Get Out of ISO 9000 and Back to Real Quality
Amy Zuckerman's articles on the ISO 9000 scene in the July and November 1998 issues of QP and the correspondence generated so far demonstrate the deepening quagmire that the quality systems standards business has become. While this particular swamp is full of alligators, there is an easy way to drain it. The first move in such an operation is to get out. Then, if sufficient other people perceive the wisdom of this, the alligators will starve.
The November article ("Late 1998 Should Prove a Watershed for Standards," p. 17) stated that the Industry Cooperation on Standards and Conformity Assessment (ICSCA) had met with the Independent Association of Accredited Registrars (IAAR) to start a dialogue on the future of third-party registration.
In particular, ICSCA wants to eliminate it altogether. Of course, if they did so the registrars would no longer have a role, or a job. That's why they want a dialogue. They need ISO 9000; industry does not.
The registrars are like the alligators: They are not safe to be with in the swamp, they cost a lot to feed, and they provide no value. U.S. industry, through the ICSCA, seems to be waking up to this, a long time after most leading teachers of quality management started warning them.
In all of my work with U.S. and European companies, I observe a strong antipathy to ISO/QS/AS 9000 from managers and engineers. Dilbert, probably one of the wisest commentators on management at work, observes the futility and waste of the approach.
Most Japanese competitors to U.S. and European companies stay out of the ISO swamp (and laugh all the way to your customers). Note that Toyota, Nissan, and others do not subscribe to QS-9000. The only people who defend the approach are those who live by it.
My advice to the ICSCA, to industry in general, and to companies that take real quality seriously is simple: Get out of ISO 9000. Which will be the first company or industry group to step out? How long will it then take for the swamp to dry out? When that happens we will be able to abandon the whole ISO 9000 fiasco and get back to the basics of real quality.
More on Preventive vs. Corrective Action
While contributing the concepts of adaptive and contingent actions, Dave Wessel in "An Ounce of Prevention" (December 1998, p. 33) gives a similar explanation of corrective and preventive action as C. Michael Taylor ("Preventive Action vs. Corrective Action: The Horse, the Barn Door, and the Apple," March 1998, p. 66): separating the two based on whether problems have occurred or not. There are several benefits to using other definitions of corrective and preventive action.
To illustrate these alternative definitions, consider Wessel's example of the oil-leaking equipment due to a worn seal and Figure 1. Wessel claims that effectively replacing the seal is corrective action.
In Figure 1, he describes, as others have, that corrective action is action taken on problems that have occurred in the past. Both the cause and the effect are in the past. I know of no action that can alter the past. Thus, replacing the seal (or any other action) will never prevent the leaks that have occurred. Replacing the seal prevents future leaks (leaks after the seal has been replaced).
The definitions that require a problem to exist for corrective action and to not exist for preventive action, are unnecessary and confusing. Why else the discussion about the terms?
For example, Phil Schoner in Letters ("March Author Confused Corrective and Preventive Actions," July 1998, p. 5) replied to Taylor that if data on past problems are used, it cannot be preventive action. So, he implies that what distinguishes corrective and preventive action is the use of existing data.
However, Schoner does make a point consistent with my point: "[If] the problem has already surfaced... it's too late to be preventive." This is certainly true of the past occurrences, but false of future occurrences of the same problem.
Consider two cases. The person fixing the problem (a) erroneously believes that the problem has occurred or (b) does not know the problem has occurred. In case a, the person erroneously believes the action that keeps the problem from occurring again (in the future) is corrective when, under these definitions, it is preventive.
In case b, the person erroneously believes the action to be preventive, when under these definitions it is corrective. In either case, whether corrective or preventive, the action never prevented the past occurrences but only prevented future occurrences.
So, what definitions would be more useful? Wessel gets at the idea in his examples and figures. One must alter a cause to change an effect. In the context of processes, which is one focus of ISO 9000, the Baldrige Award and continuous processes cause outputs.
Let's define these two terms with respect to action taken on the process or the output. No action taken on the output will prevent the existing cause-effect relationship from changing. However, by taking action on existing output, we can change whether it has the problem.
Similarly, action on the customer can change whether the customer views the defective output as acceptable. Thus, I define corrective action as action taken on existing output to correct its problem or on the customer to compensate for the problem with the existing output received.
Since only action on the process can alter the cause-effect relationship, then I define preventive action as action taken on the process to prevent existing problems from continuing to occur or potential problems from actually occurring. Thus, knowledge of whether the problem has occurred or the kind of data used has no bearing on whether the action is preventive. The only determinant is whether the action does prevent the problem from occurring henceforth.
Additional insights into these two new definitions are:
They can be done independently, simultaneously, concurrently, or sequentially.
Preventive action is independent of whether the problem has occurred because past problems cannot be prevented from having occurred.
Preventive action has the main characteristics that Wessel identifies: future oriented, anticipating and eliminating the likely causes of potential problems.
Corrective action is consistent with the notion of applying to existing problems by changing existing output that has the problem, in other words by removing the defect.
The confusion about the time component inherent in corrective and preventive action is alleviated.
The confusion about whether an action is truly corrective or truly preventive is alleviated.
Wessel's concepts of adaptive and contingent actions are still applicable and valuable.
Any action on the process cannot be corrective; any action on the output (or customer) cannot be preventive.
This definition was adopted by one organization that got ISO certified, so it does meet the intent of correcting existing problems and preventing future problems.
Aejes LLC. Ridgefield, CT
A New Member Joins the ISO 9000 Dialogue
As a quality professional in England, I have admired Quality Progress from afar and have finally taken the plunge and joined ASQ. Congratulations on a fast and efficient service in organizing my membership.
When my first edition of Quality Progress arrived in December, I read it avidly and am keen to take part in the dialogue about quality and ISO 9000 appearing in the Letters column (and elsewhere).
The first point I would like to take up is that made by Saad Benmiloud regarding the revisions to ISO 9000 in the year 2000 ("Year 2000 ISO 9000 a Nonstandardized Standard," p. 5). I applaud the International Organi-zation for Standardization for having the courage to move away from the 20 elements.
The move to grouping standard requirements more logically into processes is welcome, and the effort required to reconfigure our quality mind-set might help us make our quality systems more attuned to our businesses than to a collection of requirements.
As a third-party auditor I would always dread reviewing quality manuals duplicating ISO clause numbering and the standard format. The risk was always that the company had read the words without understanding the meaning and had put together an ISO 9000 quality manual rather than one that described its business. Often these companies had problems in the course of an on-site audit.
I agreed with a lot that Jay Stanhan writes and, not having seen previous correspondence can only comment on the content in his letter ("More on Gunter's Column," p. 6).
The revision of ISO 9000 makes more explicit reference to objectives (which should include measures if they are to be meaningful) at all levels and for all functions of the organization and so should help us all to demonstrate the continuous improvement that ISO 9000 would like us to achieve.
But, let's be clear--ISO 9000 in itself adds nothing. What the standard does is provide a focus for organizations and a structure around which they can construct their systems. In the same way, for example, the ASQ certification program adds nothing to the lot of the quality professional, but what it does very well is create a standard against which individuals can be judged and a target to which they can aspire. This will cause them to add their own value by taking courses and acquiring knowledge as a means to achieving ASQ certification.
So it is with ISO 9000. The value added is in the understanding and improvement of business processes. Implementation of the documented system adds further value by reinforcing the company policy for quality and training employees.
ISO 9000 is not a miracle cure. The point is well made that there are many other models and programs that can help a company achieve quality. It is, however, the one international standard that can be applied by any organization and used by customers and third parties to assess achievement of requirements.
PAUL SIMPSON, quality
Calor Gas Ltd., Warwick, United Kingdom
Deming Would Be Displeased at Award Competition
As I was reading a news item in the December 1998 issue ("Hurd Foundation Awards Two Quality Medals," p. 8) it bothered me when I saw "compete" and "Deming Award" in the same sentence.
W. Edwards Deming was firmly opposed to the concept of competition. Competition means that there will be winners and losers. It is the surest way to demoralization. One of the reasons Deming stayed away from the Malcom Baldrige National Quality Award was that it is a competition where best wins in each of the categories. The Deming Prize, on the other hand, according to Deming, is awarded to any number of people as long as they reach certain levels.
Sales Process Engineering Not Really Complicated
Paul Seldon's article "Sales Process Engineering: An Emerging Quality Application" (December 1998, p. 59) is a good example of the application of the concept of analyzing any work in terms of a process and then applying the appropriate tools and techniques to its improvement. It is one of the basic elements of total quality management (or six sigma).
The same methods are being used in all kinds of organizations, and it is good to see specific examples in functions such as sales, which has not previously thought of what is done in process terms. However, the tone of the article suggests the methods are complex.
Seldon defines sales process engineering as "the systematic application of scientific and mathematical principles." Sales process engineering is an overblown title. It is simply process identification, description, analysis, and improvement. Calling it "engineering" is magnifying the technique beyond its actual complexity. It can be learned by anyone of average intelligence. That is one of its virtues.
The article contained some simple arithmetic but I found no "scientific or mathematical principles" expressed anywhere. I make these criticisms because I think it is important that the uninitiated reader not be intimidated.
These techniques are valuable and quite simple. The difficulty and complexity occurs in getting management and then employees to accept the application of these methods to their organizations and work; in other words, getting them to change.
Reader Objects to Article About Quality in China
I am appalled that the state tyranny of communist China and its brutal regime, which controls the Chinese people and their economic entities without regard to liberty or human rights, was treated as an interesting case study on how to increase product quality ("State Supervision and Inspection of Product Quality in China," Zhihai Zhang, December 1998, p. 53).
The reason that I am condemning the "quality" organization in China is because it is an enforcement arm of the government, which has only recently begun economic liberalization. Until the reforms of Deng Xioaping, Chinese rule was virulently communist, with all forms of private ownership and commerce being persecuted as crimes.
China is still in the grip of a government that is swiftly moving toward a ruling style that might be more accurately described as "communism lite," with some degree of personal ownership and private commerce, but with the government firmly in control of all other possible rights that should be accorded to the citizen.
The first paragraph begins, "Relatively low product quality is one of the most recognized and serious problems affecting China's economic development."
What a cruel joke when communism--because of its forced relocations, property seizures, and other barbaric measures by which the general populace was stripped of all material possessions during communism's takeover, known as collectivization--resulted in widespread famine in the 1960s. Historical estimates of the dead range in the tens of millions. Such is the fruit of Communist economic policies!
The very idea of quality is meaningless in a state-run economy, where consumers and producers are treated as chattel. Only in a free-market, capitalist society, where free choice as to what to buy and sell is the norm, can quality truly be measured and defined.
Take this example from the article: "Consumers, on the other hand, are also responsible for the market strength of low-quality and fake products. They prefer such products simply because of their low price.
"One of the main reasons for the extensive market for fake products, especially fake foreign goods, is that these products pander to the vanity of consumers, just as they frequently do throughout the world." This statement shows the contempt for personal choice inherent in their system.
How dare the average Chinese wish to purchase goods that don't meet the government's arbitrary standards or that the government considers frivolous! Thus, the free exchange of goods, money, and ideas that are the lifeblood of a free society are strangled, lest the population feel that they, rather than the government, should control the minute details of their personal lives.
Note that I do not defend the infringement of trademarks, brand names, and intellectual property, as that is clearly contrary to the principles of private ownership that I believe in.
But to feel too strongly a sense of moral outrage at fake products, when the majority of the Chinese populace have been stripped of their most basic human rights to own property and run businesses, reminds me of a man complaining of a headache caused by the loud crack of gunshots when around him people are being murdered.
On a more personal level, I am disheartened and saddened that Quality Progress is unable to differentiate between the free-market quality concepts which I serve and the oppressive alternative that communism offers. True quality (defined as offering a product which meets or exceeds the customer's needs and wants) can only come when the freedom to choose is present and the production of a high-quality product is rewarded by the free choice of the consumer.
Any other arrangement is a farce, built on coercion and terror, and it stands in direct opposition to human dignity and freedom.
Quality control engineer
North American Lighting, Salem, IL
To Prevent Problem, Use Ishikawa to Fullest
Rebecca Staton-Reinstein's letter in the December 1998 issue ("'A New Slant on Fishbones' Not Really New, but a Good Example," p. 5) sent me looking for the September issue, in which an article on the subject by Jonathon Lore appeared ("A New Slant on Fishbones," p. 128). For some reason I missed the article.
I was appalled to read the example for five-why offered by Lore. He fell into the trap so many people fall into, which is one of the reasons problem solving is ineffective.
The assemblies came loose. Why?
The components did not fit together tightly. Why?
The fitting's diameter was too big. Why?
The parts were gauged incorrectly. Why?
The gauge had not been calibrated. Why?
Root cause: No one had gauge calibration responsibility.
The cause-and-effect scenario states that the cause of the fitting's diameter being too big was a bad gauge. The bad gauge did not cause the problem; it didn't find it. The problem already existed before the part was gauged. Even after we fix the gauge, we would continue to produce fittings with the diameter too big.
The "why" regarding effect No. 3 should have been directed to the cause of the diameter being too big, which may be any one of the four M's and E (manpower, method, machine, material, and environment).
I have been using several whys, or as many as needed, to reach three root-cause paths:
The specific root cause of the problem being investigated
The escape root cause of why we did not detect the problem
The systemic root cause of why the system allowed the specific root cause
If we are to prevent recurrence of problems, we must use the Ishikawa diagram to its fullest potential by seeking the correct answer to each why.
Vice president, global quality
Siebe Automotive, St. Lucie West, FL
Upon reading Ricci's letter and many others over the years, I have noticed that some readers seem to spot shortfalls in articles, rather than look for value. Often the alleged shortfall is really a difference of perspective, such as the reader working in high-volume manufacturing and the author in custom machine building, distribution, and service industries.
So, if my example confused Ricci or any other readers, here are three ways no one having gauge calibration responsibility can be the root cause of a fitting with too large a diameter:
A machinist uses an uncalibrated gauge to measure the custom fitting as he or she is making it.
Several similar but slightly different-sized fittings are manufactured to specs and sent to a distributor or company that does assemblies. Receiving inspection uses an uncalibrated gauge that misidentifies the parts.
The fitting is a stocked part on a machine builder's shelf. The assembler chooses the wrong fitting for the mating part by measuring it with an uncalibrated gauge.
As for Rebecca Staton-Reinstein's previous letter stating that the five-why approach for fishbones is not new, I wonder if the reader, misled by the secondary title line that the editor inserted, glosses over the later section of the article regarding the proposed use of fishbones for benchmarking.