Lean Implementation Strategies and Tactics

Where Should You Start?

You know your company needs an injection of lean. There’s waste everywhere. Upon serious reflection, you pretty much failed most of the tests. The culture is nowhere near the level that we describe as the Toyota Way. Leadership is not there, you don’t have effective work groups, functional groups are at war most of the time, problem-solving processes are superficial, and you’ve tried some lean tools here and there with good short-term results but no staying power.

Welcome to most of the world. Even Toyota has to work hard to maintain the Toyota Way, and it has particularly struggled to spread the true Toyota Way outside of Japan. It is continuous hard work.

So where do you start? In this chapter we talk strategy and tactics. You need to decide where to focus actual activities implementing lean. You need a plan. There is a great deal to do and many different ways to do it.

Returning to the 4P model, we are arguing that all four levels of philosophy, process, people, and problem solving are intertwined in complex ways. It is a system. So where does that get you? Unfortunately, reality being what it is, you need to start someplace. Even if you’ve been at lean for some years but it hasn’t taken hold, you have to start someplace to reinvigorate lean. You have at least four choices:

1. You can start with an off-site meeting of top leaders and clearly define your vision for becoming a lean enterprise.
2. You can begin implementing lean correctly as a connected value stream, as we describe in Part III of this book.
3. You can work to train and indoctrinate your people into the new lean way of thinking, directly effecting culture change.
4. Problem solving. You can train people in a problem-solving methodology and give them time to meet in groups and solve

All these approaches have been tried at various companies over the years with mixed success. And to some degree, you need to work on all of them. But if you have to pick one place to begin focusing your efforts, it has to be at the process layer—reducing waste in the transformation process. Having said this, there are still many choices to make.

Lean Implementation Levels, Strategies, and Tools

Another way of slicing the problem of planning implementation activity is to think about your organization in levels from the biggest slice to the smallest component.1 In Table 19-1 we look at levels from the extended enterprise, which includes all of the organizations and companies that somehow touch your product and work down to the level of the detailed individual process. Let’s consider each level, starting at the bottom with the individual process.

Process Improvement Approach

The individual process is what is done at a particular machine or by a particular worker for a manual operation: stamping or welding parts, performing an assembly, mixing a batch of paint, taking calls at a call center, entering data, etc. There should be a specific improvement objective for that process. It can be to reduce defects by 20 percent, reduce cycle time by 20 percent to improve output, eliminate work-in-process inventory by 50 percent, reduce downtime from 10 to 2 percent, and so on.

One common approach to process improvement as a lean strategy is the oneweek kaizen event. The kaizen event structure (a.k.a. kaizen workshop, rapid improvement workshop, lean event, rapid improvement event) is:

1. Prepare in advance. Two to four weeks of advanced preparation for the workshop to define the scope of the problem, decide on a team, collect data on the current situation, decide what lean tools to use, and make logistic arrangements for the event. In some cases there is advanced purchase of tools, materials, or equipment that cannot be done in the lead time of the one-week workshop.

1 Much of this discussion of implementation strategies and the figures used for the different strategies are based on a training course developed and taught at the University of Michigan by Bill Costantino, former Toyota group leader where he was an associate of David Meier.

Table 19-1. Lean Implementation Strategies and Tools

2. Conduct workshop:

• Monday: Give an overview of lean and teach any special tools needed for that week. Begin to collect data on the current process in the afternoon.
• Tuesday: Complete the current state analysis, collect data, draw a process flow map, draw walk pattern on layout, develop Standardized Work Combination Tables, etc., and develop ideas for the improved state. Perhaps detail the future state by the end of the day (Plan).
• Wednesday: First pass implementation (Do). It may be in one pilot, to try it first, or full implementation right away. Sometimes this starts by clearing the floor of the current process, painting the floor, then moving equipment back in the new layout.
• Thursday: Evaluate process (Check), improve (Act), and keep going through Plan-Do-Check-Act (PDCA) until you have a good approach.
• Friday: Develop a presentation for management. Present to management. Celebrate. (Often the event ends after a lunch celebration.)
  

3. Follow-up to the workshop. There are always items that could not be done during the week, which are put together as a homework list sometimes called a “kaizen newsletter.” An action plan for what, who, and when is prepared during the one-week workshop, and follow-up is needed to be sure the items get done.

The kaizen workshop approach has gotten a bad name in many quarters. Jim Womack used to laughingly refer to it as “kamikaze kaizen,” or “drive-by kaizen.” The implication was that you swoop down fast and furiously, solve some problems, and swoop back up, or drive by, take aim and fire, and you’re done. The problem is not that kaizen workshops are inherently bad, but that many companies turned their entire lean process into a series of kaizen workshops along with a kaizen promotion office to administer, support, and monitor kaizen events. They may even count kaizen events as a key performance metric. There are some serious weaknesses in this approach (see Figure 19-1):

1. Kaizen workshops generally are point kaizen focusing on the individual process. Since there is no broader vision, this will not lead to flow across the enterprise.

2. The kaizen workshop generally ends with a homework list of to-do items, which often do not get done since there is no serious ownership of the process by the people in the work area.

3. While people in the work area are involved in the event and get very excited and enthusiastic during the workshop, reality sets in the week after and more often than not there is backsliding toward the pre-workshop state.

4. There is a tendency to judge kaizen events based only on short-term cost savings, which does not drive true systems change.

5. There is no lasting cultural change.

This is not to say that good companies serious about lean should ignore the kaizen event as a tool. There are some remarkable strengths of the kaizen event, including:

1. This is an exciting experience for all involved. The concentrated analysis and improvement, combined with the feeling of being part of a team, can literally change people’s They can see waste and also see what is possible when waste is removed.

2. Management is enlightened on the speed with which things can be accomplished if a concerted effort is Amazing things can be accomplished with proper focus and leverage of resources.
3. People learn a great The intensity of the experience opens people up to learning in ways that are usually not possible in a traditional classroom approach.
4. Resources are usually made available, including management authority, cross-functional resources, and some So things can happen in the week that might otherwise take months of written requests, approvals, and cajoling people to help out.
5. Skeptics can be won In a classroom, the skeptics raise their hands and explain all the reasons lean will not work. Those same people in a workshop are making it happen.
6. As we will discuss later in the chapter, the kaizen event is a great tool for implementing aspects of an overall value stream vision.

Focused process improvement Specific improvement targets Isolated process improvements Toyota drives with hoshin kanri

Toyota uses variety of approaches Some companies use kaizen events Some companies use Six Sigma process

Figure 19-1. Strengths and traps of kaizen project approach

The Tenneco example from Smithville, Tennessee, which we describe below, illustrates the positive and negative of kaizen events. In that case, radical kaizen events every other week dramatically turned around a plant. About 40 percent of the workforce were “kaizened out.” Within one year they worked through every area of the plant, moving hundreds of pieces of equipment, making new shipping and receiving docks near the point of use, and basically remaking the place. The dramatic savings led to great management attention, and helped spur the CEO to invest in lean globally. We should note that an “event” is not necessarily a successful event. The Tenneco Smithville events were well-facilitated by a veteran lean coach who guided the plant and the event toward serious change. There are also events led by “kaizen coordinators” who lack the deep expertise and aggressive facilitation skills, and these can easily degrade into glorified 5S activities.

“Six Sigma” programs have some of the same strengths and weaknesses for process improvement as the kaizen workshop approach. They are generally of longer duration (e.g., several months), led by individuals who are earning or have earned “Black Belts,” and focus heavily on statistical methods and measurement. The origins of Six Sigma are in Total Quality Management (TQM), but advocates argue that Six Sigma adds a bottom-line financial mentality. Projects are typically expected to save the company several hundred thousand dollars. In fact, many companies track dollars saved through Six Sigma and even report these figures to stock analysts. Train 1,000 people doing $200,000 projects and in no time huge savings pile up. While Six Sigma uses statistical tools that can be quite powerful in the right hands at the right time, there are some serious traps in the Six Sigma approach:

1. Six Sigma focuses so intensively on analyzing data, picking the right statistical procedures, validating the statistical properties of the data, and developing slick reports, that the analyst can get distracted from the true purpose of the project and lose focus on the gemba.
2. Six Sigma anoints individuals as Green Belts or Black Belts and gives them a special status in the organization, yet their main skills are the analysis methods and not necessarily deep understanding of the processes they are improving.
3. The Black Belts can do too much on their own, turning the projects into technical engineering projects with minimal employee involvement.
4. The result is often lack of ownership by the people doing the work, and thus the recommended changes do not stick.
5. There is no real philosophy behind the Six Sigma program except to find, measure, and eliminate variation, and save a lot of money.

The find it and measure it, and analyze it and fix it to save dollars fast approach often leads to point kaizen that may even be contrary to lean principles. We have seen the following projects that are successful in saving money on a per piece basis but actually led the organization away from lean and ultimately increased total cost:

• Reducing changeover time, reporting labor savings, and increasing batch sizes, instead of decreasing batch sizes (see “Case Study: Six Sigma Changeover Reduction”).
• Reducing transportation cost by filling trucks through less frequent deliveries and increasing inventory levels in the plant.
• Reducing labor by assigning material handling and setup duties to workers in a cell, and as a result adding non-value-added activities to the core value-added workers.

“Lean Sigma” promises to provide the best of both worlds, but the “lean” in lean sigma is often narrowly construed to be a few technical tools like making a cell or developing standardized work. The result is point kaizen using both lean and Six Sigma tools without true flow and without the cultural changes necessary to support and sustain lean transformation. It has many of the weaknesses of the general process improvement approach through kaizen workshops and Six Sigma tools.

Case Example: Tenneco Smithville, Radical Kaizen, Phase I

Tenneco Automotive opened its exhaust system plant in Smithville, Tennessee, in 1994. The first customer was Toyota, and Nissan, Saturn, Honda, and Corvette followed later. In 1996 the plant was ISO 9000 certified and then QS 9000 certified and things were grand.

Unfortunately, the plant was set up around Tenneco’s traditional concept of process islands, with stamping, pipe bending, and different groups of welding machines together by function. Inventory of raw materials and intermediate products were everywhere, and large batches of each product type were run between changeovers. On the surface the plant was performing better than expected, and there did not seem to be a pressing need for change. It was more profitable than forecasted, and in terms of their primary measure—labor variances—they were $1 million favorable to plan.

But in 2000 trouble signs started to appear. Profits were low. Quality for Toyota was acceptable, but delivery reliability was in Toyota’s words “dangerous.” At one point, because of quality problems, Tenneco had to express ship parts by jet from Japan for Toyota at $30,000 a trip. It was clear they needed to do something, or they wouldn’t get any future business—half the plant’s business— from Toyota. At the same time, a new vice president of manufacturing, Joe Czarnecki, was brought in, and he had a completely different type of measurement. He noted that while the plant was profitable, by his calculations they should have been 20 percent more profitable. He looked at indirect labor efficiency, overtime, and inventory, which were all negative relative to his targets. Nissan was asking for a 20 percent price reduction, and Toyota was introducing a new program of price-downs. The need for change was rising fast to a crisis level.

Tenneco had recently hired a lean manufacturing expert, Pasquale Digirolamo, who agreed to dedicate almost all of his time to the plant for 8 to 12 months and treat it as a Tenneco lean pilot. Digirolamo and the plant manager, Glenn Drodge, met three times every day—a morning planning meeting, a midday review, and an end-of-day review. Digirolamo played a coaching role but was aggressive. He found the overall level of discipline in the plant to be weak and was fond of saying, “You get what you tolerate.” The Japanese consulting firm Shingijutsu had trained Digirolamo to lead radical kaizen workshops. He scheduled aggressive workshops every other week, in most cases setting up a complete manufacturing cell within the week. In the first six months, all subassembly operations were converted to cells. In the second six months, all final assembly operations became cells. The entire plant was laid out almost from scratch, and about 450 pieces of equipment were moved to the new layout. New shipping docks were built near the point of use. Primarily through the radical kaizen workshops, the plant was virtually remade from the bottom up. This was kaikaku (radical transformation), not kaizen (continuous improvement).

In preparation for this one-year radical remaking of the plant, Digirolamo estimated that the plant had 40 percent more workers than it should have. He recommended a one time layoff before the workshops began. Mostly temporary workers were let go as the plant relied heavily on agency workers. Other workers were offered Tenneco’s standard severance package, and enough took it to preclude involuntary layoffs for hourly employees. Some front-line supervisors were let go—people who did not have the management and leadership skills to perform in the new lean environment. The verbal commitment between the plant manager and Digirolamo essentially meant that Digirolamo was taking over the plant.

The bottom-line results were striking. Digirolamo came in as sensei in November 2000. Some time was spent on stability issues. In January 2001 lean deployment started seriously, led by the Smithville Lean Steering Committee. By April the plant had made a turnaround from below target to above target and other Tenneco plant managers were asking what was going on at Smithville. In the first year, labor cost was reduced by 39 percent, direct labor efficiency improved by 92 percent, total labor productivity went up by 56 percent, inventory dollars on hand were cut in half—freeing up $5 million in cash—external defects were reduced from 638 to 44 parts per million, and lead time was cut in half. In 2002 the plant for the first time received Toyota’s coveted quality and service award.

In terms of the different approaches to change covered in this chapter, Smithville in this first year had used a radical version of the “kaizen project approach.” It was kaizen upon kaizen relentlessly. Flow was created but mostly locally within cells. There were a few kanban systems that had been set up prior to this radical year, but the main focus of Digirolamo was on stability and cells. There was a clear bias for action, radical changes were made fast, skeptics were convinced in the plant and in other Tenneco plants, and the results were obvious. Table 19-2 summarizes the results. This plant’s success also got the attention of the CEO, who raised the priority of lean implementation. On the other hand, in terms of our implementation spiral (the continuous improvement cycle shown in Figure 3-4, Chapter 3) just part of one loop—stabilize, create flow, standardize—had been made across the entire plant. There was much work to be done to get to true Toyota Production System (TPS) anyplace in the plant.

Table 19-2. Smithville Lean Performance, 2001 One-Year Improvements

As we will see in phase two of this case presented later in the chapter, the plant made little progress in the next three years in lean, and some systems actually degraded. At this point they took a value stream approach and started with a model line. The current state map that reflected all of the kaizen improvements showed a bunch of push, welding cells, a bunch of push, more welding cells, and a lot of inventory. A future state map was developed and changes were implemented, resulting in another huge step up in performance. By itself, the radical kaizen events turned the plant around and greatly improved performance, but they did not lead to a sustainable culture change and did not drive true connected flow.

The kaizen project approach uses several specifically selected lean tools to address the exact process improvement purpose. Many of the problem-solving methods described in Chapter 13 are process improvement approaches. In that chapter we noted there are approaches to solving small, medium, and large problems. The medium problems are typically addressed by kaizen events or as Six Sigma projects outside of Toyota, as depicted in Figure 13-2 in that chapter. And Tables 13-1 to 13-3 show a variety of different approaches that Toyota uses for process improvement projects, including various types of cross-functional teams, Quality Circles, work groups under a group leader, and others. Depending on the project, these can be handled in different ways. It might be a very formal project assigned to a cross-functional team. It could be an assignment to an engineer who will pull together an ad hoc team. It could be a kaizen activity done by a work group with little outside help.

There are some common characteristics of these process improvement activ-ities at Toyota:

1. They are generally driven by hoshin kanri (policy deployment) objectives for the site that are linked to the site improvement objectives, which are linked to improvement objectives all the way up to the president of the company.
2. The process improvement project follows the steps described in Chapters 13 to 17. Ultimately, it will look like the problem-solving A3 report described in Chapter 18. It may be displayed on a board or a wall or actually on an A3 report, but all the elements will be included (problem statement, improvement objectives, alternatives considered, selected alternatives, justification, results, additional actions to be taken).
3. It will follow the Plan-Do-Check-Act cycle.
4. It will be part of an organizational learning process, with key learning shared across the organization.

Hot Projects Approach

Every operation has some immediate and severe pain which eliminated making the problem solvers instant heroes. It could be a bottleneck operation that is constantly holding up schedule attainment. It could be major equipment that breaks down at the most inopportune times. Or perhaps quality problems lead to whole groups set up to do nothing but inspection and rework.

Someone well trained in lean thinking and problem solving is well-equipped to quickly reduce this pain. In some cases companies use the one-week kaizen workshop as an approach to quickly analyze and solve these types of problems. As Figure 19-2 summarizes, there are both strengths and weaknesses to the hot projects approach.

Short-term urgency: solve current crisis One-dimensional

Specific improvement targets Isolated process improvements

Some companies use kaizen events

Figure 19-2. Strengths and traps of hot projects approach

We’ve been in consulting situations where management was skeptical about lean and had a “show me” attitude. They thought lean had potential and it was worth giving it a try. But they were going to wait and see if it applied in their operation, with their culture. In cases like this we might ask, “Where is your pain? What is it about your operation that keeps you awake at night?” This will generally lead to some juicy opportunities for immediate improvements that will knock their socks off. And of course if you’re working on a “hot project,” as defined by the leadership, they are likely to pull out all the stops and provide open access to resources and offer their own clout to get things done. When things almost magically get better, management becomes believers.

But those who live by the sword can die by the sword. Once management sees what lean can do for hot projects, they want more. “Lets go over there, where we have another serious problem.” Or: “Now let’s move over here, where this darn machine has been a problem since we first installed it.” And you can end up with the endless cycle of point kaizen we saw with the kaizen workshop approach. It’s almost like giving the really good stuff to a drug addict. You win them over, but at what cost?

Many Six Sigma projects are “hot project” approaches. The Black Belt is under pressure to produce major savings for each and every project. The most obvious way to do this is to find a hot project. The “Six Sigma Changeover Reduction” case study below illustrates this. The hot project was intended to relieve a bottleneck—injection molding—by eliminating changeover time. The project was a success and saved almost $300,000 a year in labor cost for changeovers. Unfortunately from a lean perspective, the result of this was larger batches and a lot more inventory of molded parts and a higher total cost. And the elaborate Six Sigma approach only reduced changeovers to 1.2 hours, which is very far from world class.

This is not to say the hot project approach should be completely dismissed. First, it’s a way to get some quick results and earn a license to do more thoughtful, longer-term lean system building—it’s money in the bank. Second, it’s something you should do anyway when you’re well along on your lean journey. Once basic lean systems are in place and there is a basic level of stability, flow, and leveling, and when people are in teams and have developed good problemsolving skills, they will often be working on hot projects. These will be the objective of kaizen. But it will not be the driver for lean transformation. It will be part of a more natural process of kaizen.

Case Study: Six Sigma Changeover Reduction— Reducing Changeover Time to Break the Bottleneck

In an auto parts plant that makes headlamps for vehicles, a young engineer was working toward her Six Sigma Black Belt. She selected as a project a major problem that the plant had had for years: an inordinate amount of time and resources focused on changing over plastic injection molding machines. This made injection molding the bottleneck in the process.

Detailed data were collected. The model changeovers averaged 3.5 hours. There were three changeovers per week, times 34 machines. This resulted in lost production of about 100 hours per week. The target for improvement was set at 2.5 hours per changeover, with anything longer defined as a defect. The project goal was to reduce 50 percent of the changeovers to less than 2.5 hours, thus cutting defects in half. A stretch goal was set at 90 percent.

A lot of data analysis was done to determine the probability distribution of changeovers; whether there were statistically significant differences across shifts, machines, and different molds. The system of measuring length of changeovers and process stability were both statistically verified, and a detailed process map for the changeovers was developed. Various statistical concepts were used like paired t-tests, Weibull distributions, box plots, and a four-way probability plot. Also, more traditional lean tools were used, such as listing the process steps and determining which could be done externally while the machine was running and which had to be done internally while the machine was down. These activities were prioritized, from those taking the most time to those taking the least. A fishbone diagram of the materials, man, methods, machine, measurement, and environment causal factors effecting inefficient changeovers was developed. The top two causal factors were identified as waiting for a changeover cart and the process of heating the die, which accounted for 38 percent of the changeover time, or 1.3 hours per changeover. They also discovered 12 of 22 other steps that could be done while the machine was running (external).

The Black Belt in training generated a brainstorm of ideas for improvement with some input from the floor. This was narrowed down to action items to be implemented:

• Schedule mold changes to coincide with lunch breaks so the dies could be heated during lunch (they could not justify the cost of equipment to preheat the dies).
• Add one additional cart, which would be enough to optimize the carts needed.
• Assign a dedicated changeover team instead of asking operators to do it, so they could prepare a lot of the external changeover items while the machines were running.

The results exceeded the goal. Detailed data were collected, put on run charts and statistically analyzed. It showed significant improvements. The result was a 98 percent improvement resulting in 2,828 parts per million defects (defining a defect as a changeover taking more than 2.5 hours). The average changeover took 1.2 hours, well below the 2.5 hour target. Analysis of the savings focused on the reduced amount of labor for changeovers, which amounted to almost $300,000 per year. Actually, the number of changeovers done in a week was over the budgeted number, and they had a parallel program to stabilize the schedule and reduce the number of changeovers. So there were arguments about whether her project should get credit for the labor savings based on the current number of changeovers or on the anticipated reduced number of changeovers.

So this was a big success, right? Or was it? Let’s consider what’s wrong with this picture:

1. The total process took several months. Much of that time was spent on sophisticated statistical analysis and preparation of presentation materials. If an experienced lean specialist did this, it could have been done within a one-week kaizen workshop.

2. The young engineer did most of the work while working mostly alone. There was little involvement or buy in of the workforce in the area.

3. The young engineer ruled out some of the most important ideas. For example, she ruled out preheating the molds, which would have had a major impact. A more experienced manufacturing change agent would have fought for this.

4. The objective of 2.5 hours is not a challenging goal, and even 1.2 hours is not a stretch objective for an injection-molding changeover. A more reasonable goal would have been 15 to 20 minutes, and a stretch objective would be five minutes, which is done routinely in lean plants. A 15-minute changeover could have allowed for more changeovers, reducing batch size, and still reduced the amount of labor significantly.

5. The overall value stream became less lean. There was no value stream map done. After the fact, a map showed that there had been five days of injected molded parts after molding, before the changeover reduction activities. By reducing the time of changeovers, doing changeovers only around lunch, and then reducing the number of changeovers, days of molded parts inventory actually increased, increasing flow days. Value stream mapping would have suggested reducing changeover in order to increase the frequency of changeovers to drive down inventory.

Plantwide Lean Tools Approach

A close cousin to the hot projects approach is what you might call the “hot tools” approach. Often when we teach professional short-courses on lean, we discover the main goals of the participants are to “learn some tools they can apply back at work.” Tools seem to be the punch line, something really practical. Theories are nice, but tools work.

Again, we do not want to suggest that there is something wrong with lean tools. Carpenters, musicians, athletes, engineers, and any other professionals certainly need to master the “tools of the trade.” This is not optional. What we’re talking about here is whether the focus of your lean activities early in the lean process should be on mastering and broadly implementing one tool at a time.

There is a lot of attraction to going wall-to-wall through the plant implementing one tool at a time, as summarized in Figure 19-3. Or, in a multisite company, you can go across plants. Any of the lean tools can be implemented in this way,including standardized work, Total Productive Maintenance (TPM), 5S, quick changeover, cells, kanban, mistake proofing, Six Sigma, and even work groups. It seems a relatively fast, easy, and inexpensive approach to learning a lot, generating a common awareness, developing standard templates for implementation, and laying the groundwork for further lean system development. Chapter 4 emphasizes the importance of developing stability before flow. So why not go across the organization implementing stability tools like TPM and standardized work? We also emphasized, in Chapter 3, creating initial process stability in two operations in order to create connected flow between them. We’ve been emphasizing lean as a system, and the real benefits of lean come from creating flow in a lean system. You can see this when the system is in action. Spending years creating isolated stability in place after place will delay creating connected flows and limit the ability to learn what real lean is. If stability is like the foundation, then you are building foundation after foundation, and in the meanwhile no one sees what the house is like.

Install one lean tool plantwide Narrow focus

“Cookie cutter” implementation

Wall-to-wall all areas

Figure 19-3. Strengths and traps of plantwide lean tools approach

An important part of the “house” concept is that the parts mutually reinforce each other. For example, stable processes are necessary for flow, but flow lowers the water level and forces improvement in stability. Machine down time will kill flow, but why knock yourself out every day on preventive maintenance if when the machine goes down the next process keeps working off of inventory anyway? When machine shutdowns choke the next process so it also shuts down, there is a sense of urgency to fix the machine and do your preventive maintenance.

Also, the tools are there to support waste elimination, not as stand-alone tools. Take reducing changeover times, the main benefit of which is that you can change over more often, reduce batch size, and support leveled production. But as a stand-alone tool, we’ve seen many companies use changeover reduction to simply produce more parts and make even larger batches. This clearly sends the wrong message.

Company X Production System Approach

Now let’s jump way up to the total organization level. Let’s say the vice president of manufacturing decides to get serious about lean. Through reading, benchmarking visits, or a few successful kaizen events or hot projects, this executive proclaims, “We need a true lean production system.” This is a noble vision and ultimately something we want with lean.

We have assisted in a number of “Company X” production system creations. One of the largest was the creation of the Ford Production System in the mid1990s, or should we say re-creation, since TPS was originally based on Ford’s system. The story in each case is the same as the summary in Figure 19-4. The consultants work with internal lean staff, with involvement of others in the company to “create” a system. While the system is based on TPS, there are modifications in language, the imagery (e.g., Ford used a five-interlocking gear model), and perhaps certain policies to fit the company. Considerable time is spent on the precise language and image. There is broad circulation of the documents and PowerPoint presentations to get agreement from senior management.

Create standard operating system Focus on education and training

-Understanding & buy-in

-Convincing upper management

Typically large, multiplant organizations Staff-centered deployment

All advances together in standardized approach Focus on right lean metrics

Figure 19-4. Strengths and traps of X production system approach

Varying degrees of standard operating procedures are put together. A lean assessment is created. The company realizes the current measurement system rewards mass production behavior, so it develops “lean” metrics such as leadtime, first-time quality capability, and overall equipment effectiveness (OEE). Worker morale is determined by conducting a survey. At Ford, key metrics were developed for each gear.

“Rolling out” the new production system (sometimes called “operating system”) is a process of education and training: education on basic lean concepts and training on specific details of the operating system. For example, Ford needed a several-day course on using the new lean metrics since every plant in the world was required to begin tracking the new metrics and reporting them. The focus is on one production system standardized for all the plants. This is the way Toyota operates, and it’s a good vision. It allows for easy sharing of best practices.

There are many good things that come from the effort to develop and spread a common operating system. It begins to give the organization a distinctive identity and a way to identify with its own tailored operating system. It provides a common language for communicating about progress. The lean metrics can help promote stability and flow instead of overproduction.

So what can be wrong with such an obviously good thing? The main issue is whether the cart is being put in front of the horse. The Toyota Way is based on action and learning by doing. The built-in belief is that people do not truly understand until they experience lean as a system. Otherwise, it’s just an abstraction, which you may grasp with your head but not your gut. If you grasp it with your head, it’s easy to intellectualize it. Basically you have three problems:

1. How can you create your production system if you do not truly understand lean?
2. Since this is often a consensus process, even if a few individuals have a good understanding of lean, others may not.
3. Developing an operating system is attractive to those with a bureaucratic mentality who love developing metrics, planning training, and envisioning the organization of the future but are eager to avoid real action.

All this amounts to a slow and expensive process of talking and developing PowerPoint presentations and teaching and talking some more. You learn lean by doing, not by talking. Or as our friend and former Toyota V.P. Russ Scafade puts it: “You can not PowerPoint your way to lean.”

Value Stream Model Line Approach

You now know many things you should not do, but what should you do? Like Goldilocks and the three bears, some approaches are too narrow and specific (e.g., process, hot jobs, tools approaches) and others are too big and grandiose (like the Company X production system). We believe the value stream model line is just right for most organizations. What do we mean by this?

While hot jobs and processes are scattered across various points in the organization, value streams cut across the organization from raw materials to the customer. Lean is a value stream philosophy: Start with what the customer values and eliminate waste in the value stream. So why not focus on building lean value streams, since tools come together to create systems at that level?

Value stream mapping, described in Chapter 3, is a core tool for envisioning your lean value streams. It starts with a current state map, which provides a picture of the current situation. Waste becomes apparent, but in this approach any process kaizen to fix problems in the current state is strongly discouraged. The value stream map is not intended to determine a set of point kaizen activities. Rather, the current state is the starting point in developing a lean future state vision—a holistic picture of connected flows. Ideally, a cross-functional leadership team led by a “value stream manager” or other high-level manager creates the current state and achieves consensus on the lean future state. Action is driven by project plans to achieve the future state.

The action plans are straightforward Gantt charts. But we strongly advise that actions be organized around material and information flow loops.3 An example of a future-state map divided into loops is shown in Figure 19-5. In this case three loops are shown:

1. Pacesetter Loop. This loop is closest to the customer, and it paces all upstream operations. It is also the one schedule point in the In this case, the leveled schedule is sent to Process 3 but then it flows through— first in, first out—without a break in the sequence to the finished goods supermarket. The pace of Process 3 then establishes the pace of pull from the intermediate goods supermarket, which pulls from Process 1, which pulls from the supplier.
2. Intermediate Process Loop. This processes supplies materials and replenishes the supermarket that holds products for its customer—the Pacesetter Loop.
3. Supplier Loop. This includes the supplier of raw materials, and the replenishment loop to keep the supplied parts supermarket stocked with materials.

Intermediate Process Loop

Figure 19-5. Basic future state divided into loops with kaizen bursts

Pacesetter Loop

Notice that each of these loops is a complete closed loop of material and information flow. Material flows toward the customer and information flows backward to trigger the next order from the immediate customer. Each loop can be independently worked on from a lean perspective, and the supermarkets buffer one loop from minor disruptions while another is being changed. A set of “kaizen bursts,” specific point kaizen activities, are needed to stabilize the process.

Kaizen projects are not replaced by the value stream approach. Individual processes must be stabilized and variation removed through kaizen projects. A particularly challenging problem of process variation might benefit from a sophisticated Six Sigma project. Nor does it replace the lean tools approach since lean tools are needed to implement each piece of the future state value stream—cells, kanban, etc. What it does do is put the use of these tools and process improvements into a broader perspective—the material and information flow as a system. It also impacts the sequence in which implementation occurs. There is often a tendency to implement one tool at a time, for example, to do quick changeover across the plant. In the value stream approach you work pull loop by pull loop and do whatever is required to stabilize, create flow, standardize, and incrementally level that particular loop. In some cases you may have the resources to work on multiple loops in parallel, and in other cases you may want to work on them sequentially.

“Learning to see” method Select product family Current & future state maps

Develop detailed action plan (“loop by loop”)

Project management approach

Visual management (“glass wall process”)

Figure 19-6. Strengths and traps of the value stream model line approach

Even the hot projects approach has its place in this approach. For example, it may be that Process 1 is a severe bottleneck and regularly shuts down other operations, causing late shipments. By all means start with Process 1 and the Intermediate Process Loop. There is no lean law that says you must start with the Pacesetter Loop, but other things being equal, this loop is the logical starting point. That is, start closest to the customer and create a leveled pull at the pace maker, beginning to create a sense of takt time in the value stream at that point. If we return to the Tenneco case three years later, you’ll see they initiated a value stream approach at Smithville. They began with a model line and value stream mapping. They found that the result of their initial foray into lean through events still left them far from a lean model. The new wave of value stream improvement got them additional results as dramatic as the initial wave of radical kaizen events.

Case Example: Tenneco Smithville, Value Stream Approach, Phase II

After the one year of radical kaikaku transformation through kaizen events in 2000, the plant did not make a lot of improvement, and in fact slid back from where it had been after the events. The 5S and other lean systems were not always followed, and the plant started to become less organized. This began to be seriously addressed in 2003-2004 when the plant shifted to a value stream approach emphasizing overall material and information flow across processes creating connected flows. The new approach was the “value stream model line approach.” The Toyota product family was selected as the pilot. Rick Harris’s firm was brought in, and the model of a purchased parts supermarket and tugger route described in Making Materials Flow 4 was adopted.

When Smithville mapped the current state, they found that despite the earlier radical kaizen, they had islands of lean connected by push systems. The current state value stream map is shown in Figure 19-7. Notice all of the push arrows. Basically we have inventory coming in, being pushed through various manufacturing processes, being pushed to one stage of assembly (welding of subassemblies) and then pushed to final assembly, where the muffler (brought in from outside), tail pipe, and such, are all welded to a complete exhaust system. The total lead time from steel coming in until exhaust systems were shipped out was 17 days.

The future state vision, which has been implemented, is shown in Figure 19-8. We will not go through all the details of the map, but here are some of the highlights:

Figure 19-8. Toyota 500N center cell future state fourth quarter 2004

1. The two stages of assembly (Assembly and Secondary Assembly in Figure 19-7) were combined into a single Assembly & Secondary Assembly cell (flow where you can).
2. The manufactured component operations (Tube Bending, Stamping) and purchased components are on a pull system using a supermarket and kanban (pull where you must).
3. One of the manufactured components that had been in batch mode (the Spun , or spun resonator) is built on a machine dedicated to Toyota with parts flowing through a small first-in, first-out buffer to Assembly. There are three boxes of inventory as compared to 11⁄2 days in the old system.
4. A daily order goes to one place—Assembly—and is leveled, with everything else pulled to MRP (Material Requirements Planning) has been turned off for everything except long lead-time purchased parts.

The purchased parts supermarket is modeled after Toyota’s system. There is one central supermarket, and then a “water spider” makes regular timed routes from the supermarket to the various operations, delivering parts on a one-hour route. She picks up kanban and manages the entire kanban delivery system inside the plant. The route repeats over and over each hour, and there is even detailed standardized work that shows minute by minute where she will be—like a well-executed bus or train system. The result was a reduction in material handlers even though deliveries went from once a day to every hour.

The results in Figure 19-9 are impressive. Complete implementation took nine months and purchased parts inventory was cut in half, one-quarter of the floor space was freed up, parts per employee almost doubled, and overtime was reduced from 252 to 10 hours per week. Bear in mind that these levels of improvement are possible in a relatively short time because this plant had previously developed a broad base of lean capability that allowed Tenneco to work on multiple value streams simultaneously. While this value stream was being worked on, Tenneco extended the model line approach to their other main value streams, which were mostly complete about six months after the original model line. System-level changes like these are generally far more sustainable because they drive more significant cultural change.

Are there disadvantages to the value stream approach or is it nirvana? Obviously, no one approach is perfect. As seen in Figure 19-6, above, the value stream approach can be time consuming, require leadership of a cross-functional team, and a lot of involvement at all levels; and while the model is being developed, other managers and team associates are kept waiting to see how it develops. We’ve seen ineffective execution of this method, most often when a team becomes preoccupied with mapping, creating beautiful maps with highly accurate data but little action—value stream mapping wallpaper. Some plants decide to map every product family in the plant, which can lead to endless meetings, mapping wallpaper, and no action. We believe in the principle of “no map before its time.” Develop a map when you will use it for implementation— immediately!

Figure 19-9. Making materials flow pilot cell: D27 resonator assembly benefits

Having the Patience to Do It Right

These approaches are not mutually exclusive. The point is to have a logical and well-planned process of deploying lean tools, which leads to lean systems and finally lean value streams. Tenneco developed a high-level, future-state value stream plan. A product family was selected, mapped, and a value stream model put in place. But soon after launching this value stream model line there are other process improvements and whole plant activities that begin. For example, kaizen workshops may be used in problem areas, tackling hot projects, and individual tools like 5S or TPM may be implemented across the plant (Figure 19-10).

The advantage of using these approaches concurrently is that you can get the strengths and reduce the weaknesses. You can get the benefit of building a pilot to go and see and learn from, and experience TPS in a holistic way. You can also get a broader set of people involved in experiencing basic lean tools. And you can solve hot problems and get management attention and resources. The disadvantage is the very real risk of spreading resources too thin, in which case nothing gets done well or in a timely manner. You must be aware of this and be prepared to pull back on some projects if there’s evidence of poor quality work or significant delays. In this case, pull back the plantwide activities and focus on the model line.

Figure 19-10. Sample lean implementation approach

Note the Plan-Do-Check-Act cycle shown for each level of implementation in Figure 19-10. Lean implementation is about more than getting money back for an investment. Companies often invest considerable money in training and consulting support, and top management asks for the infamous “business case.” When can we expect a payback? If you run the numbers for this business case, you’ll probably only get credit with accounting for tangible cost savings. Mostly this means cutting heads. You might also get 10 cents on the dollar for inventory savings. The continuous improvement team, or whatever they are called, is now under the gun and transfers that to the consultants: “We need a payback in one year.”

A good lean consultant can get this payback. They can reduce people, cut inventory, and make the numbers add up. But what are they really doing? In terms of the PDCA cycle, they’re going through rapid successions of Plan-Do, Plan-Do. There’s barely time to catch one’s breath and check anything other than the resulting cost savings.

We’ve discussed many lean improvement strategies at many levels. The model in Figure 19-11 puts these into a framework based on two factors: Is the improvement strategy focusing primarily at the value stream level or primarily at the individual process level? Is the improvement strategy primarily aimed at applying technical tools to get short-term results, or does the goal include longerterm development of your people?

We’ve described the strengths and weaknesses of the process improvement approach and described the value stream approach based on value stream mapping and a model line. Both approaches are often used by companies primarily for short-term, bottom-line results. But doing this misses a much greater opportunity—to develop your people and organization so they’re capable of making many of these improvements and thus multiplying the benefits. Many companies with employee involvement programs focus solely on process improvements and people development. People get team training and training in problemsolving tools but don’t understand broader value stream improvement concepts. As we’ve seen, Toyota works on improvement in all of these quadrants, but more than most companies, they have worked to build a lean learning organization that combines value stream improvement with people development. Where is your company?

Value Stream

Isolated Processes

Technical Tools— Short-term results

People Development Focus

Management Orientation

Figure 19-11. Management approaches to improvement: Where is your company?

Denso is Toyota’s largest supplier and grew up with TPS along with Toyota. Yet its Battle Creek, Michigan, plant was considerably less advanced in TPS than Toyota. In the past they had implemented isolated tools, but did not put lean together as a system. As we’ll discuss in the case example below, they developed their version of the Company X production system, which they call “Efficient Factory,” in order to involve everyone in continuous improvement. To implement this they used real projects and the “value stream model line approach.” Pilot product families were selected for each major product line and began to implement lean value streams from the customer back through to raw materials. Only when the model lines were implemented and the methodology tested did they move to other product families. Even a plant thought to be a lean model needs to periodically take stock of where it is and bring lean to another level, and the value stream model line is the recommended implementation approach.

Case Example: Denso’s Efficient Factory Value Stream Approach

Denso is Toyota’s biggest parts supplier, with almost $24.2 billion in sales in 2004 and 95,000 associates. Originally, the electronics division of Toyota, Denso was a spin-off, but Toyota retained a significant portion of the company (currently owning 23 percent). As TPS grew up within Toyota, Denso grew up with it, and as Toyota began to build cars in the United States, Denso built a plant (DMMI) in Battle Creek, Michigan, in 1984, to make automotive heat exchangers (radiators/condensers) and air-conditioning units. DMMI has experienced remarkable growth in a highly competitive auto parts industry year after year from its largest customer, Toyota, as well as DaimlerChrysler and General Motors. Annual sales between 2002 and 2004 went from $1 billion to $1.25 billion, and Denso’s reputation for exceptional technology, high quality, and near perfect delivery placed it at the top of the list for high-performing companies. The automotive parts supply market is a difficult one in which to make a profit, but DMMI has been profitable year in and out. It would seem that Denso is an excellent example of lean manufacturing and has little more to learn. Those who don’t understand the power of continuous improvement might say, “We have arrived,” but DMMI knows differently.

In 2003, DMMI, Battle Creek, introduced a new activity: “Efficient Factory.” DMMI is a company steeped in TPS tradition, thus, moving to a concept like “EF Activity,” one might envision next generation automation, information technology, and new lean concepts. Yet, “EF” is simply DMMI’s modified version of TPS. The EF symbol (Figure 19-12) has the appearance of Egyptian origin, possibly discovered on the walls of one of the great pyramids. Though the EF symbol looks intriguing, its meaning has nothing to do with advanced manufacturing technology, but everything to with people and philosophy. This symbol is also called the “Takahashi Triangle” after Denso chairman Takahashi, who retired as a Toyota senior executive. Driving it hard through DMMI is its President Akio (Alex) Shikamura, a true disciple of TPS. Certainly having a true believer at the top has been a key driver for deep change.

Figure 19-12. Symbol for Denso’s Efficient Factory activity

It is called EF Activity, not EF Program. What activities did DMMI begin in the name of EF? In the past they had many excellent technical programs to improve performance, including engineering-led kaizen through 1996, Total Industrial Engineering (TIE) from 1996 to 1997, and TPS concepts in 1998 (small lots, kanban). From 2000 to 2003 DMMI realized they needed more team member involvement so they created a program called WOW (wipe out waste). And each program had a major impact on manufacturing performance. But still, they realized they were significantly behind Toyota plants. So in 2002 they started EF activities with the following purpose:

1. Increase the “kaizen mind” of all associates.
2. Create a common target (vision).
3. Reduce costs by eliminating waste throughout the value stream.

EF focused on associate involvement to reduce waste throughout the entire product stream, from supplier DMMI to the customer. They realized that to bring TPS to the next level, they needed to invest in TPS experts in the plant. They selected Andris Staltmanis to lead the Manufacturing Engineering Department to a higher level of TPS. Andris has 18 years of production engineering and manufacturing engineering experience and was one of the originals at Battle Creek. In Yamanouchi Yutaka, vice president of Production Control and Planning from DENSO in Japan, he had a sensei to teach him.

However, it was understood that the key to success was production ownership. Joe Stich (general manager of Production) was also well versed in TPS and needed to drive this activity from within.

For deployment, they split the plant into three focused factories: HVAC (heating, ventilation, air-conditioning), condensers, and radiators. Within each of these they selected a product family to become a model line—to go and see and improve. At first Manufacturing Engineering facilitated the model area, and then responsibility was gradually transferred to Production. The approach included basic process kaizen, floor management improvement, and value stream improvement. Some of the tools used were visual control, standardized work, small lot size, frequent delivery and pickup, and a heijunka (product load leveling) board.

Bryan Denbrock, section leader in the M.E. Department responsible for implementing the high-level model system in the HVAC plant, described establishing plantwide heijunka as particularly challenging due to the variety of products and customers. With the target of becoming a “world class” company, the HVAC model line created a system for finished goods production. This model line served as the tangible reference example for the rest of the plant.

The finished goods are shipped from a warehouse to the customer. Three hours worth of customer orders are brought to a large customer staging post. While the product is being staged for the customer, the kanban are removed. These kanban are then taken to the heijunka post. Kanban are arranged in order to level the production signal, which has a pitch of 10 minutes. This means every 10 minutes the material handler brings an order (kanban) to Production to collect the required product to be replaced in the warehouse, which represents what the customer has actually purchased. The warehouse kanban are exchanged with the production kanban, and creates the next 10 minutes worth of production. This paces the one-piece flow assembly line.5

When this heijunka process is applied throughout the plant (using the visual management boards known as heijunka posts), it’s possible to see the state of the whole process for a whole day in one place. It is natural within the Toyota philosophy to use the material handler—or “water spider,” as they are sometimes called—in this capacity, since they can see the entire material and information flow in their route.

By creating this leveled condition throughout the plant, all forms of work can be standardized based on the 10-minute interval. This simplifies each operation, and it becomes immediately apparent at a glance if the standard is being followed. Once this condition is met, highly capable individuals who can carefully observe, understand, and think can understand the condition of the entire plant.

The heijunka post levels production across many part numbers. To achieve this, changeovers (fixture changes on the assembly line) were reduced to less than the takt time (takt time equals available work time divided by customer demand). A two-shift assembly line is changed from 90 to 125 times per day. The takt level is achieved through conveyor line spacing and the rebalancing of work elements for the team of associates on the assembly line. Internally there is a two-way kanban to an intermediate parts store (withdrawal) and then to a manufacturing process (production) for the parts pulled for use by the final assembly area. Kanban are brought to the store 88 times per day. Achieving these kinds of pickup and delivery frequencies requires a fine-tuned process that is highly stable. Even small problems will disrupt production and show up almost immediately. For this reason, for a company to be successful with these interruptions it must be committed to fixing problems immediately and then following up with permanent countermeasures. In terms of the continuous improvement spiral in Figure 3-4 (Chapter 3), the plant is several iterations down the spiral of stability, flow, standardization, and advanced production leveling.

Performance results on the model lines have been impressive. Product cost has been greatly reduced, while quality and delivery have risen to noteworthy levels. Most important, DMMI can utilize the success of this activity on other existing assembly lines. DMMI team associates have been directly involved and are transitioning to a new level of “kaizen mind.” This allows the company to foster a new culture where associates’ kaizen power can be tapped and implemented quickly.

Lean implementation is a learning journey, even in advanced stages. Every experience is an opportunity to learn and grow. But you have to take the time to check and then think about what actions will improve on what you’ve already done. We’ve heard statements like the following when preaching this learning perspective: “But we are in business to make money. This is the real world.”

Toyota is making lots of money. But it took decades of work to get to the point where they benefited from early investments in learning. When we give this advice—to make the necessary investment—it’s obvious to us that there’s waste everywhere and the company can benefit from better quality, shorter lead times, more flexibility to respond to change, and increased productivity. Making some up-front investments in learning will greatly multiply long-term savings. Remember in the 4P pyramid, the base is “thinking long term, even at the expense of short-term financial considerations.” Organizations that view lean as a short-term cost-cutting program are never going to achieve what is possible. They will never become high-performing organizations.

Many companies are anxious to spread lean quickly to the enterprise and extended enterprise levels. Simple analysis will show that most of the costs are typically in supplied parts. And it is well known that the impact of upstream processes like product development have multiplier effects on manufacturing that are far greater than the investment in product development. So why not start in those areas right away? Our experience is that starting enterprisewide and extra-enterprise level programs prematurely does more harm than good. There are a number of reasons for this:

1. Lean is easier to see in physical operations. Remember that much of the early stages of lean are about learning. It is also, unfortunately, about politics—selling the decision makers who hold the purse strings by getting visible, measurable results. This is easiest to do in routine physical processes. In pure service organizations it’s easiest in the most routine parts of the business, for example, order entry, or the test labs in a hospital.
2. There is a risk of overtaxing resources. Management is likely to assign only so many people to lean. Focus on those from whom you’ll get the best results and learning. Even if a separate staff is assigned to a “lean office,” they’re better off first spending some time in the trenches working on the core value-adding operation. They will start to understand lean at a deeper level, and much of that learning will transfer to the office environment.
3. Lean service operations should support the core value-adding operaYou can lean out a support function by making it more efficient,but any lean project should start with the business purpose: Who is the customer? What do they need? If the customer of the service operation is some type of physical transformation process, first go see what that will look like when it’s lean, so you can understand how to support it. When Glenn Uminger was asked to set up the accounting system for the Toyota Georgetown plant, he first spent a year doing TPS projects on the shop floor, which dramatically changed the way he looked at and developed the accounting system to support TPS. It was simpler, less cumbersome, and leaner.

4. There is a risk of turning lean into the latest “program.” Often, the best lean consultants and experienced lean people are assigned to the manufacturing or core value-adding process in a service organization. Support functions are left to largely fend for themselves based on a short training program. The continuous improvement group does a superficial job, and lean starts to look like the program of the month. Doing it right is more important than doing it early.
5. Trying to lean out suppliers before you’ve done it yourself is hypocritical and dangerous. What right do you have to teach lean to your suppliers if you’re not lean yourself? You need to earn that Also, since the lean supply chain is a hierarchy of many different elements that must be in place, if you start “developing” suppliers before you have mutual understanding and trust, suppliers will view the development as your excuse to hold them up for price reductions.

What we’re preaching is patience. Think about the Buddhist monk teaching a young disciple, or the karate teacher, or for that matter any good teacher of a complex skill like a sport or musical instrument. You do not begin by playing the sport or playing songs. There are tedious exercises necessary to prepare yourself. You need basic muscle control and concentration. A top golf instructor taught by one of the world’s great golfers said he spent the first three months learning golf without ever hitting a ball. Think of the Ohno circle. Stand in the circle and look. This need for patience and discipline extends to the problem-solving process. Do not race in and start implementing solutions. Take the time to find the true point of cause and then ask the Five-Whys for the root cause. Take the time to teach each employee step by step, using job instruction methods, before throwing them into the work routine. Take the time to check and audit and develop countermeasures to learn and improve. Make many little improvements, not just the big, visible ones. This patience takes vision for what can be in the long term. It takes a philosophical understanding of the purpose. It is the hardest part of lean. But in the long term, the payoff is remarkable.

Reflection Activities

Most of the readers of this book will be part of organizations that have done something with lean in the past. Many will have done quite a bit over a number of years. For those with some experience we would like you to reflect on where you have been and then develop a plan for what you should be working on next in the “process level” of the 4P model. For those complete novices here is an opportunity to work out a plan. This is a reflection that will need to be done together with a team from your organization— a team of decision makers that can legitimately set a direction for your lean initiative.

1. Take some time to list the process improvement activities you have worked on in the name of “lean.”

2. Classify the most important lean activities in the 2 x 2 matrix of Figure 19-11. Where has most of your activity been located?

3. Now classify the most important lean activities in the matrix in Figure 19-11. Where has most of your activity been located?

4. Now think about how you can build on what you have accomplished. Where should you go next in the models in Figure 19-10 and 19-11? For example, if you have mostly focused on tools or hot projects it may be time to undertake a value stream model line. If you have a good deal of experience on the left side of Figure 19-11—the tool side of the matrix—it may be time to work on the people development side. Note the lesson from Denso that working on the people side still means involving people in concrete improvement activities at the process or value stream levels.

5. Develop a high-level work plan. You can use as a framework the simple conceptual diagram in Figure 19-10 with some rough dates.