When Dean Dunaway sells $20 worth of oxygen, it's delivered in a $200 tank. If a customer or employee loses track of a single one of those costly containers, it can erase the profits on a huge number of product sales. The bad news is, it happens all the time.
"We have probably 15,000 or 20,000 transactions a month moving those containers," explains the president of 72-person industrial gas distributor Capweld Inc. in Jackson, Mis-sissippi. Keeping track of that many tanks was a logistical nightmare: "Every time somebody reads a seven instead of a one or something like that [on the tanks' identification numbers], your [records] get off."
Billing customers for lost containers didn't work--customers couldn't relate to the idea that the tank was anything more than a tool to use. And employees, being human, occasionally committed errors. So instead of upsetting customers or haranguing employees, Dunaway turned to the idea of mistake-proofing, also known as fool-proofing or poka-yoke (from the Japanese).
First, he began using more technology to record and track tank identification numbers. "As we've gotten more stuff into the computer, that helps," he says. Next, Dunaway hopes to attach microchips to each of the 40,000 tanks his company owns. Scanning the chips to grab ID numbers will further reduce input errors.
"It's helped a lot," Dunaway says of mistake-proofing his design of products and processes. "I hope it's going to help us a lot more."
Dunaway's hopes are likely to be realized, according to Richard B. Chase, a business professor at the University of Southern California in Los Angeles and co-author of Mistake-Proofing: Designing Errors Out (Productivity Press). Poka-yoke is a low-cost, highly effective way for companies of all types to reduce mistakes and boost quality, he says. "Personally," says Chase, "I think it's the next wave of the quality movement."
Poka-yoke (pronounced either "POH-kuh yohk" or "POH-kuh YOH-kay") comes from two Japanese terms meaning "avoid errors." The term and the concept are generally attributed to the late Shigeo Shingo, an industrial engineer for Toyota who felt that avoiding worker error should be a prime target for managers.
Shingo's variety of poka-yoke relies on devices designed into factory machines or the products made in factories. A classic example is a screw that an auto manufacturer had trouble threading because it consistently slipped off of screwdriver blades. Changing the slot on the head of the screw resulted in a firmer grip by screwdrivers and fewer worker errors.
Many other examples of poka-yoke exist. Drilling machines use jigs and depth indicators to keep workers from drilling holes of the wrong number or depth. Kits containing exactly the right number of each kind of part let assembly workers know at a glance whether they have included all the needed parts in a product.
Poka-yokes are easily identifiable in everyday life, says John Grout, a business professor at Southern Methodist University in Dallas. The automobile and gasoline industries use several poka-yokes to prevent errors while filling up. Specially shaped nozzles keep drivers from accidentally putting leaded gasoline in a car designed for unleaded. Leashes prevent loss of gas caps. And noisy ratchets alert drivers when caps are on tight enough, while also preventing overtightening.
Poka-yoke is also widely used in service industries. Martin M. Glesk, a management professor at the University of Pittsburgh, Bradford, cites hospitals that use specially indented trays to hold surgical tools. Surgeons finishing an operation have only to glance at the tray's indentations to see whether all instruments are accounted for or whether something might have been left inside the patient.
Another service industry example concerns a retailer that wanted clerks to make eye contact with customers. Point-of-sale terminals were programmed to require clerks to enter the customer's eye color for each transaction. "That forced them to make eye contact with the customers to see what color their eyes were," Glesk explains.
Whether it's a device or a process, used in service or in manufacturing industries, poka-yoke always has the same ends in mind. Says Chase, "It's a way to either control the process so that a defect doesn't occur or warn you that a defect is about to occur."
Steps To Start
An entrepreneur who wants to mistake-proof a process or product should first perform an inventory of existing poka-yokes, says Grout. This will lead to other fool-proofing methods or, perhaps, ways existing poka-yokes can be improved.
Another early step is to prepare a flow chart of the product's creation process. For customer service issues, Chase recommends developing a map that tracks interactions with customers. This will highlight possible failings and may help you find ways to improve the system.
Next, ask workers how their jobs, equipment or even products and services might be changed to avoid errors. At all times, avoid coming up with overcomplicated solutions. "Go simple," advises Glesk. "Be as specific as you can with the problem, the issue and the failures you're trying to correct. Very often the solutions are not complicated."
Perils Of Poka-Yoke
Mistake-proofing can, at times, be a mistake itself. The term "fool-proofing" can offend workers, who may feel they are being treated like idiots. (In fact, Shingo called the technique poka-yoke to draw attention away from the worker and toward the error.) Entrepreneurs can avoid ruffling feelings by bringing workers into the poka-yoke design process and carefully informing them of its purpose.
"You can go wrong if you don't discuss it with the people who are actually doing the work," warns William Hailey, chairman of the Department of Decision Sciences at Appalachian State University in Boone, North Carolina. "Get their ideas on how this can be simplified. Ask, `How can we change this to get the best quality?' "
Poka-yoke can also be misapplied. It's most useful in routine, mundane tasks, such as assembly line work or ringing up sales at a cash register, where memory and attention to detail are more important than creativity. Tasks that have to be done once a month rather than, say, once a minute are usually poor candidates for poka-yoke. Says Glesk, "Sometimes you can make the system more complicated, increase costs, increase difficulty and create problems you hadn't otherwise anticipated."
Where To Learn More
Many entrepreneurs, of course, already do poka-yoke without calling it that. But poka-yoke is increasingly being specifically targeted, especially by large companies that have had total quality management programs in place for several years and are looking for further ways to improve. In addition to Toyota, Mo-torola and AT&T Power Systems are among the large firms pursuing poka-yoke.
Shingo's books explaining poka-yoke tend to be difficult for non-experts to absorb. A few popular books, such as Chase's, focus on poka-yoke, and it is mentioned in passing in broader works on quality. Similarly, many quality training courses spend some time on poka-yoke. A few universities, including the University of Southern California, also hold one-day poka-yoke workshops costing around $500 a person.
Mistake-proofing need not be expensive, time-consuming or
complicated. "You're looking for simple, inexpensive
fixes," says Grout. The fixes may be small, but if numerous
enough, they can make a major difference. AT&T Power Systems
has implemented 3,300 poka-yokes at an average cost of around
$2,500 apiece for a total savings of more than
$8 million, Grout says.
In much the same way, Dunaway hopes to keep trimming away at the small errors that can have such big effects at Capweld. "There's nothing I can point to that helped us a lot," he says. "It's just hundreds of little bitty things."
Mark Henricks is an Austin, Texas, writer who specializes in small-business topics.