Kanban

What you choose to call objects, ideas, positions, methods, tools, etc., makes a difference in your ability to lead and change the way your organization works. Engineers often believe that it makes no difference what words you use as long as what you do works; marketers know better. They know that “sushi” sells better than “raw fish,” “Lean” better than “TPS,” and “Black Belt” better than “Staff Statistician.” The “Toyota Way” or the “Toyota Production System” sound like appealing approaches; “toyotism,” like yet another plot to enslave workers. Words matter; they engage, motivate, and inspire, or they confuse, offend, and alienate. Following are a few thoughts on Manufacturing Wordsmanship, the art of naming manufacturing improvement tools and concepts:

• The contrary body of opinion
• How terms are used
• Methods of word formation
  • Borrowing from everyday life
  • Borrowing foreign words
  • Combining roots from dead languages
  • Creating acronyms
  • Naming things after people

The contrary body of opinion

In the TPS Principles and Practices discussion group, Dermot Freeman‘s summarized the engineering perception as follows: “Our (insert customers or clients or bosses or colleagues – whichever is appropriate for you) don’t really care what the tools or methods are called as long as they achieve the correct result.” Further in the same thread, Sid Joynson quoted the taoist philosopher Zhuangzi (莊子, 389-286 BCE), as saying:

“The fish trap exists because of the fish. Once you’ve gotten the fish you can forget the trap. The rabbit snare exists because of the rabbit. Once you’ve gotten the rabbit, you can forget the snare. Words exist because of meaning. Once you’ve gotten the meaning, you can forget the words. Where can I find a man who has forgotten words so I can talk with him?”

Sid also quoted Fen Yang, a follower of Zhuangzi 1200 years later, known in Japan as Zen Buddhism founder Funyo Zensho (汾陽善昭, 947-1024 CE) :

“When you are deluded (don’t understand) a thousand books are not enough. But when you have realised understanding, one word is too much.”

Attractive though these  pronouncements may seem, in manufacturing, we not only keep the trap, the snare, and the words, but also their models, revision and serial numbers… It is not a world where understanding exists independently of language. Understanding in manufacturing is not a matter of feeling or enlightenment, but of using words, drawings and numbers to turn materials into products with machines and tools. In Training Within Industry (TWI), the system developed in the US during World War II and later adopted by Toyota, being able to do a job is not enough. To be certified as proficient in it, you must also explain it step by step in words. If you have been doing the same repetitive work for a while, you certainly forget the words and go on automatic. But you have to pay close attention to the words to respond to changes that occur, for example when improving an operation, implementing engineering revisions in the product or process, or making custom-configured products.

How terms are used

An effective term serves not only to communicate with others, but also in your own mind as an address in memory to which you attach everything you learn about an object. If you have a name for it, you will notice it when you see it, and you will remember its characteristics. When you learn a new word in a foreign language, you suddenly start hearing native speakers use it in a variety of contexts. Knowing that this object on the shop floor is called a “broaching machine” lets you attach the broaching process, tools, times, and quality issues to this name and remember them. Conversely, terms that are difficult to remember, ambiguous, offensive, or inconsistently used —  or sound too much like other terms — sow confusion and impair both communication and learning. What you are trying to achieve on a manufacturing shop floor is free and open communication. Mutual understanding is the primary goal. While a challenge, it is an easier one than the restricted or confidential communication characteristic of many other activities. There, in addition to encoding and decoding for electronic communications you still use code words, cant, or jargon to confuse unintended audiences when speaking to someone next to you.

Methods of word formation

Different objectives call for different methods in choosing your words. As an example, I am going to focus on the vocabulary used in Japanese for the Toyota Production System (TPS), and its translation into English. To get the translations right, we need words that not only mean the same but resonate with local manufacturing people as well as the TPS terms do in Japanese. The worst we can do in translating TPS is over-intellectualize it. In the US, factory people rarely read the Harvard Business Review.  We should not translate  TPS terms into words that would fit in this journal. The TPS vocabulary doesn’t include entropy, DNA, or Value Stream.

Borrowing from everyday life

To communicate with auto workers, TPS uses concrete words, and sometimes imagery like “spaceships” (宇宙船) and “pirate ships” for different types of fixtures. “Spaceships” hover over workstations, providing tools that hang over it; “pirate ships” are level with the work station and “board” it on its side, moving through the station along with the assembly line. These words are borrowed from everyday language. Kanbans are the shingles stores put out in the street. Andons are lanterns. That “Kanban” is borrowed from shop signs may not make it easier to understand how the system works, but it is a revealing choice in word formation: grab a word from common, everyday language and extend its meaning in a way that sort of makes sense to the people who are expected to use it.

On its own website, Toyota gives “ergonomic seat” as translation for the raku-raku seat (楽々シート). The meaning is accurate, but the connotation different. “Ergonomics” is a word made up from Greek roots; “raku-raku,” in Japanese, is colloquial. Its connotation would have been better rendered by “easy seat” or “comfy seat.”

When borrowing a word to give it a new meaning, make sure it comes from a domain that is sufficiently different to avoid ambiguity and confusion. No one will confuse a Kanban used in production with a shop sign on the street. On the other hand, calling a shelf on which you store parts a “Kanban” creates confusion, because it is like using the same word for a theater ticket and for the seat it is for. You can contrast this with importing foreign words, using acronyms or mashing together Greek and Latin roots. Acronyms are the way to go if you want to create a code that excludes the uninitiated. Foreign words work that way too, but not with native speakers of these foreign words.

Borrowing foreign words

TPS also borrowed words that are foreign to Japan, such as:

1. “Takt,” which is the German word for a musical bar, strokes in car engine or, as David Anderson noted, a regular interval between trains. 
2. Phrases built from foreign words, like “just-in-time.”
3. Acronyms built from oddly used foreign words, like “SMED.” I don’t recall a native speaker of English referring to an interval of less than 10 minutes as “Single Minutes.” In addition “Exchange of Dies” is wording that is specific to stamping presses, for which the method was originally developed, but you apply SMED to many machines that don’t use dies.

Combining roots from dead languages

European scientists mashed together roots from Greek and Latin to concoct new words like “entropy” or “isotope” for Greek and “tyrannosaurus rex” for Latin. Sometimes, they even mixed both Greek and Latin in words like “television” or “heterosexual.” Why didn’t they choose words from their own vernaculars? Today, using Greek and Latin roots shows off your classical education and excludes those who haven’t had one…. It is difficult to imagine another reason to refer to a periodic maintenance task as being “isochronal” in the US today. 200 years ago, a European scientist pushing the use of his own language would have been in trouble with foreign peers. Drawing on dead languages that scientists from Italy to Sweden and Portugal to Russia had studied as kids was a safe way to avoid any appearance of cultural arrogance. In Japan, the equivalent of Greek and Latin roots is Chinese roots, and the Japanese have been at it for 1,500 years, thus creating almost half of their vocabulary. In TPS, this has given us words like Heijunka, Jidoka, or Jishuken. Jidoka (自働化) is a case in point, because the Toyota version sounds exactly like the standard word for automation (自動化), but differing by two strokes added to the middle character, changing it from 動 for “moving” to 働 for “working,” the two strokes added being the radical for human (人). This is as brilliant as it is untranslatable.

In Japan, as in the US and Europe, this mode of word formation is perceived as solemn, official, and old-fashioned. It is still done, but not by anyone who wants to be cool. Combining fragments of English words, as in the “CHAdeMO” charging stations for electric cars, is cool. “CHA” is for “charge,” “MO” for “motion,” and the “de” is there to make the whole word sound like a fragment of a Japanese sentence for offering tea. I don’t know of any such words in the TPS lexicon, where coolness is not pursued.

Creating acronyms

Acronyms in everyday use include OK, laser, snafu, or flak; in manufacturing, JIT, WIP,  SMED, TQC and TPM. They are prized as convenient abbreviations but, unlike everyday words used metaphorically, they hurt communication by having no connection with their meaning. Too many sprinkled in a document or speech can make it unintelligible. It is striking how people qualify acronyms to reconnect them with meaning. Instead of “ATM,” they will say “ATM Machine,” even though the “M” in “ATM” already stands for “Machine.” Likewise, in Manufacturing, you hear of “MES Systems,” where the “S” in “MES” already stands for “System.” Acronyms work best when built from words that provide a clear and accurate definition, as is the case, for example with WIP. When creating acronyms, you should also be careful to avoid the following:

• Acronyms that are jokes or can embarrass people in any of the languages used in your company. The “Project Information System” (PIS) was unfortunate, and so are the “Semiconductor Equipment Communication Standards” (SECS).
• Acronyms in which some letters stand for another acronym. They just take too long to decypher.
• Acronyms that are an exact match to an existing word. It is a good idea to make acronyms pronounceable, but an exact match with an existing word creates confusion, particularly in Google searches.
• Acronyms that already exist with a different meaning.  It is particularly confusing when it is in the same context. In one plant, for example, successive inspection was called “Touch Quality Control,” abbreviated TQC, which outsiders understood to mean “Total Quality Control.”  Even when used in different contexts, search problems still occur. “5S,” for example brings up not only items about visual management in factories, but about 5-year old children in kindergartens and about the latest version of the iPhone.

Sometimes acronyms are used as mnemonics rather than abbreviations. TIMWOOD, for example, is a way to remember Ohno’s list of waste categories. It stands for “Transportation, Inventory,Motion, Waiting, Overproduction, Overprocessing, and Defects.” Acronyms are sometimes used to deliberately exclude the uninitiated. When Gus Pagonis, the general in charge of logistics for the first Gulf War, on p, 90 of his book Moving Mountains, discusses the Time-Phased Force Deployment List or TPFDL, pronounced “tipfiddle,” the purpose of using this term is obviously not to be universally understood.

Naming things after people

I can’t think of any TPS concepts that are named after people. On this subject, Frederick Stimson Harriman wrote the following:

“The Japanese practitioners that I know won’t even call a Gantt Chart a Gantt Chart. I have often heard Japanese call Gantt Charts 予定表 YOTEIHYOU (“schedule” or “plan”). What is called a “Pareto Chart” in the US among TPS/Lean practitioners is often called simply バーチャート BAA CHAATO “bar chart.” What many people call an “Ishikawa Diagram” is often referred to in Japanese as 特性要因図 TOK– USEI YOUIN-ZU literally “characteristic factor diagram” but “cause and effect diagram” is not an inaccurate translation. The Consultants I have worked with expect it to be translated either as “fishbone diagram” or “cause and effect diagram.” Of course these are all tools that are not specific to TPS, but are often used by practitioners in problem solving, etc. From my experience, Japanese have a general aversion to naming things after people. Note that Americans have no trouble immortalizing Gantt, Pareto, and Ishikawa, and it seems to be something that Japanese feel more comfortable avoiding. Even the name Toyota of “Toyota Production System” is for the company and not the family, which continues to pronounce its name ‘Toyoda.'”

This comment about the Japanese not naming things after people, besides being a challenge to find exceptions, raises the question of whether it is a good practice. Many Japanese companies are named after founders, like Matsushita or Honda. 20 years ago, a Japanese colleague of mine used a system of stickers to make flow charts that was called, as I recall, the Kitagawa method; my son took violin lessons on the Suzuki method; Norman Bodek is promoting  the Harada method; and, like everywhere else, Japanese scientific discoveries and theories are named after the researchers who are credited for making them, like the Ito integral.  But none of that is TPS. Generally, we can think of several good reasons to name things after people:

1. It is gratifying to have people use something you invented and call it by your name.
2. It tells everyone in the organization that they have a chance at immortality.
3. It connects the thing with the context in which it was invented.

But there are downsides as well:

1. A person’s name tells you nothing about what the idea is or what the thing does. “Pressure cooker” or “schedule chart” does.
2. Individual recognition can create jealousy and conflicts about attribution, which discourages teamwork.
3. People’s names are not always easy to pronounce.

Gantt charts were actually invented by Karol Adamiecki in Poland a few years ahead of Henry Gantt. Adamiecki called them “harmonograms.” So why do we call them “Gantt charts” and not “Adamiecki Harmonograms”? Who has time for eight syllables when two will do, especially when they roll off the tongue like “Gantt Chart”? Adamiecki may have been first, but he had the wrong name.