The cartoon is a simple illustration of a fisherman sitting in a small boat in the middle of a lake. In the first frame, the water level in the lake (meant to represent inventory) is high, concealing rocks (potential problems) on the lake's bottom; in the second, the water level has dropped, revealing the rocks and allowing the fisherman to more safely steer his course.

What the cartoon illustrates is a system, called Just-in-Time, that has already changed the fundamental ways in which companies view their manufacturing operations. As a philosophy JIT targets inventory as an evil presence that obscures problems that should be solved, and that, by contributing significantly to costs, keeps a company from being as competitive or profitable as it otherwise might be. Practically speaking, JIT has as its principal goal the elimination of waste, and the principal measure of success is how much or how little inventory there is. Virtually anything that achieves this end can be considered a JIT innovation.

Take as an example the reduction of machine setup times. Long setup times require long runs in order to justify the expense; product not immediately required by customers becomes dead inventory, which then takes up space, gets in the way of employees, and contributes nothing to a company's bottom line. If, by eliminating one adjustment on a machine, setup time can be reduced, smaller runs, which minimize inventory and are more responsive to changing customer demands, become possible. That is a JIT innovation, as are requiring more frequent deliveries from vendors, which reduce a company's stockpile of raw materials or parts; moving machines involved in a process closer together (no forklifts ferrying dead inventory); and transporting parts to the point where they are needed only as they are needed (no piles of inventory waiting by each machine).

The point is to make sure that each person involved in the manufacturing process gets what he needs "just in time."

It may seem obvious, something that industrial engineers should have been doing all along, but, somewhere between Henry Ford I and us, tradition took precedence over innovation. Manufacturing, theory and machine and factory design became more set in their ways than circumstances warranted.

But JIT, which is generally regarded as a Japanese development (there it is known as kanban hoshiki, or "card system"), has recently become a hot, top-of-the-agenda topic in this country. Automotive Industry Action Group, a three-year-old productivity task force made up of Big Four representatives, has a JIT project team, presents seminars, and has developed a four-hour videotape series on the subject. And the American Production and Inventory Control Society has launched a JIT "crusade"; at its annual convention in New Orleans this past November, some 4,500 chief executive officers, plant managers, and inventory specialists attended standing-room-only addresses on JIT. "Just-InTime, Just-In-Time," one non-JIT lecturer groused to his wife over coffee. "No one came to hear my presentation."

A growing number of U.S. companies are making use of the approach. Omark Industries, a $300-million-a-year corporation based in Portland, Oreg., last year saved itself an estimated $7 million in inventory carrying costs with its own version of JIT, which it calls ZIPS (Zero Inventory Production System). T. D. Shea Manufacturing Inc., of Troy, Mich., a $4-million-a-year producer of plastic products for the automotive industry, calls its brand Nick-of-Time. And, at the Harley-Davidson Motor Co.'s engine plant in Milwaukee, where it is used to help restore the company's edge in the large-motorcycle market, it is known as MAN (Material-asNeeded). At least three universities are now offering courses in JIT.

"In the U.S., we've sold our soul to Murphy's law," admonishes Edward J. Hay, a JIT consultant with Rath & Strong Inc., of Lexington, Mass. "We believe that there's some mystical force out there that's going to do us in -- that's going to shut down a line -- and so we protect ourselves with inventory- The Japanese believe that human error, machine breakdowns and defective parts can be prevented and that inventory simply hides the problems and keeps companies from achieving their goals."

During the 1950s, Hay explains the Japanese, recognizing that they would have to upgrade their manufacturing expertise significantly if they were to make additional inroads in world markets targeted quality and the elimination of waste (read: unnecessary inventories). U.S. statisticians Edward Deming and Joseph Juran helped with the first, preaching statistical process and quality control (ironically, Japan's most coveted industrial honor is the Deming award for quality), while Taiichi Ohno, then a vice-president of Toyota Motor Co., assembled the fundamentals of JIT. Together, the two wreaked havoc with the U.S. automobile industry. "The Big Four," Hay recalls, "couldn't understand how the Japanese could import all of their raw materials, produce a quality car ship it 7,000 miles, and still enjoy a $1,200 to $1,500 cost advantage in the American market." When they sent study groups over to Japan to find out, they discovered that slave wages was not the answer.

"I first heard about JIT at a seminar at Ford Motor Co.'s headquarters three-and-a-half years ago," notes Hay, still relishing the memory. "When I came away from that, my feet didn't touch the ground -- it just made so much sense." While Ford and General Motors Corp. began spreading the word, many businesspeople, such as John L. Warne, president and chief operating officer of Omark Industries, were happening upon JIT during their own Japanese outings. The U.S. auto industry was not the only one threatened; quality had become a national priority in Japan, and JIT, which originated at Toyota, had spread to many of the auto maker's 220 suppliers, and beyond. Consumer electronics and personal computers would be next. "It's pure fear for survival that's driving our interest in JIT," says Robert W. Hall, associate professor of operations management at Indiana University Graduate School of Business and author of Zero Inventories.

Despite the surge of interest and enthusiasm, JIT remains, at the moment, a Big Business advantage; the companies using it most include the Big Four, Hewlett-Packard, Motorola, Westinghouse Electric, General Electric, Deere, and Black & Decker Manufacturing. The question now is, Does JIT have a major role to play in small business? The answer is a nearly unqualified yes. Apparently, size and type of manufacturing pose no particular obstacle to implementing JIT. "The philosophy applies almost anywhere," counsels Hall, observing that he has seen it work well in 20-employee job shops. As with any system, though, there are practical considerations and limitations. Henry Ekstein, president of Management Resources Inc. of Teaneck, N.J., would like to be able to offer JIT to more of the companies he advises, but finds it difficult to do so.

"Many of them would like to have it, but can't," he says. JIT works best, he notes, in repetitive manufacturing situations involving at least some of the following criteria: There are significant levels of inventory to begin with; demand and production can be forecast accurately; suppliers are located nearby, manufacture quality parts, and are willing to cooperate. "Controlling the vendor is a very tricky thing " Ekstein points out. Some companies, he admits, might be able to adapt JIT to only 20 items out of the 1,000 they carry in inventory. At the moment, he is doing just that with Ecker Manufacturing Co., a small storm-door and window company.

T. D. Shea bespeaks the benefits that even a small business can derive from Nick-of-Time. Edward J. Stenger, vice-president of operations, first heard about the system at a Ford seminar in November 1982, and, for the past six months, has been introducing it to the company's two plants, which employ 120 people. The steps have been small, but productive: A simplified setup procedure has reduced the setup time on one press by 50%; tote boxes and carts have been acquired to move parts more easily and in smaller quantities; an automatic stacking machine has been purchased (before, finished product sometimes spilled off the end of a line). And T. D. Shea is beginning to use kanban (the "card" in the "card system" of kanban hoshiki), a simple system that involves matching parts with numbered tags. The changes, which have cost Shea some $25,000 in capital expenditures, are expected to save it nearly $75,000. "It was a real eye-opener," claims Stenger. "We found ourselves analyzing operations that had been done the same way for 20 years."

A. P. Parts Co. of Toledo, a $160-million-a-year manufacturer of automobile exhaust systems for the OEM and replacement markets, found itself forced to confront JIT nearly two years ago. Honda of America Manufacturing Inc. set up an automobile plant in Marysville, Ohio, only 100 miles away, and A. P. Parts found that it couldn't compete, in quality or price, with the mufflers that Honda was importing from Japan. "That was the final straw," avers JIT consultant Hay, who worked with A. P. Parts.

One of the first things Hay did after he arrived at A. P. Parts was videotape and study the setup of equipment in the muffler-assembly area, looking for possible shortcuts. Working with the employees, he instituted changes that significantly improved the efficiency of the operation. Workers, who had set up and operated machines in old, sometimes self-taught ways, were choreographed precisely; if setting up a machine required 17 moves, then 17, not 25, was the number that should be used. Rather than make adjustments and subjective determinations each time a setup was done, the correct settings for all products were recorded, and the resulting chart gave the correct setting to use for each individual product. Work that could be done before or after a run was identified, and done then. By-the-book operation and innovations on machines eliminated the need for most adjustments and for tools and tape measures. In one case, the setup time for an individual machine, was slashed from 7 minutes, 13 seconds to 7 seconds; and, on a second machine, the setup time was reduced from 26 minutes to 3 minutes.

Hay won't reveal how he did it -- ", a trade secret," he says -- but one of his other innovations is revealing. On one machine a pilot had to be changed for different product runs. With the help of A. P. parts's project team, the pilot was redesigned as a snap-in snap-out part. The switch-over time was reduced by 95%.

Together with other programs it introduced, A. P. Parts saw a 32% reduction in finished-goods inventory (worth $10 million), did away with the need for a 460,000-square-foot warehouse, and permitted the consolidation of two plants.

Omark Industries is even larger than A. P. Parts, but it has pursued its own vision of JIT (ZIPS) in small business ways, and is now regarded by many as the foremost U.S. practitioner. John Warne, the president of Omark, which has six divisions encompassing 21 plants, 17 of them in the United States, had heard rumblings about JIT years ago. Then, during a trip to Japan in 1981, he was introduced to one of JIT's more enthusiastic disciples, Yoshio Ohya, a general manager at Nippondenso Co., that country's equivalent of the Delco-Remy division of General Motors. "We visited a starter motor plant, and Mr. Ohya spoke," recalls Warne. "He said, 'We have seven hours' worth of raw materials inventory, nine hours of work in process [the number of pieces actually being worked on at a given point in time], and eight hours of finished goods. . . . We feed 10 auto plants, making deliveries every two hours, and, in the last five years, we've held up one product line for 30 minutes.' " The claims bordered on the miraculous, and Warne was appropriately skeptical . . . until he visited the shop floor.

"I'm an old tool-and-die man," he explains, "and once I got out into the plant, I understood what was really happening." The revelation was nothing short of amazing. "It was like being born again. I thought, God Almighty, this is the secret of their productivity."

After Warne's pivotal visit to Japan, and an enthusiastic reception of the idea by Omark chairman and CEO Ted Smith, a task force was formed to study the system. Recognizing that employee involvement at every level was crucial to implementing ZIPS, management ordered 600 copies of Study of Toyota production System.- From Industrial Fngineering Viewpoint--the bible of the new faith -- by Shigeo Shingo, formerly a member of Toyota's management and now president of the Institute of Management Improvement in Tokyo. In November 1982, Omark passed them out.

"It isn't the easiest book in the world to read," confesses Larry White, plant manager of Omark's Oregon Saw Chain plant in Portland, which manufactures $50 million worth of saw chain, saw guide bars, and accessories a year. Intimidated by the challenge of Shingo's "Janglish," a hybrid of Japanese and English, eight members of Oregon Saw's management divided and conquered. Each was responsible for reading and reporting on one chapter. "It took us 32 hours to get through," White recollects with a trace of pain.

At the 20 other Omark plants, which manufacture everything from log loaders to railroad-tie fasteners, similar scenarios began to unfold. Each plant was asked to evaluate the relevance of ZIPS to its operation, and, if it wished, was authorized to proceed with a program of its own design. Smaller plants, with their greater flexibility, were the first to launch pilot projects, nnd Omark's smaller operations -- a twist drill plant in Mesabi, Minn., and a sporting-goods plant in Oroville, Calif., among thein -- now lead the pack in terms of being "ZIPSed." The results reported by the various plant managers add up to a Guinness Book of World Records--like chronicle of improvements:

* At the Mesabi plant, large-size drill inventory was cut by 92%, productivity increased by 30%, scrap and rework (imperfect pieces that must be redone) dropped 20%, and lead time (the time it takes to go from order to finished product) was slashed from three weeks to three days.

* At the Oroville plant, which produces reloading equipment for firearms, inventories were reduced by 50%, lead time plummeted from six weeks to two days, and lot sizes (the number of parts produced during an individual run) were progressively cut from 500 to 30. In less than one year, some 200 products, more than 70% of the plant's entire line, were ZIPSed.

* At pilot operations in a log-loader plant in Prentice, Wis., total inventories were reduced by 45%, and, thanks to a new layout, parts that had once traveled 2,000 feet now move only 18 inches to get from one machine to another. As distances shrank, and the need for long runs disappeared, lead times fell from 30 days to a matter of minutes. Work in process went from 60 pieces to just 1; rather than being manufactured in 60-unit lots, then waiting, frequently for days, before being transported to the next machine in the process, pieces proceeded from step to step one-by-one. Now when a problem yields defective pieces, it can be corrected before 60 bad parts have been churned out.

Oregon Saw Chain Division, the largest and most cumbersome of Omark's divisions, was slower to get off the ground, but has performed equally well. The division had always considered itself a market leader, and enjoyed its stature in the industry. It owned more than 50% of the pie in all the areas it sold to, and supplied most Japanese chain-saw manufacturers. Its plant, built in 1965, and twice expanded to a total of 300,000 square feet, was efficient and attractive. The company was equally proud of its technical and manufacturing expertise. It purchased high-speed presses and upgraded its production lines during the 1970s when the home chain-saw market exploded, and it had a knowledgeable and innovative tool-design department. It created its own control systems and hard automation, burned its own chips, and wrote its own programs. A Cincinnati Milacron Inc. robot stood sentinel on its shop floor.

But, like most U.S. manufacturing concerns, the company had proceeded on the basis of unexamined and, sometimes, antiquated beliefs. Machines were to be kept running as much of the time as possible; the number of pieces manufactured during a run must justify long setup times; it was best to consolidate similar operations in one department; employees should specialize; inventories should be adequate to cover problems with suppliers or unexpected customer demand. The philosophy, White explains, shaking his head in disbelief, boiled down to, "Let's produce more, more quickly, with fewer men."

ZIPS, he notes with an honest appreciation for the system's simplicity, demanded that he take a fresh look at his plant and ask a single question: Can't this be done a little bit better?

"Traditionally, American industries have considered shutting down a line the worst possible thing that could happen," observes consultant Hay. "The Japanese have taught us that that's the third worst. The worst thing that can happen is that bad products get made; the second worst is that inventory hides a plant's problems and inefficiencies; the third worst is interrupting the flow of production."

When White took a look around, he saw a lot of room for improvement. An employee task force set up to identify likely targets came up, however, with a proposal for two modest projects, but projects with aggressive goals: to reduce, within three years, the lead time required to produce rivets used in assembling chain from six weeks to one, and the lead time for three-quarter-pitch chain, the largest made in the plant, from eight weeks to one. "We wanted to be able to produce a week's worth in one week, with no parts left over," says White. The work force, which had received some 40 hours of ZIPS-related instruction, was turned loose on the task, and changes, large and small, began to happen.

"One of the first things we did," White recalls with a chuckle, "was arbitrarily eliminate a week's lead time. We told people, 'We're going to take a week's worth of parts out of the system and see what happens.' "The piles of parts sitting in front of, behind, and in between machines were trimmed by that amount, and what happened was that things ran smoother. "People asked that we try taking another week's worth out." After that, ZIPS spread throughout the plant's operations "like an amoeba."

After five months of ZIPS, Oregon Saw had achieved its three-year goals -- and then some.

At the same time that it was reducing lead times, the company had hoped to see a 40% reduction in inventory in the pilot programs; after five months, it had witnessed a 55% plantwide reduction. Now, it no longer even bothers setting up pilot programs: It just lets ZIPS happen. Before ZIPS, White estimates, he spent about 25% of his time implementing changes. Now, he spends fully 50% of his workday overseeing alterations.

The massive inventory reductions at Oregon Saw also freed up large blocks of time. "That gave us an opportunity to take a closer look at what we were doing," notes White with satisfaction. The company's relationship with vendors immediately came under scrutiny. More frequent vendor deliveries is one of the highest-profile aspects of Just-in-Time. The novelty of daily or hourly drop-offs is a seductress the American press finds hard to resist, but, in the minds of many practitioners, it is the crowning glory of JIT, and best left until late in the game.

"My feeling," says Hay, "is that you put the burden on your suppliers last. . . . A lot has been written, mistakenly, about auto makers demanding JIT deliveries, thereby pushing their inventory onto suppliers." That, in the long run, would make little sense under the JIT philosophy, which argues on behalf of close, frequently monogamous, relationships with vendors. Suppliers, saddled with extra inventory to cover a customer's needs and forced to make more frequent deliveries, would inevitably be forced to raise prices, thereby offsetting any advantage. In Japan, where distances are less formidable -- 80% of Toyota's 220 suppliers are located within 50 miles of its plants -- and where close customer-vendor relationships already exist, frequent deliveries are feasible. In the United States, the auto industry has more than 28,000 individual suppliers, located from coast to coast. Even so, some of them have begun to prepare for stepped-up delivery demands. A. P. Parts, which once made mufflers in Grand Haven, Mich., consolidated that plant with one in Toledo in order to be closer to its OEM market. "Draw a 300-mile circle around Detroit," suggests Michael J. Cicak, the former vice-president of manufacturing for A. P. Parts. "If you're not inside that circle in five years, you're not going to be a supplier."

At Oregon Saw, deliveries weren't increased, they were simply made more efficient. "We had plenty of space in our initial staging area," explains White, "but, suddenly, it seemed stupid to stage materials at all. Why not put them where they were going to be used? Now, as soon as it comes in the door, it goes to the area where it's needed."

The portland plant has few major vendors -- 12 for strip steel, 6 for wire (used in the manufacture of saw files and rivets) -- and, rather than hound them about deliveries, it began to work with them on quality, the second major concern of Japanese manufacturing. "The most massive rock that surfaced when our inventory levels dropped," says White, coming back to the fisherman-in-the-middle-of-the-lake analogy, "was quality."

With more time to analyze incoming raw materials, the company began to find problems that had gone undetected. In one case, it spotted a defect known as lamination, caused by a thin layer of impurities in the steel. "In the past, it hadn't shown up until the final operation -- you put in all of the value, and then the piece fell in half." Oregon Saw alerted its vendor, which was then able to trace it, not to a problem on its own rerolling line, but to one of its suppliers. able to detect a change in an annaeling process used by a wire supplier which made the wire unusable. It came as news to the supplier, and temporarily shut down the supplier. "But," White hastens to add, "in the past, we'd have wound up manufacturing bad product." Now the] vendor provides Oregon Saw with sample "coupons" of its steel, enabling the comany to spot potential problems before the wire is even shipped.

Oregon Saw has also attempted to make life easier for its suppliers. Although it once was secretive about production forecastss, it now shares information, so that vendors can predict more accurately their own necds and manufacturing schedules. "Before, we worked at arm's length with suppliers," White. says with regret, "but now we want a strong marriage relationship." That relationship extends, if a supplier is interested, to assistance with implementing ZIPs. "We don't want them fooling around with hourly deliveries. . . There's plenty of room for improvement elsewhere."

The emphasis on quality carries over to plant's own operations. Under ZIPS, as soon as a defect is spotted, the machines grind to a halt. And because work in process has been sharply reduced, the loss is kept to a minimum. When a metallurgical problem surfaced at Omark's Australian railroad-tie-fastener plant, management pulled the plug and called Michael Rowney, a technical expert in metallurgical engineering. "The only work in process was five pieces on the mahines, each valued at $1.50," recalls Warne "Under the old American batch approach to manufacturing, we would probably have had 50,000, worth $75,000. Under ZlPS, we only had to ditch $7 worth of bits."

White, an industrial engineer by training, offers a tour of the Portland facility. After one year of ZIPS, the plant has reduced its total finished-goods inventory by 50%, and its work-in-process inventory by 50% (its raw-materials inventory thus far has been left relatively untouched). In its pilot projects, it managed a 6% reduction in net manufacturing costs, and it freed up more than 30% of its floor space, including 50% of its 100,000-square-foot warehouse. It recently launched a Single Minutes Exchange of Die program, which has targeted a changeover time of under 10 minutes. It now turns over its inventory seven times a year, up from two to three times before ZIPS, and it has improved its inventory efficiency (value of work in process divided by units produced) by more than 50%. Oregon Saw and Omark remain properly proprietary about some of the results, but White indicates that he would be hard-pressed to find an index that doesn't look good.

David Pinch, vice-president and controller of Omark, notes that the company as a whole has witnessed an overall reduction in inventories of 25% -- down $20 million from $80 million two years ago. That translates, he explains, into a savings in carrying costs of $7 million a year. In ventory, an Omark slide presentation on ZIPS points out, is an insidious expense, devouring 3% of a product's value each month. Thus, saw chain kept on the shelf for one year "has lost all potential profit." Omark is committed to the concept of ZIPS: It has reaped as-yet-unmeasured rewards for a first-year investment of only $200,000, significantly less than its capital expenditures for the same period would ordinarily have run. It has earmarked nearly $800,000 for the second year, and has no doubts that it is money well spent.

ZIPS isn't always an easygoing process, however. A few automobile manufacturers have already encountered Just-in-Time roadblocks. General Motors was obliged to close down two plants because of a shortage of computer chips; and Chrysler Corp. had to switch off all of its assembly operations the day after a crucial stamping plant went out on strike; neither had inventories to get them through. And, at Ford, one attempt to reduce setup times ran head-on into the company's traditional cost accounting and bonus systems, and was shelved for the foreseeable future.

But the Big Four apparently remain sold on the philosophy of UT. "They can't afford to ignore it," asserts Professor Hall. "They'd do so at significant risk." Even at Omark, some plant managers objected to it initially. At Oregon Saw, the heating and ventilation systems don't work very well now that so many departments have been consolidated, and, occasionally, a line has been forced to shut down. But even that -- once the nemesis of any plant manager -- is now regarded as another opportunity, a chance to identify problems and solve them. One notes a real sense of regret in White's voice when he admits that there hasn't been a major shutdown. "But I'm sure that we'll see one somewhere down the line," he adds. "And I'm ready and willing to face that. . . . You might even say that we're looking forward to it."