2009
Make Everyone Aware of Quality
I was almost excited when I saw the title on the cover of the November 2001 issue: "Make Quality Everyone's Job." Imagine my disappointment then when I found only slight variations on the standard quality by proclamation approaches already so common. Based on the articles, the theme should have been "Make Everyone Aware of Quality."
Say point A is the day senior management makes the informed (keyword) decision to do quality. And say point B is the day a complete quality process is in place that involves everyone on the payroll on a continual basis; has formal components of leadership, process analysis, communication, training, measurement, recognition and celebration; and is having a positive, measurable impact on the bottom line.
Then ask how long it's going to take to get from point A to point B. If the answer for an organization of 2,500 or fewer is more than six to eight months, senior management's commitment or the consultant's competence should be seriously critiqued because it is deficient.
Is such a measure realistic? Yes. I'm sure because I've done it. Twice. I've had to meet this goal using two different, successful quality efforts in two service companies, and I have written and spoken about this notion for years.
Instead of asking, "Who should be involved in this effort and at what pace?" senior management should begin by asking, "Whom can we afford to exclude from this effort to improve?" Put more bluntly, "Whom do we have on the payroll who we think is incapable of improving?" The list will be short, especially if the senior executives are required to write down names. Everyone else should be included in the effort because each person has something to offer the effort to improve. Building a process that allows all employees to contribute is senior management's responsibility.
PAT TOWNSEND
UICI
Dallas
ptownsend@uiciinsctr.com
ANOVA Methods Detect Percentage Differences
The October 2001 "Statistics Roundtable" article (Richard F. Gunst, "Inferences on Percentage Changes," p. 56) looks at some statistical properties of percentage differences. Gunst notes the percentage difference is seldom addressed in statistical literature. There are two good reasons for this oversight:
1. When the numbers in the percentage are measured incorrectly, the percentage difference is biased.
2. The confidence intervals are surprisingly wide.
Define the growth rate as Z = (XPOST XPRE)/XPRE. This represents, for example, the change in some output measure after a change in the process. Suppose both XPOST and XPRE are measured with error, but have the same statistical expectation. Even when E[XPOST] = E[XPRE], E[Z] > 0. The amount of this bias depends on the distributions of XPOST and XPRE. The more variable they are, the greater the bias.
In other words, when you are looking at the effects of an experiment, using relative change will likely lead to a misinterpretation of the results. Furthermore, even when the amount of the bias is negligible, confidence intervals for Z under the null hypothesis (there is no change from PRE to POST) are distressingly wide. This means when Z purports to indicate a reasonable, but incremental, change, you cannot reject the hypothesis that there has been no change.
Both of these effects--the bias and the wide confidence intervals--may explain some of the difficulties our economy is facing. If percentage differences are managed, then managers are looking at the wrong thing. Managers end up failing to detect changes when the changes are there and reacting to changes that are simply noise.
You are much better off using analysis of variance (ANOVA) methods than the relative difference to detect differences from PRE to POST.
RICHARD O. LIGHTBURN
Deerfield, IL
lightburn@earthlink.net
Note: For more details, please see Lightburn's full
letter on the Quality Progress Discussion Board at www.asqnet.org.
"Crosby Taught Me Almost Everything I Know"
I read the October 2001 article about Philip B. Crosby
(Kristen Johnson, "Philip B. Crosby's Mark on Quality,"
p. 25) with great interest because Crosby had the most direct influence
on my business career. I was disappointed, though, that his 14 years at
ITT were written off in a simple sentence: "A year later, he left Martin-Marietta
and began working as the corporate vice president of quality at International
Telephone and Telegraph (ITT)."
When Crosby joined ITT in 1965, the highest level quality manager was a chief inspector, and quality meetings were held in the basements of the manufacturing plants. When he left in 1979, all the major subsidiaries had quality directors who reported directly to the presidents of these units, and quality meetings were held in the subsidiaries' boardrooms. Crosby was ingenious and a spellbinding orator, and he surrounded himself with people from around the world who helped implement his innovations.
Crosby made a big difference in my decision to join ITT in 1967 as director of quality and reliability of the semiconductor division. In 1969, he invited me to join his staff as director of quality--technical support (quality engineering). Between 1969 and 1979, I also reported to him as director of consumer affairs and director of product safety and environmental affairs.
Crosby taught me almost everything I know, and I am grateful for his legacy.
ERNEST W. KARLIN
Karlin Associates
Delray Beach, FL
ekarlin404@aol.com
Ford Should Manufacture Quality Into Its Products
I think there's an enormous communication lapse between Bridge- stone/Firestone and Ford (R.W. Hoyer, "Why Quality Gets an 'F'," October 2001, p. 32). Any company that relies on an integral part from a supplier should take the time and money to fully understand all aspects of the product and its application. It doesn't matter if the tires are attached at the beginning, middle or end of the assembly process.
Do huge corporations rely too much on the word of the supplier instead of simple inspection/experimentation data to back up the supplier's word? I believe this simple procedure has been overlooked in the face of rising costs and a sluggish economy.
Incorporating more inspection and testing is only a temporary quality control. Ford should manufacture or assemble, not inspect, quality into its product. As one of Deming's 14 points for management states, total quality cannot be controlled by mass inspection only. A certain amount of field work, product validation and continual process improvement has to be performed.
Just because the tires worked for the first few vehicles does not mean the process is frozen, and there is no room for improvement. I offer my unspecialized words of advice: Both companies must do everything in the realm of the real world to fix the problem, instead of playing the blame game. Priority one is to re-establish vehicle safety.
STEPHEN FOSTER
Indium Corp. of America
Yorkville, NY
sfoster@indium.com
Why Aren't Quality Professionals in Charge?
I couldn't agree more with R.W. Hoyer's assessment of the current state of quality in the American automotive manufacturing industry.
I am currently the director of manufacturing at a midsize Tier One, QS-9000 certified automotive seat frame manufacturer. I was promoted to this position after spending almost 30 years in the quality profession. As each new quality program was introduced, I thought at last we would do it right. Only later did I realize it was business as usual.
However, the state of quality has improved over the years. We are now forced to supply defect free product to our customers. Our current parts per million defects at the customer level is 6.5, which is world-class.
The problem is we have not found a way to keep from making defective product. We have just been more successful at keeping it from reaching the customer. We do not engineer processes to produce defect free product. We still require the production operators to find all the bad ones. The root cause of most manufacturing quality problems is that, at crunch time, we don't follow our own quality systems and procedures.
I recently learned of Ford's decision to replace Jacques Nasser as chairman with William Clay Ford Jr. The new chairman has made it clear quality will be his primary thrust. Ford is not the only corporation to openly declare its focus on quality. However, if quality really is the main issue, why haven't these companies placed a quality professional in charge instead of accountants, lawyers and engineers?
JAMES E. BRAKER
Alex Products Inc.
Definace, OH
jim.braker@alexproducts.cc
Author Uses Temerity To Critique Quality
Two articles in the October 2001 issue deserve broad scale plaudits, "Philip B. Crosby's Mark on Quality" and "Why Quality Gets an 'F'."
As I am a past national officer of ASQ, my association with Crosby goes back many years, and I have great respect for the man and his opinions.
In the October article, Joseph DeFeo of the Juran Institute is quoted referring to an observation Crosby once made: "(Crosby) stated that it would not be long before Ford Motor Co. would once again have serious quality problems since it dropped its slogan, 'Quality Is Job One!'" Now, think about the ongoing problems Ford has had with Firestone tires and relate them to the other article, "Why Quality Gets an 'F'."
Hoyer gave the Ford subject an erudite treatment, and he obviously did his homework well. But this is not the part of the article I wish to discuss. Rather, it is the theme he mentioned in the opening paragraph of the article, followed by the contents of the second paragraph. Such an approach requires temerity and integrity, and I applaud him for possessing both!
Hoyer directs a finger quite pointedly at the cracks in the professionalism of our Society, and I can't help but think of him as a kindred spirit. For a number of years, I have attempted, without success, to do much the same thing, and it is reassuring to finally see the subject given the public attention it deserves.
DALE L. LOBSINGER
ASQ Fellow and past president
Paradise, CA
Quality Alone Shouldn't Get a Failing Grade
I would like to commend R.W. Hoyer on his article "Why Quality Gets an 'F'." I think he hit home on several points, especially with the fact that an automobile, including the tires, is a complete engineered system.
However, I was a little disturbed ISO 9000 and QS-9000 were referred to as snake oil. These standards should not be used as scapegoats for what happened at Ford. I believe these standards were put in place in an attempt to establish a minimum level for a quality system.
I think people should know the standards do not guarantee a defect free product. However, being a certified company does let people know, at a minimum, an organization has a documented quality system. It is still our obligation as quality professionals to help our management improve these quality systems to minimize the defects in the products we produce.
Quality should not be the only one to get this failing grade. All the functions of the company from top management down should also get a failing grade. As quality professionals, let's not keep blaming quality standards or quality systems for lack of individual commitment and responsibility in doing our jobs to the best of our abilities.
DREXEL COLLINS
Limitorque
Lynchburg, VA
drexel.collins@invensys.com
