|
|
How much of lean is lost in translation?
In Sofia Coppola's movie “Lost in Translation,” Bill Murray’s character comes to Japan to star in a whisky commercial. During the shoot, the director gives him detailed instructions in Japanese, telling him to treat the glass of whiskey like an old friend and look at it longingly. The interpreter’s translation of the entire director's speech, however, is “He wants you to turn. Look in camera.”
Likewise, in lean, words shuttling back and forth between languages take on different meanings, often with the result that the original is lost in translation. This article examines TPM, standard work and 5S as three specific cases where translation errors have had serious consequences.
TPMWhile TPM is an English-language acronym, it was coined in Japan by Seiichi Nakajima in 1971. It reflects Japanese usage of English and is anything but self-explanatory. If you spell it out as “total productive maintenance” and ask an uninitiated native speaker of English what it means, you are unlikely to get an answer that is consistent with Nakajima’s intent. Instead, productive maintenance is likely to be interpreted as maintenance conducted with as few technicians as possible, and total to mean thorough.
In Nakajima’s own words, productive maintenance originated at GE in the United States and is aimed at enhancing overall equipment effectiveness through the elimination of losses due to breakdowns, setups and adjustments, microstoppages, slowdowns, production of defectives and reduced yields during startups. Oddly, the English-language literature covers preventive, predictive, condition-based and reliability-centered maintenance, but not productive maintenance, except in the TPM context.
But what about total? We tend to think of it as meaning complete, thorough or all-encompassing. In the Japanese context, however, total stands for “zenshateki,” meaning “company-wide” or, even more specifically “zenyinsanka,” which means “with participation by all employees.” So it is not about doing the job thoroughly but about involving everyone. TPM as a whole, when spelled out in Japanese, is “zenyinsanka seisan hozen,” which literally means “maintenance of production with participation by everyone.”
As can be seen in the Appendix below, none of the definitions of TPM that I could find in the English-language literature by authors other than Nakajima accurately reflects his thinking about participation by everyone. Where it is mentioned at all, it refers to the involvement of production operators in autonomous maintenance, but Nakajima means everyone, from the CEO to the janitor. According to Ron Moore, more than half the TPM programs fail because management treats it like just another maintenance program and delegates its leadership to the maintenance department.
Standard workFor different reasons, “standard” is also a word that generates miscommunication because it means different things in different languages. As described in Liker and Meier’s “Toyota Way Fieldbook” [2], standard work is a repeatable work method with defined expectations, documented instruction sheets, work combination charts and cycle balance charts, and supported by shared tooling, visual controls, procedures, boundary samples and TWI.
It is very specific and limited in scope, and meant to ensure that, at any given workstation, the same process is executed the same way every time, regardless of shift and operator assignment. This is needed for quality assurance, capacity planning and as a starting point for improvement. It is most desirable and most achievable within one plant because the work it supports is done within each shop floor. It is also desirable but much more difficult across multiple plants, and Toyota is currently struggling to achieve it worldwide.
But it is not, for example, about mandating a common cell layout across all plants, regardless of product and takt time, or about demanding that specific types of kanbans be used everywhere in the company regardless of flow requirements. This level of standardization is ineffective at improving performance, particularly when undertaken at the start of a company’s lean transformation, when its managers and engineers have yet to master the tools of lean, and it stifles innovation. Yet it is exactly what many corporate lean groups are pursuing, ramming half-baked “standard” designs down the throats of plant managers whose teams already know better.
While the Japanese word “hyojun” translates literally to “standard,” cultural attitudes toward standards vary across the world and interfere with a proper understanding of the role of standard work in lean. In Germany or France, for example, standards are set top-down by committees, ruthlessly enforced and nearly impossible to change, with government agencies like DIN regulating everything from the dimensions of manhole covers to the size of steel balls on motorcycle brake handles. Standardization is viewed as intrinsically good and a bad standard as better than no standard.
While this is not well known, Japan is at the opposite extreme, with half the country on 50 Hz power and the other half on 60 Hz, with no standard gauge for railroads, no standard keyboard for word processors and, back in the early days of the VCR, both Betamax and VHS. Standards emerge bottom-up through competition. The winner becomes the de facto standard and, as the industry matures, becomes official. In the United States, the military and the industries it influences are close to the German/French model, but the rest of the economy, and particularly high-technology industries, are much closer to the Japanese model, with “official” standards often issued by professional societies, such as IEEE (the Institute of Electrical and Electronics Engineers Inc.), SAE (Society of Automotive Engineers or SEMI® (Semiconductor Equipment and Materials International) rather than by the government.
The German/French approach has had successes like the metric system, and failures like the ISO/OSI communication protocols that were swept aside by the Internet in the early 1990s. On the other hand, the standards that emerge from industry have flaws that are due to the absence of a systematic design. For example, the ISO/OSI protocols had built-in data security, but the Internet protocols didn’t and had to undergo painful retrofits to support electronic commerce. The free-for-all Japanese approach to standards has not prevented the electronics industry from being successful, but the use of special gauge tracks has doomed the Japanese high-speed trains as an export product.
When applied inside a company, the German/French approach leads, for example, to designating a committee to come up with a standard design for manufacturing cells, mandating that all factories apply it and writing up violators. The rationale given is that, once implemented everywhere, this standard design can be improved, but by whom? At that point, the plant teams have not developed the required skills, and corporate insistence on compliance has killed their motivation.
Likewise, structured reports that fit on a single A3 or 11 x 17 in. (28 x 43.2 cm) sheet are a tool to prompt teams to generate all the required information to support a decision, and to help readers find it. The top-down imposition of a single format early on, however, often results in reports that do not meet most needs and in teams spending hours trying to fit their square pegs into this round hole instead of solving their problems. Standard formats are needed for plant-wide or company-wide deployment but not for pilot projects, and, in fact, pilot projects are an opportunity for the organization to develop its own single-page reports. The most effective can then be made into standards.
5SIn the case of 5S, attempting to translate five Japanese words that start with S into five English words that also start with S turns a difficult challenge into a hopeless one. Whether it comes out as “sort, straighten, sweep, standardize and self-discipline” or as “sort, straighten, shine, systematize and sustain,” it just doesn’t match the original. For the first three Ss, the mistranslations are relatively harmless because the correct meaning can be explained; the last two, however, are critical because it is where implementations fail. If we go back to the original Japanese words, the issues are as follows:- Seiri is about removing from the shop floor all items that are not routinely needed; disposing a rusty fixture that hasn't been used in two years isn't sorting.
- Seiton is tidiness, in the sense of having assigned, easily accessible and labeled locations for all routinely used items. While it is one of the possible meanings of straighten, it is not the first one that comes to mind.
- Seiso is cleaning both the machines and their surroundings. Mopping up an oil spill with a rag is not sweeping, but it isn’t shining either. The key point is that the action of cleaning makes operators notice frayed cables, leaks and broken signals that they might otherwise overlook.
- Seiketsu. While Seiton and Seiso respectively refer to the actions of tidying up and cleaning, seiketsu designates the resulting cleanliness and neatness, and could seem redundant. What it really designates is the management actions needed to reduce the first three Ss to daily practice. It is about making the first three Ss happen every day by management enforcement: audits, scoring and so on.
- Shitsuke. Shitsuke means discipline, with the connotation of upbringing and nurture. If you have been well brought up, you are said to have good shitsuke. In the context of 5S, this means a state in which the behaviors have become second nature to participants, and they sustain them even in the absence of enforcement, like children who no longer have to be told to brush their teeth. T. Osada describes it as “doing the right thing as a matter of course.”
In most companies that undertake to implement 5S, managers understand neither seiketsu nor shitsuke, as reflected in their use of expressions like “We’ll 5S this area before the visitors come,” or “We’ll run a 5S event.” Such actions can produce a 5S Potemkin village that may fool external auditors, but won't implement 5S.
ConclusionThe above examples show that the translation losses on many aspects of lean are not minor. They are in fact as serious as the interpreter’s mistakes in Sofia Coppola’s film. If the lean movement were more successful at transforming the whole manufacturing sector, we could ignore these losses. While we can’t expect all manufacturing managers and engineers to decipher the original texts, the consultants, academics and other thought leaders should at least be more careful. If Toyota were not the best car company in the world, no one would pay attention to lean. As we implement it in industries other than automotive, we must adapt our toolkit, but it behooves us to make sure that it remains consistent with the underlying principles that have made Toyota successful. And this requires us to pay more attention to the ways we and our Japanese colleagues mistreat each other’s language.
Appendix A: Definitions of TPMTPM: A series of methods, originally pioneered by Nippondenso (a member of the Toyota group), to ensure that every machine in a production process is always able to perform its required tasks so that production is never interrupted.
Source: Lean Advisors Inc.
TPM (total productive maintenance) is a maintenance philosophy designed to integrate equipment maintenance into the manufacturing process. The goal of any TPM program is to eliminate losses tied to equipment maintenance or, in other words, keep equipment producing only good product, as fast as possible with no unplanned downtime.
Source: Vorne Learning Center
Total Productive Maintenance (TPM) is maintenance activities that are productive and implemented by all employees.
Source: Wireman, T. "Total Productive Maintenance” 2nd ed. New York: Industrial Press, 2004, p1.
Total Productive Maintenance, an offshoot of the Toyota Production System in Japan in the late 1960s, is a strategy to improve equipment performance that links proven maintenance elements together to achieve a high level of equipment effectiveness. The six key elements of TPM include:- Improving equipment effectiveness by targeting the major losses.
- Involving operators in the daily, routine maintenance of their equipment.
- Improving maintenance efficiency and effectiveness.
- Training for everyone involved.
- Life cycle equipment management and maintenance prevention design.
- Winning with teamwork focused on common goals.
Source: Arabe, K.C. “Total Productive Maintenance: Increasing Equipment Effectiveness.” ThomasNet IndustrialNewsRoom, Jan. 18, 2001.
Productive maintenance is when the result of maintenance of equipment is measured (normally in economic terms, for example, cost-benefit analysis, and the result is positive (for instance, tuning your car regularly will not only prevent damage, it will also save fuel costs. If the fuel saving is superior to the tuning costs, that is productive maintenance).
Source: Asia Productivity Organization P-Glossary
Appendix B: Definitions for standard workStandard work: A precise description of each work activity specifying cycle time, takt time, the work sequence of specific tasks and the minimum inventory of parts on hand needed to conduct the activity.
Source: Lean Advisors Inc.
A detailed account of standard work within the Toyota Production System can also be found in Chapter 6 of Liker and Meiers’ book, “Toyota Way Fieldbook” [2].
Appendix C: Additional Sources[1] Moore, R. “Selecting the Right Manufacturing Improvement Tools.” Amsterdam: Elsevier, 2007, p174.
[2] Liker, J. and Meier, D. “The Toyota Way Fieldbook.” New York: McGraw-Hill, 2006, pp111-140.
[3] Osada, T. “The 5S's: Five Keys to a Total Quality Environment.” Tokyo: Asian Productivity Organization, 1991.
Appendix D: Illustrations of Japanese approach to standards
About the authorSince 1987, Michel Baudin has consulted on lean manufacturing implementation for such clients as Honda of America, Unilever, Boeing, Hewlett Packard, Intel and National Semiconductor. His prior experience includes serving as director of the Menlo Park Technology Center of Teknekron Corp., giving technical support for CIM installations in Japan on behalf of Consilium Corp., designing the MS/X OnTime production scheduler marketed by Tyecin Systems and implementing the OPT scheduling system in two General Motors factories.
Copyright © 2010 Society of Manufacturing Engineers
SME Customer Service: 1-800-733-4763
|
|
|
|
 |
| May 10, 2007 Issue: |
|
|
