「華人戴明學院」是戴明哲學的學習共同體 ,致力於淵博型智識系統的研究、推廣和運用。 The purpose of this blog is to advance the ideas and ideals of W. Edwards Deming.

1989年1月5日 星期四

quality function deployment, or Q.F.D.,Raising Quality: Consumers Star

Raising Quality: Consumers Star


Published: January 5, 1989

LEAD: Deep down, what do people really want in a car door? That is not a trivial question for the Budd Company, which is using an increasingly popular Japanese quality control system to find the answer and make better doors as a result.

The system, known as quality function deployment, or Q.F.D., has been in use in Japan for several years but is only now gaining favor in the United States. The system uses a chart that resembles a drawing of a house with a peaked, tiled roof and is often referred to as the ''house of quality.''

Using Q.F.D., Budd, which makes doors for the Big Three car companies - General Motors, Ford and Chrysler - found that defining the perfect door is not simple. Most car buyers say they want a door that closes easily. But they also do not want to get wet when driving in a rainstorm. The two desires are in conflict, because doors that seal tightly enough to keep out water require a good slam to close properly.

Q.F.D. helped the Budd engineers to learn what consumers really want in a door and to get that translated into engineering specifications. The system involves an elaborate series of charts to spell out consumer desires, rank them in importance, identify conflicts and set specifications.

The acceptance of Q.F.D. by companies like Ford Motor, Budd, ITT and others is a recognition that many of the problems that have plagued American manufactured goods are embedded early in the design process, not on the factory floor. Rather than trying to increase quality by adding legions of inspectors at the end of assembly lines, these companies are seeking to prevent defects with more thoughtful designs and production processes.

''We've been doing a lot of these things for years, but with large, complex systems you can end up working at cross purposes,'' said Dana M. Cound, vice president for quality and productivity at Gencorp. ''Q.F.D. provides discipline to the process. It forces you to pay attention at meticulous levels of detail.''

Companies using the system say it not only helps them tailor products more closely to customers' wants, but also dramatically slashes product development time and costs. They note that the Toyota Motor Corporation cut the cost of a new mini-van by 61 percent in the early 1980's by applying the Q.F.D. system.

''Q.F.D. helps point us to the things that give us the biggest bang for our engineering dollars,'' said William C. Phillips of Budd, which conducted a study for a mini-van door. He described the process as ''agonizing'' but ''valuable.''

''It is an extremely valuable tool,'' said David Cole, director of the University of Michigan's Office for the Study of Automotive Transportation. ''It forces people to think in a formal and organized way.''

Some American executives concede they were skeptical of the charts, which can involve hundreds of hours of tedious meetings. ''We were not particularly impressed when we first saw Q.F.D. in 1983,'' said Robert J. Marshall, engineering director of Ford's body and chassis engineering group. ''It seemed as if they were just writing down the obvious. The trouble was, we have had failures with things that should be obvious.'' Indeed, quality specialists have an aphorism that says, ''If it goes without saying, it may go without doing.''

Since then most of the 2,600 engineers in the body and chassis group at Ford have taken training in Q.F.D. methods, and Mr. Marshall said there were 80 to 90 Q.F.D. studies under way. 'Bound by Convention'

The first step in Q.F.D. is to ask customers what they want in a product. Although this seems obvious, it appears to represent a major change for some companies.

''The problem is that we have been bound by convention,'' said Ralph E. Reins, president of ITT Automotive. ''Engineering dominated design without much input from marketing. The key event is that now we're taking customer requirements and factoring them into design.''

Q.F.D. proponents encourage engineers and designers to get involved in consumer research, either through groups of selected customers, or, as the Japanese do, by putting a car or an appliance in a public place and noting the comments of passers-by. Such comments are said to be more helpful to designers than a dry report from a product planning group.

Customer desires are grouped by category and written along the vertical axis of the array. The product characteristics to meet them are written across the top. The matrix is then used to check correlations. If, for example, customers place a high value on non-leaking doors, the chart would suggest that special attention be paid to the dimensions of the door and the flexible sealing material.

The top of the ''house'' relates the various product attributes to each other to check for conflicts. That is where the leaks-versus-door-closing problem becomes clear.

The chart, which ranks the value that customers place on each feature, may indicate that the customers are willing to slam a little harder to keep the rain out. But it may also suggest that the supplier of sealing material should develop a softer compound that would require less effort to close the door. Getting Involved Earlier

The charts are prepared by a group that includes marketing and manufacturing specialists as well as engineers and designers. ''In the past, someone else would design a product and send it to us to develop a process to manufacture it,'' said Mr. Phillips of Budd. ''Most of the time we would have to go back and ask for design changes. That said we should get involved earlier in the design process.''

Q.F.D. advocates say that taking the time to fill out the charts and, more important, to develop the information that supports them, pays off in fewer changes close to production. ''It may take 60 hours of meetings up front to go through the process, but it may save you 600 hours downstream,'' Mr. Phillips said.

Other benefits of the Q.F.D. approach are that customer desires are enshrined as the major goal of a program. ''It's easy for the focus to go to the art of the possible rather than what you really want,'' said Mr. Cound of Gencorp. ''Q.F.D. drags you back to the touchstone.'' 'Lives a Long Time'

It also serves to preserve knowledge. A person working on a new project can look at an old Q.F.D. chart and understand the philosophy behind a design. ''Q.F.D. properly done lives a long time,'' said Mr. Reins of ITT. ''It becomes part of the data base.''

This is an important consideration in a mobile society. ''It seems as if American industry makes the same mistakes and corrections about every five years,'' Mr. Cound observed.

An important part of a full Q.F.D. chart is an assessment of the way a proposed design ranks with competitive products. If the design falls below the competition on important features, a revision is indicated. If it ranks much better, those features can be emphasized as selling points.

Japanese manufacturers see quality as existing on three levels: expected quality, like a car that starts in the morning; performance quality, perhaps the smoothness of a ride, and unexpected quality, which is something that surprises and pleases a buyer. The net in the trunk of Ford's Taurus and Sable cars that prevents grocery bags from tipping over is often cited as an example of unexpected quality.

Q.F.D. advocates say that other American companies must develop this kind of thinking if they are to remain competitive. Simply preventing defects is not enough. ''Up-front efforts concentrate on what the customer likes, rather than fixing what he does not like,'' said Norman Morrell, a quality specialist at Budd.

If there is a downside to Q.F.D. it is that it is so different from standard American practice that it is hard to get started. Mr. Marshall of Ford describes establishing the procedure at that company as a ''tortuous process.'' He added, ''The quality of some of our early Q.F.D. studies was lousy.''

But the pain of change is also a blessing, said Mr. Cound, a former chairman of the American Society for Quality Control. ''The greatest long-term benefit is that it will change the culture of companies,'' he said. ''If you are going to change behavior, you have to change procedures and methods.''

網誌存檔