What's Wrong With the Rote Application of Lean Tools?

There is more to playing the piano than practicing scales, but you can't get there unless you do practice scales. Likewise, there is more to Lean than tools, but you need the tools. They are not sufficient, but they are necessary. Unless you learn them, you are depriving yourself of the benefits of clever tricks, methods, and analytical tools developed over 65 years. Each one doesn't necessarily take long to learn, but would to reinvent, and you don't have that time.

1. Rote application of tools

When discussion rote application, you need to distinguish between learning by rote and applying by rote. I learned calculus by studying the theory, working out textbook exercises, and then I applied it. On the other hand, I learned to make Beef Bourguignon by following a recipe step by step, without having any understanding of why, for example, I should brown the meat before braising it. You learn different skills in different ways. Rote enables you make an acceptable Beef Bourguignon, if not a great one, but it doesn't work for calculus.

Whether you can or should unthinkingly apply a skill learned by rote to a real situation is another question. Once, a former plant manager who attended by Lean Logistics course told me "I have used all the tools in your course but no one had ever told me the logic behind them and the way they worked together. I was given instructions from corporate, and had to follow them or else I would have been fired, even if my plant made money."

His former employer was an auto parts maker and, while, on the face of it, the approach seems Darth-Vader bad, it had worked for that company. Lean came from the automotive industry, so, if there is one place where brute force imitation could work, it was in auto parts manufacturing. Even though it worked in this case, the approach has two fundamental limitations:

  1. Within the company itself, blindly following corporate directives does not set the stage for continuous improvement. For continuous improvement to happen, employees at all levels must have some understanding of underlying principles, and the freedom to deploy these principles in solutions they create.
  2. It does not carry over to other industries. In aerospace or frozen foods, the rote application of automotive tools does not work. The pace of work, its complexity, and the business issues are different.

One area where case-by-case thinking cannot be avoided is strategy. Rote application of Lean tools is an implementation strategy that should not be rejected out of hand, because there are cases where it works. It just should not be applied in the more common situation where it doesn't.

2. Tools, methods, and methodologies

Tools are devices we make or buy that enable us to do tasks we couldn't do with our bare hands. To become adept at using a tool, we need to learn or create methods. When tools and methods are arranged in a fixed sequence to obtain a standard outcome, the result is a process or a methodology.

What are the Lean tools anyway? Wrenches, hammers, screwdrivers, power drills, or even text editors are tools. Tools are all the devices we would miss if stranded on a desert island. On the other hand, we don't usually refer to the assembly instructions for a bookcase as a tool. Instead, we call it a method. Without the instructions, we could still figure it out, but it would take longer and involve more errors.

Tools are things we buy or build, whose possession does not make us adept at using them. Methods, on the other hand, are things we learn. The difference is clear in daily daily life but blurred in management-speak, where, for example, a method like SWOT analysis is often called a tool, event though all it has that resembles a tool is a 2x2 matrix on a white board with squares labeled "Strengths," "Weaknesses," "Opportunities," and "Threats." The value of a SWOT analysis depends exclusively on the skill of filling out this matrix in a brainstorming session.

By contrast, a piece of software that generates orbit charts, for example, is much more similar to what we call a tool, because it draws instantly and accurately a chart that would take a person hours to generate manually, with errors. In Manufacturing, we also often conflate tools and methods as, for example, in Kaoru Ishikawa's "7 Tools of QC," which range from checksheets to Shewhart control charts.

We like to call things tools, because, until recently, we thought we were the only species on this planet to have this concept. A few years ago, ornithologists observed a crow grabbing a stick to coax a treat out of a narrow glass tube, debunking our perception of tools as uniquely human. In the following video, you see a crow using a short tool to retrieve a longer one in order to eventually retrieve a treat:

We all have a set of tools and methods that we choose to use when the need arises. This choice is problem-solving strategy and requires thinking. You have more than just a hammer, not every problem is a nail, and you may have more than one tool you can use. A methodology is quite different, a "12-step, proven process" that you surrender to and relieves you from the need to think.

Processes, of course, are the foundation of repetitive production, and methodologies are successful for some endeavors, such as helping an alcoholic get sober, but one-size-fits-all "Lean methodologies" don't work because the Lean transformation of any factory is more like exploring uncharted territory than building a house. There is a proven process to build a house, at least if it is a tract house without special features. When you are exploring uncharted territory, on the other hand, you don't know what lies beyond the next hill, and must constantly adapt what you are doing to what you currently know.

House vs. Uncharted territory

3. The Tools of Lean

There is no consensus on what the tools of Lean are. When you google tools of Lean, you get lists from 5 to 40 items from many sources, some of which do not pertain to Manufacturing. In English, John Bicheno has two toolbox books, one for production and one for service. In French, Radu Demetrescoux's Lean toolbox, La Boite à Outils du Lean, holds 67 tools. The best and richest source by far, however, is the Japanese literature, with numerous compilations of tools and tricks published by Nikkankogyo, and monthly installments in manga form in their Factory Management magazine (Kojo Kanri). The sources include not only Toyota but also many other Japanese manufacturing organizations.

In addition, the lists contain entries that truly do not belong together, like Jorge Luis Borges's classification of animals. They contain small building blocks along with the systems they are part of, and items that are not tools at all. Here is an excerpt from one of the lists available on line, which reads more like a list of keywords than tools:

  • 5S
  • Andon
  • Continuous Flow
  • Gemba
  • Kaizen

The "tools of Lean" in most online discussions of the subject are limited to a short list, typically including Value Stream Mapping (VSM) and 5S that play a minor role in the Toyota Productions System (TPS) and Kaizen events, Design for Six Sigma, and Activity-Based Costing that are alien to it. Unsurprisingly, the deployment of these tools fails to yield the kind of performance that generated worldwide interest in TPS.

4. Conventional wisdom about tools

The conventional wisdom of the Lean community today is to downplay the "tools" and emphasize management and culture instead. In this view, if top management understands the philosophy of Lean, it can make it cascade down the organization and the tools are not essential; the tools of TPS are just for the auto industry anyway, the employees are the experts, and all they need is to be unleashed in order to achieve world class performance. In other words, there is no need to study the specific tricks developed at Toyota for the past 65 years and the underlying principles are all we need.

The notion is welcome to many managers and consultants, particularly when they don't have an engineering background, because it takes years of learning off the table. Unfortunately, this shortcut cannot be taken, and the belief that it can be is a fallacy, for the following reasons:

  1. You cannot teach top management the philosophy of Lean.
  2. There is more to world class performance than existing employee skills and knowledge.
  3. There is no consensus on what the underlying principles are, any more than on a list of tools.
  4. The underlying principles are embedded in the tools.

The following sections elaborate on these points.

4.1. Teaching management the philosophy of Lean

You can lecture managers but the best ones will not believe you. There has historically been no shortage of gurus peddling management philosophies that they claimed would make the company better, so why should they take your words more seriously? The only way their thinking can be changed is by experiencing some implementation success. The best you can do is get them to suspend disbelief long enough for carefully selected pilot projects on the shop floor to show the power of the approach. But the pilot projects invariably require the application of one or more of the tools of Lean.

4.2. Employee skills and knowledge: necessary but not sufficient

Production operators have expertise about the work they do and, often, ideas on how to improve their own workstations. At the outset of Lean implementation, however, this expertise does not extend to the design of production lines, the art of reducing setup times, or mistake-proofing, and these techniques are faster learned than reinvented.

Not all the tools developed at Toyota for making cars are applicable to, say, frozen lasagna or airplanes, and useful tools have been developed at other companies too, like Nissan's QRQC or Lucas Industries' Runner-Repeater-Stranger analysis. So you have to select relevant tools,  and adapt them as needed. And you may have to invent new ones.

4.3. No consensus on underlying principles

See Lean versus the Toyota Production System, for a discussion of several lists of Lean principles, from Toyota itself, from Jeffrey Liker's The Toyota Way, and from the Lean Enterprise Institute, among others. Following is my own, centered on manufacturing:
  1. Focus on people are the main driver of performance.
  2. Look for profits in the details of shop floor operations.
  3. All manufacturing is repetitive at some level, even where it doesn't appear to be.
  4. Make materials, information, and people flow.
  5. Make it easy to do what you do often.
  6. Improve, don’t optimize. Optimization comes to a full stop; improvements never end.

4.4. Principles embedded in tools

While many of the "tools of Lean" would be better described as methods or tricks that have emerged as point solutions to problems Toyota encountered over the decades, and coalesced into a system, as described by Takahiro Fujimoto. For this to happen, these tools have to be the expression of common underlying principles, which no one to date has successfully and completely articulated. As a consequence, learning the tools is, in effect, the only way to grok the principles embedded in them.

"Make it easy to do what you do often" is one of the principles I have encountered in multiple tools. It is what you do when you locate the most frequently used tools and materials closest to the operator in a manual assembly station, and it becomes more elaborate when automation is involved, for example when boxing multiple electronic products, when you use a robot with suction cups to pick the relevant manuals, warranty cards and promotional coupons in a row of bins and drop them in the product box. Then, obviously, given the hardware configuration, you minimize the robot travel time by locating the bins with the most frequently used documents closest to the boxing station, but you can take it further by changing the hardware configuration, split the row of bins in two and place the halves around the boxing station, which further cuts the robot travel time in half, as shown below.

Product box and doc bins

The same principle is at work in the first "S" of 5S, about decluttering the shop floor by removing all items that are not routinely used, because the effect is to make the ones that are routinely used easier to access.

Jidoka isn't just about "stop and fix"

Jidoka (自働化)  isn't just "stop and fix" or "stop and call."  It is a complete approach to automation that includes building in the ability of a machine to stop when it malfunctions but also includes many other things. Sakichi Toyoda's Type-G loom didn't just stop when the yarn broke, it also had automatic shuttle change, which reduced the need for human intervention in its normal operations, and was a breakthrough that had eluded everybody else.

The goal wasn't switching shuttles but switching bobbins within shuttles, which nobody managed to automate successfully. Toyoda was the first to realize that you could prepare shuttles with full bobbins off line, and that switching shuttles was easier than switching bobbins in a flying shuttle. It was similar in concept to replacing reels with cassettes in tape recorders.

The difference between Toyota's Jidoka and automation as originally conceived at Ford in the US in the 1940s can be seen in the total absence of consideration of the role of people in the American literature on automation. Jidoka as "automation with a human touch" also means automation with due consideration of production lines as human-machine systems.

A dramatic example I remember seeing in a Japanese auto parts plant was a fully automated cell with old machines where you could tell that various retrofits, including robots, had gradually taken over the routine work done by operators. Next to this cell was a brand new one that looked very different but had been built based on the lessons learned in the old one.

You can call it separating people and machines but, rather than just a separation, I see it as thinking through the way to effectively use people with automated systems. I understand you can organize concepts in many different ways and I have recently seen Poka-Yoke/mistake-proofing as part of Jidoka, but I would not put it there, because Poka-Yoke is about manual operations. Poka-Yoke belongs instead as a quality improvement tool, under the Quality management header.

Is there such a thing as managerial DNA? | LinkedIn

Scientific and technical terms are frequently used metaphorically in business, in ways that don't always make sense. Companies, nowadays, are commonly described as having certain practices, "in their DNA," and you hear discussions of "changing their DNA."As is known to anyone who has taken High School biology or watched a recent cop show on TV,  the one thing you can't change is your DNA. We each have our own version, formed at conception and replicated in every cell of our body.

Sourced through Scoop.it from: www.linkedin.com

See on Scoop.it - lean manufacturing

The Internet of Things in Lean Manufacturing | SME |F. L. Thomas

"[...]Within the context of Lean manufacturing, focused on elimination of waste and continual process improvement, the Internet of Things can lead to huge efficiency gains … some people see it as Lean on steroids. Tools and equipment will automatically collect, share and interact with other data and processes, opening up a whole new realm of achievements attainable under Lean initiatives.[...]"

 

Sourced through Scoop.it from: www.sme.org

Michel Baudin's comments:

Even though the article is a marketing pitch, and the Internet of Things (IoT) so far has been focused on systems embedded in finished goods rather than production processes, it is a topic that manufacturing professionals and Lean implementers should pay attention to.

Continue reading

Is Cost Reduction the Goal of TPS?

In a rebuttal to John Seddon's latest paper, An Exploration into the Failure of Lean,  Bob Emiliani asserts that the original purpose of TPS was to reduce cost. He quotes both Taiichi Ohno and Yasuhiro Monden saying so, and chides Seddon for not reading their works carefully enough. In the context of these documents, however, I think the quotes are misleading. Neither Ohno's and Monden's books, nor any other Japanese publication about manufacturing systems that I have seen, contain a discussion of what costs actually are.

Continue reading

On Reddit: "I am a Manufacturing Consultant, Ask me Anything!"

Reddit has a "subreddit" called "IamA," where an individual posts information about himself or herself, and all members are allowed to ask questions. I gave it a try last night, posting the summary of my LinkedIn profile. The questions were more like those students ask about career choices and strategies than the more technical ones I receive through this blog or through LinkedIn.  On Reddit, you only see usernames, so I don't know whom I have been responding to.

Continue reading

Theories of Lean and Leveling/Heijunka| Christoph Roser

ChristophRoser-200x300Christoph Roser has more impressive credentials than most Lean consultants, from a PhD in Engineering to a research job at Toyota labs, stints in operations at Bosch, and a professorship at Karslruhe University of Applied Sciences. So, if anyone is qualified to write a theory of Lean, he is, and he is trying his hand at it in production planning and scheduling.

Continue reading

“Going to the Gemba” and “Going to the Customer” | Philip Marris

In the TPS Principles and Practice discussion group on LinkedIn, Philip Marris started a discussion on this topic three weeks ago with the following statement:

"I believe that going to the customer is nearly as important as going to the Gemba.

I won’t comment on “Going to the Gemba” in this LinkedIn discussion group because normally we are all already totally convinced that it is vital and I would bet that most of us enjoy nothing more than those moments on the shop floor seeing, smelling, listening, touching, thinking, learning and coaching.

What I wanted to do with this post is: ask if there is a Japanese (or even better a Toyota Motor Company) term for “Going to the customer” and ask members what their experience is on this subject. I think it is both different and similar to “Gemba walks” and very important.

Personally, after over 25 years in industry in many countries and cultures, I am continually disappointed by the "go to the customer" practices. Some small to medium sized firms are good at it but large organizations seem to lose it. This is especially true of B2B (Business To Business) operations but also in B2C (Business to Customer or Consumer goods manufacturers) where even the best tend to completely delegate this crucial element to the Sales and Marketing people.

I think going to the customer is important for all employees: workers on the shop floor, those that purchase the packaging, those that print the bill, those that take the customer orders, the top management ... I would add that for some B2B companies they should of course go further down the Supply Chain and get to the final customer (the one that uses the product). "

There have been 28 comments so far, many of them theoretical, about the value of customer feedback, or off-topic, about the foibles of MBAs, with a few accounts of personal experience. I picked out the following:

  • Todd McCann: "While working with Boeing of Winnepeg (composite parts Mfg) we led a 3Gen road trip to Everett and Renton WA where the Winn gemba workers learned more in 1 hour how the customer viewed their parts and the Par traveling paper work that accompanied the part, than they wanted.  Eye opening to say the least. I recall, Heads kept moving back and forth side to side, eyes closed and open palms slapped the forehead."
  • José Ignacio Erausquin Arruabarrena. About 20 year ago, when I was managing a plant being Tier 1 supplier of the automotive sector, I remember as one of the practices that was really appreciated by the workforce, the one to have always one or two operators of the assembly line in the visits to the customer, when developing new products or even when answering to customer claims. Operators observing how their customer utilizes their product is one of the best sources of improvement that you can imagine.

My own experience is consistent with Todd's and José Ignacio's. When I was consulting for Boeing Portland, a machine shop making structural components, one of their best practices was sending a small group to the assembly plant that used their products once a quarter, to meet with the assemblers and collect their feedback. They recorded the interaction on video, and played it back to the entire production team back home. Different people went every quarter, and I went along on one of of these visits.

It is relatively easy to organize such things for an in-house supplier, to a plant that is not too far away. It is more complicated when you are dealing with actual customers, especially when the customers are not end-users but dealers or distributors. If you are selling to dealers, you can, for example, place technicians for a few months at dealerships when you launch a new product, to gain first-hand knowledge of any problems end-user may report to the dealership about the product.

When GM created the dealership system to sell cars in the 1920s, the primary purpose was to shield the production plants from the fluctuations in the market. Dealer inventories acted as a buffer to allow production plants to proceed at a constant pace. As they quickly discovered, however, this system also shielded GM from information about market trends, and they didn't realize the market had a downturn until the lots of their dealers were all full.

To keep a finger on the pulse of the market, Toyota in its early days sold cars door-to-door. Chatting with housewives, the sales rep learned that Mr. Yamada had been promoted, and paid him a visit in the evening to sell him a car to fit his new position. While it provided better market intelligence than dealerships, it was too expensive an approach and was abandoned.

Much later, about 2000, Toyota launched an internet portal in Japanese called gazoo.com, dedicated to "car life," with information like used car values, games for kids during long rides, recommendations for pleasant road trips, etc. It is different from the brochureware websites of other car makers. They didn't explain why they did this, but my guess is that it was to recover the direct contact with customers that door-to-door sales used to provide. Through their clicks, page views, and comments, gazoo visitors are telling Toyota about the market.

Another approach is to bring customers to the production plant. Until Honda of America closed their Marysville motorcycle plant in 2009, they held a yearly "homecoming" for bikers. All owners were invited to a big party, with a tour of the plant and meetings with the production teams. The idea was also adopted by Saturn, but every four years, and they stopped in 2004.

In a similar spirit, Porsche in Leipzig lets buyers pick up their Panameras and Cayennes at the assembly plant. The customers tour the spotless final assembly line, get an hour of coaching with a pro on the test track, eat at the fine-dining restaurant in the visitor center, and buy expensive souvenirs. And, for this privilege, they pay an extra 1,250 euros.

What's Next after Lean? | Industry Week | Larry Fast

"[...]What’s Next? The short answer is nothing. Don’t wait on anything new that is of a game-changing variety."

Sourced through Scoop.it from: www.industryweek.com

 

Michel Baudin's comments:

The emergence of Toyota and its production system (TPS) caught the manufacturing world by surprise. The first reaction was denial that it was new, followed by blind adoption of a few of its most visible features, and the development of something different, called "Lean," which borrowed Toyota's credibility but doesn't have much left in common with TPS.

Unlike Larry Fast, I am sure there will be another game changer in Manufacturing. It will come from an unexpected place, as post-war Japan was, and I have no idea what it will consist of. In the past 250 years we have had revolution after revolution in the art of making things, and I think it is presumptuous to assume that there won't be anymore.

See on Scoop.it - lean manufacturing

Applying Lean to Engineering | ENGINEERING.com

"[...]Applying lean concepts to engineering is a complex task, but as Dearborn Mid-West Company discovered the increased efficiencies are well worth the effort. To manage that complexity Dearborn focused on five key areas:

  • Eliminating wasted work
  • Defining common communication protocols
  • Adopting a common technology platform
  • Building core knowledge in a scalable small team and tools
  • Scaling the team and tools to meet short-term bursts in demand."

Sourced through Scoop.it from: www.engineering.com

Continue reading