Toyota Kata : Managing People for Improvement, Adaptiveness and Superior Results

by Rother, Mike

  —Takahiro Fujimoto1

  Comparison with Our Current Management Approach

  If the process level is the fractal at which continuous improvement and adaptation can occur most effectively, then organizations that are able to improve constantly and systematically at this level should, in crowded market situations, realize a competitive advantage. If so, then this has implications for both management and management education.

  Many companies experience a subtle disadvantage when it comes to continuous, incremental improvement and adaptation, because they rely heavily on managing by setting outcome targets, reporting of metrics, incentive schemes, and ROI-formula-based decision making. The evidence is mounting that, by themselves, management by objectives—at least as we currently practice it—and formulaic decision making do not make an organization sufficiently adaptive and continuously improving for long-term survival in highly competitive markets.

  One problem is that reported numbers arrive after the fact, are manipulated to look better than they are (because of incentives), and, as Professor H. Thomas Johnson points out, are only abstractions of reality. Metrics are abstractions made by man, while reality is made by nature. Only process details are real and allow you to grasp the true situation.

  Many executives and managers—reinforced by their MBA education—put their faith in those quantitative abstractions, pursue financial outcome targets, and in many instances have lost connection with the reality from which those abstractions emerge. Decision makers are poorly informed about the actual situation, and as a result they make incorrect assumptions, set inappropriate targets, and do not see problems until they have grown large and complex.

  Managing from a distance through reported metrics leads to overlooking or obscuring small problems, but it is precisely those small problems that show us the way forward. Overlooking or obscuring small problems inhibits our ability to learn from them while they are still understandable, and to make timely adaptations in small steps. Over time this can adversely affect the company’s competitive position.

  I meet many managers, executives, and academics who continually hunt for the right mix of performance metrics that will stimulate Toyota-style continuous process improvement. This may seem logical from the perspective of the current management paradigm, but those metrics simply do not exist. There is no combination of outcome metrics and incentive systems that by themselves will generate continuous improvement and adaptation.

  Setting targets and performance metrics alone usually does not generate the desired behavior or result in real improvement of work process. And how could it? The people trying to achieve the quantitative targets are not taught or guided by any sort of improvement kata. My colleague Robert Austin has studied this phenomenon and makes it nice and clear with the following comments:

  The manager relies on signals that he or she assumes are good results measures. In fact, the employee knows ways to make signals look good that the manager hasn’t thought of and that have nothing to do with results.

  Another lousy feature of such systems is that they punish workers who have too much integrity to game the measures.2

  If we want our organization to be adaptive and continuously improving, we should develop ways of maintaining more focus on the details of the real situation in real time. Toyota’s improvement kata does this well. It provides a means for people to work empirically and creatively toward objectives that may not be easily or readily achievable, and that would often not initially pass one of our formulaic ROI decision-making calculations.

  Toyota’s shop floors are not connected to the IT system.

  Managerial accounting control systems can exacerbate the negative effects of managing from a distance via metrics, since reported data arrives late and leaders interact even less with the reality of the situation. This is why accounting control systems have little or no place on factory floors at Toyota. Factory leaders at Toyota do not refer to accounting reports to get an understanding of a situation. They are taught to go and observe the situation firsthand. In order to develop and guide good improvement practice, Toyota leaders interact with the unfolding situation at the process level, by following the improvement kata.

  Figure P3-6. IT systems have little place on the factory floor

  The factory in Figure P3-6 is treated like a “black box,” but of course a large amount of process performance data is utilized in Toyota plants; on thousands of charts, boards, documents, alarms, etc. However, this data is maintained near the place of occurrence, and leaders have to go to the process to get the information they need and understand the situation there. To manage an organization with the improvement kata, many leaders may have to organize their workdays differently. There is an organizational impact.

  In summation, the improvement kata gives people a means for working together. Consider, for example, some managerial concepts of the late twentieth century proposing that managers and leaders should seek out and respect the ideas of their subordinates. I have witnessed dozens of organizations that sincerely tried to employ this logical but vague advice and got nowhere with it, or worse. In an unmanaged— or “self-directed”—environment, the scope of ideas about what to do is often so wide-ranging and even conflicting that it frustrates progress. In contrast, when groups of people strive for a target condition—not just an outcome metric—and also have a common routine for working to achieve it, then they are brought into a channel that focuses their thinking and taps their capability. Not only does this make it more practical to seek out and respect other people’s ideas, it makes it natural to do so.

  Here is an interesting observation to consider. At Toyota, how to act in going through the improvement process is defined by the improvement kata, whereas the subject matter is open and varies depending on what one is working on. To a degree this is the opposite of how we so far have been trying to emulate Toyota: we defined the subject matter, the production techniques like kanban or heijunka that were to be implemented, but left “how to act” up to everyone to decide for themselves.

  How does Toyota ensure that everyone in the organization learns and follows the improvement kata? That is the subject of Part IV.

  Notes

  1. Takahiro Fujimoto, The Evolution of a Manufacturing System at Toyota (New York: Oxford University Press, 1999).

  2. Jim Austin, “Robert Austin: An Interview,” Science Career Magazine (April 26, 2002).

  Part IV The Coaching Kata: How Toyota Teaches the Improvement Kata

  Introduction to Part IV

  Once we become aware of Toyota’s improvement kata, described in Part III, it gives rise to several new questions, such as:

  How do we teach everyone in the organization the improvement kata?

  How do we ensure people are engaged in the improvement process and utilize the improvement kata correctly in their daily work? How will we know what skills individuals need to work on? How do we ensure that appropriate challenges/target conditions are developed?

  How do we ensure that the PDCA cycle is carried out correctly and effectively?

  How will we ensure that leaders have a grasp of the true situation at the process level in the organization?

  How will we pass on the improvement kata from generation to generation?

  Toyota’s answer to these questions is its coaching kata, which is the subject of Part IV. The purpose of the coaching kata is to teach the improvement kata and bring it into the organization. We will look into the role of managers and leaders at Toyota in teaching the improvement kata to everyone in the organization and making that kata work as effectively as possible every day. Part IV is not about how Toyota trains production workers in their jobs. It is about how Toyota works to develop and maintain improvement kata behavior across the organization.

  In Chapter 7 we will first take a brief look at who is actually applying the improvement kata to production processes in Toyota factories. There have been a lot of misconceptions about this. Then, in Chapter 8, we will l
ook at how application of the improvement kata is taught and managed at Toyota.

  Chapter 7

  Who Carries Out Process Improvement at Toyota?

  Figure 7-1. Working toward a target condition

  A question that has been debated for many years is: “Who should carry out process improvements?” (Figure 7-1). Here are three common but problematic answers to that question.

  1.The process operators? One of the widely held opinions about continuous improvement at Toyota is that it is primarily self-directed, with teams of production operators autonomously making improvements in their own processes. Some typical comments along these lines are:

  “The operators are closest to the process and are empowered.”

  “How can we get our line operators to solve problems?”

  “How can we make continuous improvement run by itself?”

  Operator autonomy is a commonly held and unfortunate misconception about Toyota’s approach. It is not at all how operators and improvement are handled at Toyota. For one, it is unfair and ineffective to ask operators on their own to simultaneously make parts, struggle with problems, and improve the process, which is why Toyota calls autonomous operator-team concepts, “Disrespectful of people.” It is physically impossible for production operators to work fully loaded to the planned cycle time in a 1×1 production flow and simultaneously make process improvements. Furthermore, many operators are just beginning to develop their understanding of the improvement kata and their problem-solving skills. There are currently no autonomous, self-directed teams at Toyota.

  This does not mean that we should not empower or engage process operators. In fact, teaching people the improvement kata by engaging them in it is critical to Toyota’s success. It only means that concepts like self-directed work teams are not such an effective way for an organization to empower and engage people.

  2. Leave it to chance? I have not heard anyone actually give this answer, but in many cases our comments and actions—comments like these—indicate this is exactly what is happening:

  “Andon gives everyone in the plant information.”

  “This alerts everyone that there is a problem.”

  “Any person walking through the area can see …”

  The number of andon-style warning-lamp systems that have been installed in our factories in the last 20 years, for example, is astonishing. Yet in many factories the red lamps are lighting up and no one is responding. The basic point here is that if we assume anyone (or everyone) is responsible, then no one is responsible.

  3. A special team? As we have already discussed, this will not work if we want improvement to occur at every process every day. At Toyota, factory staff includes no specific continuous improvement agents. The improvement kata is embedded into every work process, and everyone is taught to work along the lines of the improvement kata.

  Who Does It?

  In schematic form, a typical Toyota factory’s line functions are organized as shown in Figure 7-2. There are, of course, additional support functions such as maintenance and production engineering that are not shown, but this diagram is detailed enough for our purposes.1

  Figure 7-2. Schematic of Toyota factory line organization

  In 2004, Professor Koichi Shimizu of Okayama University published a paper about continuous improvement of production processes in Toyota factories. In his paper, Shimizu classifies process improvement activity at Toyota in two categories:

  Improvement carried out by production operators themselves through quality circles, the suggestion system, and similar initiatives. Shimizu calls this “voluntary improvement activity.”

  Improvement carried out by team leaders, production supervisory staff, and engineers as part of their job function.

  There are some surprises in Shimizu’s paper (Figure 7-3). Specifically, his research suggests that only about 10 percent of realized improvement in productivity and cost at Toyota comes from the first category, whereas about 90 percent comes from the second. In addition, the main purpose of the first category—improvements carried out by production operators themselves—is not so much the improvement itself, but rather to train production operators in kaizen mind and ability, and to identify workers who are candidates for promotion to team leader. The purpose of the second category of improvement, on the other hand, is clearly cost reduction via diligent and constant improvement of productivity and quality.

  Figure 7-3. Who carries out process improvements at Toyota?

  Source: Koichi Shimzu,“Reorienting Kaizen Activities at Toyota: Kaizen, Production Efficiency, and Humanization of Work,” Okayama Economic Review, vol. 36, no. 3, Dec. 2004, pp. 1-25.

  What I have been able to learn so far about who makes process improvements on Toyota shop floors fits with Shimizu’s findings. The great majority of shop floor improvement in a Toyota factory is generated by the functions circled in Figure 7-4. These team leaders, group leaders, superintendents, and various levels of manufacturing engineers are the primary people who apply, and coach application of, the improvement kata to production processes. This process improvement activity represents well over 50 percent of their work time, which is not surprising since at Toyota the improvement kata is actually a way of managing.2

  Toyota production operators, called “team members,” are of course also regularly involved in making process improvements, but these are usually improvements in the operators’ immediate work envelope, which are carried out in collaboration with, and under the guidance of, the team leader. It’s the responsibility of team leaders to encourage and get improvement suggestions from their team members, and, conversely, operator promotion to team leader is determined in part on how much improvement initiative and skill an operator demonstrates. Both operator and team leader, in other words, have incentive to work together on process improvement.

  Figure 7-4. In the factory organization, process improvement activity is mostly here

  Working with Target Conditions

  In the case of a new process or product, management sets a target cost and a target date for production. The first process target condition (that is, work standard) is typically established by the process’s group leader and a production engineer. This is then given to the production team (the team leader and team members).

  As the production stage begins, the production team and their group leader work to achieve that target condition, which can take several weeks. Once regular production stabilizes, then further target conditions, called “standards” or “targets,” are developed:

  Group leaders, team leaders, and team members focus on target conditions in their process, and on understanding and resolving daily production problems.

  Themes, targets, plans, and initiatives are announced by senior management. These are worked down into the organization via mentor/mentee dialogues (more on this in the next chapter), and are converted into process target conditions. The conversion of outcome targets into process target conditions generally begins at the “superintendent” level. Managers at this level ensure that target conditions, improvement efforts, and projects at individual processes follow improvement-kata thinking, fit together for a flowing value stream, match with the organization’s targets, and serve customer requirements.

  Responding to Process Abnormalities

  A common way of reacting to process abnormalities in our factories is to have production operators record them, so they can be compiled into summaries and Pareto charts. Sounds like a good idea, but it is not effective for improvement. I once listened to a plant manager proudly explaining a Pareto chart of problems and how the top problem was being worked on. One of my colleagues said in response, “Oh, and the rest of those problems you are shipping to the customer?” which I thought was a pretty good insight.

  The information provided by Pareto charts usually comes too late to be useful for process improvement efforts. By the time a problem has risen to the top of a Pareto chart, it has already caused a lot of damage and grown compl
icated, the root-cause trail is cold, and we become involved in analyzing postmortem data instead of understanding what is actually happening on the shop floor now. It is interesting to note how often the largest category in a Pareto chart is “other,” that is, an accumulation of smaller problems.

  This does not mean that Pareto charts should be abolished, but that they should not be thought of as our first choice for becoming aware of and dealing with process problems.

  Here are two aspects of how Toyota thinks about dealing with process problems:

  1.The response to process abnormalities should be immediate. Why?

  If we wait to go after the causes of a problem, the trail becomes cold and problem solving becomes more difficult. We lose the opportunity to learn.

  If left alone, small problems accumulate and grow into large and complicated problems.

  Responding right away means we may still be able to adjust and achieve the day’s target.

  Telling people that quality is important but not responding to problems is saying one thing but doing another.