When Taiichi Ohno first came to tiny Showa Manufacturing in 1984, mighty Toyota was just at the end of a thirty-five-year process of diffusing lean thinking across the Toyota Group within Japan and was just beginning to spread it across the world, beginning at the NUAUNII plant in California.
Two of the basic lean concepts in physical production—automatic machine and line stopping whenever a mistake is made so no bad parts will be passed forward to interrupt the downstream flow (which Toyota calls jidoka) and a pull system so that only parts actually needed are made (which Tovota. calls Just-in-Time)—had been formulated by Sakichi Toyoda (the founder of the Toyota group) and his son Kiichiro Toyoda (first president of its offshoot, the Toyota Motor Company) in the 1920s and 1930s. However, these concepts of physical production were only linked and operationalized by Taiichi Ohno and his disciples beginning in the late 1940s. At the same time, Toyota was pioneering ideas on the organization of product development, supply chain management, and order-taking from customers, which ultimately constituted the complete Toyota system. Soberingly, Toyota was only able to make the historic leap to fully implement these ideas when it faced a deep crisis in 1950.
Looking back in the 1980s, Taiichi Ohno noted that “Companies making even a modest profit never use the Toyota Production System. They can’t. On the other hand, there are nearly bankrupt companies that implement the Toyota Production System to the fullest, knowing they won’t lose much even if it fails…. This is the advantage of a defiant attitude.” I
Certainly, loss-making Toyota did not have much to lose in the immediate post-war period and Ohno was master of the defiant attitude. When he was promoted to manager of the Engine Manufacturing Department at Toyota in 1948, and suddenly had the authority to make changes, he found a classic batch-and-queue operation with all machines of a type in one location. The shop’s performance was even worse than one might expect because other Toyota departments supplying the engine shop rarely delivered on time and then delivered only in huge batches. Therefore, the engine shop spent the first half of the month waiting for all of the necessary parts to arrive and the latter half working furiously to meet the monthly production quota.
It was soon after arriving that Ohno had his most fundamental insights. First, he noted that workers spent most of their time simply watching machines do their work and that many bad parts could be produced before they were discovered by inspectors from the Quality Control Department. He remembered Sakichi Toyoda’s self-monitoring looms (which he called “a laboratory in front of your eyes”) that used devices measuring thread tension to shut themselves down immediately if a thread broke and the loom began to make defective cloth. Using this idea as his inspiration, he quickly devised a set of simple limit switches and go/no-go gauges so that machines, once loaded, could do their work to completion without human intervention and stop working immediately if they detected an error in their efforts. With these simple devices added to conventional machine tools, it was quickly possible for one worker to superintend many machines and perform quality checking as well, intervening only to load machines (as in the chaku-chaku line just installed at Pratt & Whitney) and to deal with malfunctions.
Ohno’s second insight was that “when you have lots of inventory you are always one part short.” He decided that the problem could only be solved if each processing step went frequently to the previous processing step and picked up exactly the number of parts needed for the next increment of production. By adding the ironclad rule that the previous step would never produce more parts than the next step had just withdrawn, a rudimentary JIT system was put in place. The famous kanban cards were introduced in 1953 to formalize the system and make information flow smoothly backwards at the same rate products flowed forward. The quick changeovers of tools needed to permit the previous process to rapidly respond to the needs of the next process were first attempted in the late 1940s, but the dramatic ability to change even the most massive tools was not fully perfected until the late 1960s.
Ohno’s third insight was that machines should be moved from process villages into “cells.” There, in a horseshoe pattern, they would be placed in the exact sequence required by the part being made. By focusing on the needs of the object undergoing manufacture, rather than the maintenance needs of the machines, the traditional skill sets and work methods of the workforce, or conventional thinking about scale economies, he focused the value stream and eventually perfected the concept of “single-piece flow.” Note also that the introduction of single-piece flow eliminates much of the need for in-plant JIT linking departments and process villages. In addition, by adding or subtracting workers from a cell, Toyota could increase or reduce the rate of production to keep it exactly synchronized with the “pull” of the market.
Ohno’s insights and actions marked a fundamental departure from other Japanese firms in the post–World War II era (including arch-rival Nissan). Many companies focused on larger and larger high-speed machines grouped in process villages, eventually linked by MRP, or on elaborate, automated transfer and assembly lines linking dozens of manufacturing steps and increasingly employing robotics to eliminate human effort. The latter might be thought of as “high-tech” mass production and these methods were perfect for the high-volume production of standardized products, largely for export. However, such goods are today a vanishing species, and high-tech mass production is often a loser when confronted by a flexible lean producer that has introduced continuous flow through its entire value stream.
One of Ohno’s favorite sayings was that “Common sense is always wrong.” He viewed his life as an effort to reverse common sense—for example, the belief that batch production is more efficient—and find a better way. However, his temperament and the very notion that the “common” was wrong destined him for collisions with most of his colleagues and workers. From the moment he discovered that one worker could load and monitor as many as fifteen machines, and further concluded that machines needed to be arranged and rearranged in the sequence of production steps without regard for traditional skills, there was potential for conflict with the workforce. And from the moment he concluded that upstream departments should do precisely what the next department downstream requested precisely when it requested it, the life and work of managers all along the value stream was permanently changed.
Excerpted from page numbers 230-233 of ‘Lean Thinking’ by Womack and Jones