A Statistician’s Favorite Answer: ‘It Depends,’ Part 2

Stop getting sucked into the swamp of calculation minutiae

When teaching the I-chart, I’m barely done describing the technique (never mind teaching it) when, as if on cue, someone will ask, “When and how often should I recalculate my limits?” I’m at the point where this triggers an internal “fingernails on the blackboard” reaction. So, I smile and once again say, “It depends.” By the way…

… Wrong question!

I made a point in Part 1 of this article that I feel is so important, I’m going to make it again: Do not bog down in calculation minutiae. If you feel the instinct to ask that question, pause and think of how you would answer these from me instead:

1. Could you please show me the data (or describe an actual situation) that are making you ask me this question?

2. Please tell me why this situation is important.

3. Please show me a run chart of these data plotted over time.

4. What ultimate actions would you like to take with these data?

And since writing Part 1, I’ve thought of a fifth question I’d like to add:

…

Want to continue?

By logging in you agree to receive communication from Quality Digest. Privacy Policy.

Create a FREE account

Forgot My Password

Comments

Auditing a large cause

I agree wholeheartedly with the lunacy of recalculating limits for the sole purpose of recalculating limits. Davis, I think if you do some investigation, you’ll find that a major source of this comes through the “audit” process (that has been almost 100% of my experience). I have been involved with two different “statistically enlightened” companies for over 20 years. We never talked about frequency of recalculation of limits amongst ourselves as an opportunity for improvement, because we knew that was a foolish thing to do. But the audits that we were subjected to by our direct customers and their auditing bodies didn’t have this same sense of understanding of limits. The best was when purchasing agents, with a whopping 2 day course in SPC and no hands-on experience, would attempt to tell people with much more background and experience that limits must be recalculated every 3 months. Sigh…

Now, manufacturing companies aren’t blameless in creating this scenario. I have often seen where a set of limits were calculated many moons earlier and are no longer relevant, or opened up because too many alarms were being generated by improperly set limits (and we all know there are more reasons than these). So, the customer sees these examples and is thus dissatisfied with the state of control within their supplier (and who could blame them). As a result, the easiest way to audit this is to see the date of when the limits were last recalculated. Auditors love black and white – makes their job easier. That forces the supplier to take a look at it and never let it get “out of control” again. Hence, the frequency requirement and the drive for the answer to the question.

After my first round of corporate quality I got smarter in my second round. I wrote corporate procedures and policies that strictly stated NOT recalculating limits on a systematic basis. If a plant was audited and pushed on this, they would point to the corporate policy and corporate would get on the horn with the auditors. That generally didn’t save the plant from receiving points off of the audit (but at least it demonstrated they followed policy!!!!!!!!! ).

You want the plants to focus their limited resources on the right things that add value to the process, not in silliness that adds no value other than satisfying an uneducated auditor or manager who has been otherwise duped into believing that is part of process control.

Kevin Keller

The more things change...

...the more they remain the same.

Gents,

WONDERFUL insights,both of you and "spot on." Kevin, I saw the same things in my manufacturing days at 3M, whose profile I'm sure compares favorably to your "statistcally enlightened" companies. Is the goal of an audit to pass the audit or improve quality? I'm more firmly convinced than ever that quality must be built into a company's cultural DNA to the point where the words "statistical" and "quality" are dropped as qualifiers because they are "givens." Otherwise, the maddening games continue to try to outfox auditors, many of whom are as you described.

VERY few people need an advanced knowledge of statistics. It is only in the last 10 years that I have finally "got" it. As Deming himself said, "If I had to reduce my message to management to just a few words, I'd say it all has to do with understanding variation." He taught very few techniques in his seminars and he is probably rollling over in his grave at the sub-culture of "hacks" (his term) that has been created. Only 1-2% of people need ADVANCED statistical knowledge, and all these "belts" are shooting themselves in the foot with their training, which nothing short of legalized torture. I had access to some correspondence Deming wrote to a well-respected statistician...in 1984: "Sorry about your misunderstanding...TOTAL! When will statisticans wake up?" As David Kerridge says: If we're actually trying to do the wrong thing, we may only be saved from disaster because we're doing it badly." And managment like Steve describes above continues to be deluded as to its effectiveness. The more things change...

Questions...questions!

Davis, Your insight is appreciated. I get the same questions when I teach the use of Process Behavior Charts. I guess people are people wherever you go! For some reason people get too hung up on the computations rather than gaining insight into their processes. Also, whenever executives tell me that there is "too much uncertainty" in the predictive powers of a given Process Behavior Chart, I just show them a chart of variances between THEIR monthly forecasts and the actuals. Now THAT'S uncertainty!

sjm ;)