Gemba Panta Rei

On page one of the October 29th, 2004 Wall Street Journal there was an article titles "Big Dealers Pressure Car Makers To Cut Production: Higher Costs, Thinner Margins Are Forcing Major Sellers to Keep their Inventories Lean". It seems the Detroit automakers' strategy of offering large rebates and no interest loans of 60 and 72 months are no longer enough to keep overproduction going at the factories.

In Lean thinking and according to the Toyota Production System there are 7 types of waste that erode profit and reduce operational performance. They are:

1. Overproduction
2. Transportation (or Conveyance)
3. Motion
4. Waiting
5. Processing
6. Inventory
7. Defects (or Correction)

The greatest waste of them all is Overproduction. By overproducing you create a false demand and use resources and cash too early or on the wrong things (products which will not sell). While it may have seemed like a good idea at the time to keep the machines and factories running these individual efficiencies are always outweighed by large inefficiencies of the whole system.

Just as in the WSJ article, overproduction leads to excess inventory, ultimately to a downward spiral of pricing, and unhappy customers (dealers) demanding cuts so their lots are not overcrowded. This then forces car makers to reduce production or shutter factories until inventories on the dealer lots are reduced. This is disruptive to auto workers and their families, creating a further damper on demand.

Why do companies overproduce? There are many reasons, and typically it is because a decision has been made (incorrectly) that overproducing is the lesser of two evils. In the case of Detroit, the choice is between reducing production and at factories and making workers go home, or building cars that do not have customer demand. Which is worse? It is a difficult question affecting the lives of many people. Too often the decision made is a short term band-aid that does not address the root cause of the waste..

What is clear is that there is a better way. Toyota builds cars to order in their domestic market. You can order a custom vehicle and receive it in 3 days in Japan. You don't have to pick from what's there on the lot (what the car makers "push" on the market). Instead, you determine the exact specifications you want and have it custom built (customer "pull").

In order to do this, Toyota has refined its production system over many years through kaizen. Such a redesign is needed for Detroit in order for production to be responsive to customer needs and the needs of their workers and their families for a stable income.

Posted by Jon Miller at 9:04 AM | Permalink | Comments (0)

Too few companies on the Lean journey have effectively incorporated what is known as "teian" or suggestion schemes for employees' creative ideas. Although kaizen breakthrough activity may seem more dramatic and appealing, it is the slow trickle of creative ideas over many years that has kept companies like Toyota on the forefront of Lean and kaizen as well as business performance.

One type of kaizen is the "kaizen event" or "kaizen blitz" that brings together a team for several days to a week or more to radically redesign workflow. The other type of kaizen, the foundation of persistent long-term improvement, is the creative idea suggestion system.

Over the last several years, Toyota has been reporting anywhere from 10 to 12 creative ideas implemented per person, per year as part of their suggestions scheme. This has been going on for over 50 years at Toyota. It is a part of the fabric of their culture, training system, and management.

In Japan alone, there were 580,000 ideas implemented in 2003. Millions and millions of creative ideas are being implemented every few years by thinking Toyota employees. How could any company manage so many suggestions? It would seem like supervisors and managers would have not time to do anything but review the suggestions and approve or deny them. Part of the secret lies in the fact that Toyota reports at 99% implementation rate for kaizen ideas by employees.

Supervisors and managers encourage and even require thinking about kaizen ideas by their staff. This involves training and coaching so that ideas that may not be good suggestions are developed over time into ideas that can be implemented, at a rate of about one per month. It is part of a supervisor's daily job to help his or her team think about their work and find ways to make it better.

Many of these 'teian' or kaizen ideas, are small improvements in quality, safety, productivity, environmental management, etc. Some of these ideas may be the type of improvements we make without thinking about it, or writing them down to keep track. Some of them may save a $1, and a few of them may save thousands of dollars.

Almost every manager at a company with a healthy suggestion program will say that the financial results from these creative kaizen ideas are not what is most important. What is more important is to have everyone thinking everyday about improving their work, everyday.

Toyota has in effect turned every worker into a "knowledge worker", to some extent, through their commitment to the creative idea suggestion system, or kaizen by everyone everyday.

Posted by Jon Miller at 7:33 PM | Permalink | Comments (3)

The Toyota managers who share their insights with us on our study missions to Japan tell us that there are two things that are part of every Toyota' employee's job. They are:

1) Follow the standard
2) Find a better way

This is the essence of kaizen. These simple yet profound rules are what drive every employee to maintain safety, quality, low cost, and on-time, and strive to make it better. It sounds so simple, yet how many of us who think we've made good progress on our Lean journey could say that our organizations live by these rules?

Taiichi Ohno said "Where there is no standard there can be no kaizen (improvement)." When the fastest, safest, best quality, repeatable steps have been identified that is documented as the standard. That is now the record to beat. Kaizen teams find and eliminate unreasonable/unsafe working conditions waste, variability, and waste (muri, mura, muda).

The term "standard" can be misunderstood as something rigid, unchanging, absolute. If it is misunderstood in this way, it becomes an obstacle to kaizen.

Take the example of a 1st tier automotive supplier of rubber products. After redesigning the assembly lines and implementing one-piece flow, it came time to create Standard Work.

The production manager who had been actively participating in kaizen resisted documenting Standard Work. When finally confronted, he explained that he didn't want a published standard time because he wanted to keep challenging guys to beat their times and get higher production in fewer hours. What he was talking about was "the record" you had to beat to have your picture up on the "wall of fame" at the factory. Standard Work is the method used to achieve that record and must be redrawn each time the record is broken. It is how you train to beat the new record.

Machine operators in a candy factory had concerns with establishing standards settings on their production lines. Their reasoning had to do with the fact that in the past management had wanted standard followed strictly and not adjusted. This was an attempt to maintain control and keep quality at a certain level. Yet without the "Find a better way" element, it was not kaizen.

The second part of their concern had to do with the fact that now that they had freedom to change settings, they did often based on variation in the quality of the cooked candy, the recipe, etc. The fact that this level of variation exists and requires constant adjustment is a waste of processing, and demands that standards are set, followed, and improved.

It was not an easy process, but once we listened to these points and discussed how setting and improving standards could address their concerns, the kaizen efforts were back on track.

Another example is during a series of Lean Enterprise overview sessions conducted to train engineers on how to apply kaizen in their areas. We were warned by the Lean Manager "Whatever you do, don't mention standards to engineers." We were puzzled, and discovered that yet again there is a strong belief that standards get in the way of creativity or freedom to make a better product design or a better process.

Even in engineering "knowledge work", whatever is the most effective current method is the standard. "Most effective" needs to be based on fact. With engineering work, so little of it is measured in terms of time or quality that this can be difficult, nonetheless, standards needs to be documented and shared so that kaizen can happen.

Posted by Jon Miller at 10:40 PM | Permalink | Comments (4)

A client of ours has recently begun implementing TPM (Total Productive Maintenance) as the natural next step in their Lean journey. Their goals were to lengthen mean time between failures (MTBF) as well as replace time spent on reactive maintenance with improvement maintenance.

As a first step, there was a renewed focus on 5S, looking deeper into cleanliness and organization around the machines. Next came Autonomous Maintenance, or making the machine operators responsible for early detection of trouble with their machines through cleaning, daily inspections, and lubrication. This is typically where one meets resistance, and we did.

The machine operators objected at first, feeling as though they were being asked to do more work in their already busy day. "When are we going to find the time?" and "We've got to make our numbers, we don't have time for this." These concerns were heard from the kaizen team members tasked with implementing Autonomous Maintenance on the first machine.

Fortunately these concerns are easily addressed. First, the inspection and lube activity that are part of TPM are not meant to take hours each day. In fact, most Autonomous Maintenance routines are 5-minute deals where clearly marked gages and inspection locations are given a quick look only to make sure that all is normal. Weekly and monthly checks take more time, but are less frequent.

Second, the concern of "We won't make our numbers if we take time out to clean and inspect" goes away when you demonstrate that 80% of unplanned breakdowns can be avoided by these Autonomous Maintenance actions. Machines break when time is not taken to fix things that are starting to wear, and these breakdowns happen when you are busiest and it is the most difficult to recover from them.

Even if you do not have good maintenance and downtime history to demonstrate this, most kaizen teams will have at least one team member who can remember a costly downtime situation that could have been avoided through TPM and Autonomous Maintenance.

TPM is really a Lean system in itself. TPM helps build a firm foundation for companies to become Lean and sustain these gains. It involves and empowers the machine operators, requires cross-functional involvement to succeed. Autonomous Maintenance is a critical first step, and buy in from machine operators is critical.

Posted by Jon Miller at 8:47 PM | Permalink | Comments (3)

I learned a lesson about how easy it is to assume people understand something that sounds simple to you. Near the end of a recent Lean study mission in Japan, one of the participants who is a Six Sigma Black Belt, a dynamic change agent, and a PhD made the comment "We didn't see any one piece flow this week."

While he agree that each company showed excellent examples of Lean and kaizen, he had not seen any one piece flow. This stunned several of us at the lunch table. We had been to Toyota, among others, where things flowed one at a time. We argued heatedly for a few minutes until we realized the source of the confusion.

One piece flow does not mean there is only one piece of work on the line, or that there is only one piece of work between one worker and the next. For various reasons there can be more than one piece between processes (cure times, automatic machining process, transportation distance between processes, etc.). When these pieces are placed there logically and mathematically (based on takt time and cycle time) this is what is called Standard WIP.

Our very intelligent and observant friend had seen this Standard WIP, and not knowing what it is, had thought that there was "no one piece flow".

The strict definition is that one piece flow is that each person or process only works on one piece at a time before it is pulled downstream, one piece at a time, to the next process. If you have multiple pieces of work in front of you that you are working on, by definition this is not one piece flow.

There is a fine distinction between "fake flow" and one piece flow with Standard WIP. If you have material or work pieces between processes for no good reason (such as auto cycle machining, cure / bake / set time, or a pull-based transport) then it is fake flow. If the material is there in a set quantity that is a factor of takt time divided into cycle time or time-based work constraint, it is still one piece flow.

Ironically, you can have one piece flow through a batch process so long as it is one piece going in and coming out of the process. This is a related, though different topic.

Suffice it to say that with a simple demonstration with sugar cubes the Black Belt was convinced that he had been seeing one piece flow all week.

Posted by Jon Miller at 9:04 PM | Permalink | Comments (0)

We received a nugget of wisdom from Mr. Nojima of the Logistics group of Yazaki (makers of Creform and Japan's largest privately held company) during our October study mission to Japan.

Mr. Nojima made a statement that "all work has waste in it". In fact, he made the point that you might as well think of all work as waste since there is just so much waste and non value-added in what we do.

One of our study group wanted to know how Mr. Nojima's students reacted to comments like that. He said "they become angry". So how does Mr. Nojima explain this in a way that helps them see?

He gave an illustration of how he gets people to see, using the pull system. He said it is a simple comparison of 'delivering' versus 'going to get'. As long as you are delivering something, you are pushing material rather than pulling. If you stop delivering and required the customer (downstream process) to go pick it up, that's true downstream pull.

This is impractical if the customer process is located far away from the supplier process. The custome ends up walking too much to pull the material needed. This requires the processes to be brought closer together in order to save steps for the customer pulling, or 'going to get'.

If you allow people to 'deliver' or push, you will never reduce the distance, since the person doing the delivery may not see their job of transportation as waste. By insisting on a downstream pull, it forces the situation to change, connecting the process and reducing transportation and motion waste. This results in kaizen.

It was a very simple explanation. It illustrated the pull system very elegantly. It also showed how you can motivate people to change when you require your organization to operate on Lean principles.

Posted by Jon Miller at 5:43 AM | Permalink | Comments (0)

The rule of thumb to have a good mix of kaizen team members from different areas is: 1/3rd of the people from the area or process targeted for kaizen, 1/3rd of the people from upstream or downstream processes (customers and suppliers) and 1/3rd from areas that are outside of or not strongly related to the process, such a finance person for a kaizen in the machine shop.

Sometimes kaizen team members ask "Why aren't there more people from the area on the kaizen team?" Or, "I don't know anything about this process, why was I chosen to be on the kaizen team?" These voices are heard when Lean Champions follow the rule of thumb for getting a good cross-functional mix of people on the kaizen team, without explaining the reasoning behind it.

Half of the purpose of kaizen events is education and half of the purpose is to get business results. By having people on the team that will not work in the area that is the target for kaizen, you are in effect educating people from other areas about kaizen and Lean. These people begin to spread the word, supporting communication and education.

People from outside of the process where the kaizen is done will bring outside perspectives. When you have people who know nothing about machines to a SMED event, it forces the process owners and experts to explain things in simple ways so that "even the finance guy" for instance, can understand. This helps group brainstorming.

The team members from outside the are also bring skills that might not be relevant to running the machine, but may be useful in other ways. Taking the "finance guy" example, he may be able to explain the financial impact of higher inventory turns and motivate the team towards SMED.

Team members from outside the area will not have pre-conceived notions or fixed ideas about how the process should be done. The less experienced team members may have some good common sense insights, without knowing all of the reasons why "it can't be done". They help by asking the questions about processes and assumptions people have and why things are the way they are. Ideally, this asking of questions (remember, there are no stupid questions) leads to the discovery of quick kill improvements.

Pulling in kaizen team members from other parts of the company (or even suppliers and customers) helps to keep from overburdening one department or work group. If the target of the Lean activity is the assembly line, you don't want to have the entire kaizen team be made up of assemblers. You will stop the line. You need knowledgeable process owners and experts, but you also need to borrow people from other areas. In return, you educate these people. And someday you may lend an assembler for a kaizen in the office.

Perhaps most important, this cross-functional team approach to kaizen enhances Value Stream awareness and thinking by making people aware of upstream & downstream relationships in business. Kaizen events should be breakthroughs, or ‘flow’ improvements rather than ‘point’ improvements that can be made within the department by everyone on a daily basis. The types of kaizen that improve flow and speed up the process of creating value for the customer are best done when the team looks at a series of processes as a section of a Value Stream.

Posted by Jon Miller at 9:26 PM | Permalink | Comments (0)