I
once worked for a company that employed more than 1,000 people in the quality department alone. Today, that company has no employees in the quality department--in fact, it has no such department at all. Think of how much money it must save. The next company I worked for had a quality system based heavily on inspection and inspectors, and I watched it make the transition toward virtually eliminating inspectors. This resulted in a more effective workforce that cost less. Today, I work for a small, lean division of a large corporation, and we have no inspectors. I'm the only employee in our "quality department," and I serve more as a consultant than as anything else. We are efficient: Our 50 employees will make about $3 million in profit this year.
This modern trend in the quality profession is not limited to traditional areas like inspection. Statistical process control (SPC) is also affected. The simple truth is that
SPC is not a quality function; it's a production function. Production is where the action's at, because once the parts are made, it's too late to do anything about quality. Oh sure, I know you
can screen the parts. You'll find some of the bad ones--you may even be able to pass an acceptable quality level inspection--but we all know that bad parts will remain in the lot. You'll also
throw away a lot of good parts. It costs a pile of money to screen parts this way. Clearly, it's far better to study the process that makes the parts. If you really understand
how that process works, you won't make any bad parts. (You may make a few bad parts during setup, but that's quite different from in-process scrap, when parts are defective because the process
changed in a way that you didn't want it to change.) Process control is for the manufacturing stage of production, and the people best able to perform it are the production operators.
This isn't theory. I see it in action every day. At Parker Hannifin, our production operators check their parts and enter the inspection results into an Access database. I review this data
daily, and I'm confident in the results because I trained the people checking the parts. In addition, I've conducted measurement error studies that quantified our measurement error.
It's just as easy to train a production operator to check parts as it is to train an inspector. Moreover, the production operator has far greater ownership of the process than any inspector
can ever hope to have. A strange phenomenon occurs when you use inspectors: The production operators, or at least some operators, begin to believe that they're no longer
responsible for the final quality of their product. They think the inspector will catch anything that goes wrong. At the same time, the inspector never feels responsible for the quality of the
product because he or she didn't make it. As a result, neither inspectors nor operators care as much about quality as they should. You're paying people to be less responsible about quality. However, if you eliminate the inspector, the production operator suddenly knows who's responsible and that there's no one downstream to catch any errors. There is also the
direct benefit that you no longer have to pay the inspector. This kind of system might look a little frightening to someone who hasn't seen it in action, but it's important to remember that
errors get through an inspector-based system too. There's no way to completely guarantee that no errors will be made nor that none will get through, but there are ways to
minimize the chance of that happening. One of the best ways to minimize errors is to make sure employees understand that they are fully responsible for the quality of the parts. Companies can
save thousands, perhaps millions, of dollars by understanding this one simple fact. A relatively minor change in organizational structure can have a profound impact on an
organization, and the change may not be nearly as great as it seems. For example, just as production workers can be taught to perform inspection, inspectors can be taught to make parts. After a
year or so, you can merge the inspectors into the production activities and end up with a team of people who both make and inspect their own parts. Attrition sometimes makes it possible to
complete this transition without any involuntary job loss. William Conway once told a story about a live-fire exercise from his experience in the U.S. Navy. An old chief told
him, "Mr. Conway, if everything is right, things will be OK." That's the key to quality: Make sure everything is right. Inspectors are more a safety net than they are a quality tool. If you
expect things to go wrong, if you design your quality system to detect errors rather than prevent them, you're not going to make very much money. But if you take a proactive approach, you'll be
so well-heeled you can buy one of those advertisements during the Super Bowl. SPC is intended to be used at the source of production, by production workers. With a little
training and a little creative insight into organizational structure, everybody can grab a handful of free money. About the author
Gregory P. Ferguson is quality manager of Parker Hannifin's Tucson, Arizona, facility. He has published technical articles and assisted in the publication of two books.
Comments can be e-mailed to him at gferguson@qualitydigest.com . |
