Quality quiz (October 2017)—and September’s quiz winners!

Winners of last month’s quiz and a copy of Practical Tools for Continuous Improvement: Volume 1 are George Lukaschunis (Bluewater Thermal Solutions of Saint Marys, PA); Kim Riehman (KLR Group, Inc. of Schaumburg, IL); and Larry Litke (Grand River Rubber & Plastics of Ashtabula, OH). For this month’s quiz, and a chance to win a copy of Practical Tools for Continuous Improvement: Volume 1, submit your response by October 31.

Hartford Simsack’s newest charges, Perry Winkel and Rose E. Glassis, both quit their jobs last month, frustrated with the ambiguous training that Simsack had given. His mission had apparently been to demonstrate their limitation, for he taught them just enough about charting for them to be dangerous. Simsack’s boss, Ford D. River, did not evaluate their exit interviews, so—luckily for Simsack—he is still in the dark about his quality manager’s statistical prowess.

Now Simsack must be sure that the three new technicians hired to replace Winkel and Glassis are competent with the use of p-charts. Simsack loves p-charts, seeing their limitless possibilities, and looks forward to a chance to demonstrate his own grasp of this statistical concept. “All those formulas,” he mumbles in excitement, knowing that the confusion about p-charts lies in the complexity of these impressive formulas. He loves writing these formulas on the whiteboard in the training room.

Simsack explains that as long sample proportions lie between the upper and lower control limits, the process is in control, so there’s nothing to worry about. “Any questions?” he asks, but does not wait for a response.

The following week, Ben Hunten, one of the new technicians, decides to apply what he has learned to a process on one of the lines at Greer Grate & Gate. When he encountered a pattern that seemed puzzling, he took it to Simsack for help in determining whether the process was in control or not.

“All of the points are inside the control limits,” he observes, reminding Hunten of last week’s admonition, “so it’s obvious that your process is perfectly in control.”

Is Hartford Simsack’s response accurate?

A. Yes. After all, what are control limits for, if not to show when everything is in control?

B. No. The process is clearly not in control.

5 thoughts on “Quality quiz (October 2017)—and September’s quiz winners!

  1. Although the readings show in control, it would be wise to review the samples to see if there is an assignable cause to the highs and lows – could be operator related or a hard variable in the process

  2. It is of course unwise to make the declaration that as long as the points are between the control limits, the process is stable (aka in control). More investigation into the generation of the defective proportions should be undertaken to identify the extreme differences in what appears to be two processes (instability), followed by testing of the plausible findings of that investigation. Then of course, the UCL and LCL are transposed on the control plot as shown.

  3. While the current data is within the control limits there should be concern for how close those points are coming to the limits. If as a company you are taking the time to track this and chart it you should also do something with the data. An investigation as to why the data is all over the place should be conducted. I would also say that the process in NOT in control seeing how a few of the points are at the max and some at the min. There is no consistency, and without a proper investigation one can almost assume that this will soon enough be outside of the tolerances given.

Comments are closed.