to teach us the same old lesson one more time?
About the authors
David Udler is the principal of Altegra's SPC training and consulting practice. He is an ASQ-certified quality and reliability engineer.
Alex Zaks, the founder of Altegra, leads the company's Quality Information Systems Group, which provides customized QA/SPC software, data-collection hardware and system integration services.
The authors can be reached by telephone at (610) 539-4224 or by fax at (610) 539-4225.
© 1997 Altegra. All rights reserved.