Work Instructions Needed Along With Training
As I read the article "Taking the Myth Out of Documenting Work Instructions" (Terry Dunford, December 1998, p. 30), I couldn't help thinking that the author was supporting the notion that there is no need for work instructions except for some product specifications. It seemed that he is advocating the elimination of operating procedures and test methods in favor of documentation of the training of individuals who perform the function.
It is my contention that a combination of both is required. It is simplistic to believe that once an individual is trained, he or she will forever remember every step of the complex task. Furthermore, without the procedure, there is no basis for training.
The author uses anecdotal information--namely the tensile test and standards--to support his case for using source documents to replace Level III work instructions. It is assumed that the source document he refers to is the American Society for Testing and Materials procedures.
Anyone who has read these standards knows the technician performing the test will not get a cookbook version of how the test must be performed. It takes a very knowledgeable person with technical skills to read, interpret, and perform the test. Level III procedures have an ability to filter out the superfluous information and get right to the steps that need to be performed. These procedures may be used as the basis for the training of new employees and as a refresher for those who perform the tests infrequently.
I certainly agree with the author's intent to simplify procedures and eliminate unnecessary ones. However, there are three reasons procedures should be written:
1. The task is very complex, and each step must be performed as specified in order to get the desired result.
2. The process is critical to the operation. If it is not done properly, there will be a product or process failure, which may not be detected or obvious.
3. The procedure is required by the ISO standard or by another authority.
The most important point that the article reveals is the caution that "the range and detail of the procedures that form the quality system shall be dependent upon the complexity of the work...and the skills and training needed by personnel in carrying out the activity."
Perhaps some operations could eliminate the Level III procedures and allow the work instructions to be interpreted by the operator or technician. In some operations, the work instructions may be part of the job traveler in the shop, but this is not the case in all companies. Had this article been entitled "Reduction of Work Instructions at CSM Industries," I may have been in full agreement.
I have to score this article a fair to poor because the author fails to take into account the fact that Level III procedures are the "how to. . ." of the organization. These procedures can be the best asset of a company documentation program if they are used as a recipe for operator training, a refresher course for the knowledgeable, and a guideline for consistency.
Manager, quality assurance
Klockner Pentaplast of America
I appreciate Carmine Covais' thoughtful letter because it helps me clarify some of my main points. I believe it is necessary to distinguish "how to" documents (including training materials or equipment manuals) from documents unique to a process (drawings or dimensions) that are critical to product quality. The former are usually general and transferable from one company to another, while the latter may be peculiar to a company's process. When documenting a quality process, a cost (in time and effort) must be weighed against practicality.
Consider this analogy: Few of us drive our cars with a step-by-step work instruction forever at our sides that begins, "Insert key in ignition." Our cars already come with an operating manual. Critical information is knowing the changes in speed limits.
Article on Prevention
A Good Start on Subject
I was delighted to see the article on prevention and ISO 9000 ("An Ounce of Prevention," Dave Wessel, December 1998, p. 33), a good and refreshing start on the subject.
I hope, however, it isn't the implication--or that the readers infer--that prevention science is relatively recent or novel. The basic underlying sciences of prevention in medicine, my area of training, began nearly 150 years ago as what is now epidemiology, reasoning upstream from the effects to the cause(s).
The discovery and use of increasingly sophisticated inferential tools such as Bayesian thinking, biostatistics, problem, causality, and risk analysis, coupled with better and better disease etiological knowledge, resulted in a recognized specialty of medicine about 50 years ago that has become highly refined and further subspecialized.
Wessel's (and apparently the ISO 9000 revisions') use of such terms as corrective (recurrence prevention?) vs. preventive action, while not fitting exactly my specialty's prevention vocabulary meanings, appears to correlate with such preventive medicine concepts as secondary vs. primary prevention. (Prevention in medicine is frequently divided into three stages, based on the strategy of the intervention. Tertiary is mitigating the ill effects.) Sometimes in quality science the terms "predictive" or "prognostic" have been used for preventive action.1
Some of us in the specialty of preventive medicine who have wandered into or discovered Deming-based quality science are delighted by the parallels and kinship between preventive thinking and modern quality thinking. Wessel touched on some of those. For example, his discussion of corrective action vs. preventive action reflects the need to understand the sequence of causality and exactly what the intervention is intended to accomplish. That is not such a large leap from understanding the Ishikawa analysis.
In medicine and public health, causality can be very complex and usually is probabilistic, not mechanistic, which then requires an understanding of risk dynamics and analysis.
Our firm specializes in strategic consulting at the interface of prevention and quality (and, to some extent, systems thinking--another conceptual domain useful to prevention). Unfortunately, not many of us have crossed into quality country, bringing our prevention expertise with us. And vice versa.
Much of preventive medicine, including public health systems, still tends to think of quality simply in terms of conformance to specifications. The insights of Shewhart, Deming, and others have not permeated many areas of the field. We need to bring the two closer together. Tremendous synergy would result.
Although preventive science is not a new domain to medicine and to the preventive medicine practitioner, it may be recent in other areas. It certainly is time, then, to begin what should be a very fruitful dialogue.
1. Sato Mitsuichi, "How to Structure and Solve a Problem," Diamond Harvard Business, May-June 1978, p. 49.
RICHARD M. BIERY, M.D.
The BroadBaker Group, Kansas City, MO
Measuring People Goes Against Deming Concepts
The authors of "Measuring People and Performance: Closing the Gaps" (Brian Morgan and William A. Schiemann, January 1999, p. 47) obviously believe in the philosophy that "you can only manage what you measure." They call this a time-honored maxim.
I guess it is not their fault. How could they have known that this is a costly myth that will erode the competitive position of American goods and services? The authors further talk about performance measures cascading down the organization. How could they have known that these numbers are not additive and cannot be cascaded down without appreciation for system?
The authors also lament that measuring people is inadequate at present. How could they have known that it is impossible to measure people? The apparent performance of individuals is their interaction with the system. Thus, rewards and punishments that the authors are trying to preach are, in fact, rewarding and punishing people for the performance of the system, which is beyond the control of the individuals. It can only demoralize them.
Phases like "holding people accountable" and "measuring results" sound appealing. Judging people, however, has never helped them do a better job. It robs people of the pride of workmanship. It is time for managers to realize that achieving arbitrary numerical goals is not a great accomplishment. In fact, any goal can be achieved by manipulating numbers or by increasing costs.
Management by objectives as currently practiced and promoted in this article are obstacles to growth. What is needed for industry, government, and education is awareness of W. Edwards Deming's concepts of profound knowledge. The authors themselves can start their education by studying the books Out of the Crisis and The New Economics. Deming thought it might take three decades for transformation of American industry. I think that the master gave us the benefit of the doubt. I think it will take more like five.
The attitudes and performance of people are measured every day, but Velury would have us believe it is impossible to measure people. But are people's ratings on performance appraisals, their opinions regarding their jobs, and their times in the 100-yard dash only consequences of their interaction with the system?
Certainly, individuals do not perform in a vacuum, but the assertion that the system--however the term might be defined--is the only source of variance in the interaction is an ideological rather than an empirical one.
Velury also states that it is time for managers to realize that achieving arbitrary numerical goals is not a great accomplishment. We would agree, while noting that the numerical goals we describe are far from arbitrary. Rather, they are performance targets on people issues with a demonstrated impact on bottom-line performance.
Our findings indicate higher-performing companies do a better job of people measurement than do their lower-performing counterparts. How would Velury explain this away?
Velury also states that any goal can be achieved by manipulating numbers or increasing costs. But consider the case of product and service quality. Companies may be able to kid themselves by manipulating their quality-of-service and product scores, but customers will tell them in no uncertain terms if they are not, in fact, meeting their expectations by giving them low marks on customer surveys and shopping elsewhere.
Finally, we note that in presentations at the Conference Board's National Quality Conference, the National Institute of Standards and Technology (the sponsor of the Malcolm Baldrige National Quality Award) identified two areas in which the quality movement has fallen short: measurement and the identification of strategic linkages between measures.
Quality improvement and measurement go hand in hand. As we move ahead into a new century, the trend will be toward more focused measurement on the people, customer, and operational issues that are predictive of business results. Strategic measurement of employee commitment, satisfaction, and performance will be an integral component of this trend.
BRIAN MORGAN and WILLIAM SCHIEMANN
Some Thoughts Regarding Customer Surveys
Editor's note: A reader received a guest satisfaction survey from a motel chain and responded with the following letter. He also requested input from Ken Miller ("Are Your Surveys Only Suitable for Wrapping Fish?" December 1998, p. 47).
In my consulting practice, assessing customer satisfaction is a key topic. Thank you for supplying another example for my presentation. I will block out all references to you, but plan to use this form as an illustration of how not to do it.
A dissatisfied customer is exposed to greater irritation by being reminded of his disappointment. He does not want to respond to issues that do not relate to his problem. Your questionnaire had 20 headings with as many as 21 subtopics. There were more boxes to be checked than there are on some Internal Revenue Service forms.
A good survey is focused on what was expected, were expectations met, areas of disappointment, and whether the customer will be a future customer.
I wanted a pleasant supper and a good night's sleep. Neither hope was fulfilled. The dining room was uninviting. The blankets smelled moldy and dusty. The room had been cleaned and was well made up, but was not what I expected from a facility belonging to a major chain. I will not visit your motel if any alternative lodging is available.
I am not interested in a refund or other consideration. When speaking to companies about customer satisfaction, I offer the advice of several of my mentors: Don't waste money on what you can't change. Fix the system so it does not happen again.
I would appreciate seeing the corrective action plan generated by my complaint. I teach my clients to request a corrective action report for each nonconformity discovered in the product. Such a report should discover the root cause for the failure and make an effective correction. But, as W. Edwards Deming used to teach: The root cause of a system failure is always a management failure. It is never solved by employee discipline or incentive plans.
ROBERT W. BEDARD
Certified quality auditor and engineer,
I did not have the opportunity to view the survey, but based on experience, I can assume that it probably does not ask about many of the things that caused Bedard's dissatisfaction.
I applaud the active role he took as a customer to express his wants and concerns to the organization. Hopefully, it will follow his advice and not only take corrective action, but launch an aggressive campaign to proactively determine what customers want by actually talking to them and then build in those priorities.
Transition to ISO 9000-2000 Should Be Made Easy
I must agree with Saad Benmiloud (Letters, "Year-2000 ISO 9000 a Non-standardized Standard," Decem-ber 1998, p. 5).
Perhaps the greatest value that any standards-making organization can accomplish is to show how important the concept of migration is. Thus improving something may mean changes, but the changes must be easy for those who have already invested in the previous version. If not, we get a response that says, "I'll just wait for the next version."
Too many consumer products--computer software in particular--do not make transitioning easy. This should not be true for important materials such as the standards that we consider our professional foundation.
J.R. (DICK) SWENSON
The Boeing Co., Seattle, WA