Genchi Gembetsu

The Vigorous Learning Enterprise Robert W. "Doc" Hall With the publication of The Machine That Changed the World in 1990, lean became the accepted descriptor of the Toyota Production System(TPS). Earlier American practitioners called it "Just-in-Time," the name Toyota first used to describe TPS in English. "Just-in- Time" misled many to think that just cutting inventory levels created magic. Lean, while more holistic, is all too easily interpreted as an anorexic employment level. "Continuous improvement" implies dealing only with small problems. Whatever name is used, the financially-minded are apt to regard it as merely a super way to cut costs, for our system of management is as dysfunctional as our old systems of work. No term fully describes a different way to work, to develop a work organization, and even how to define what a company is. It's a tough idea to get in mind. Today, Toyota refers to TPS as the "Thinking Production System." That thinking carried over into product development and other systems, and it continues to evolve as Toyota struggles to imbue it in a global, multicultural workforce. Today Toyota is vastly different from when the principles we study were incubated in a little company whose supply network did not extend much beyond Aichi Prefecture, Japan. Although lean thinking is far from mainstream, versions of it and fragments of the thinking have entered companies all over the world. Many mix "tools" from TPS, Six-Sigma, and other methodologies into their own system. However, feedback from Target readers and others point to two major problems: 1. A TPS (or lean) mind set clashes with that of financial systems and the general business rationale by which companies are fundamentally managed. 2. Regression is easy; initial gains are hard to sustain. These problems and others ball up in that wooly bear called "creating a lean culture." Eliminating the clashes requires changing the culture of an entire work organization. Only top management can effectively address this, but even committed CEOs have difficulty envisioning it. Misunderstanding the purpose of TPS blurs the vision for this. It certainly is not In Brief This connect-the-dots review draws on a number of prior Target articles to propose a framework for deepening and extending "lean thinking." This thinking stalls when tacked onto conventional business logic for running a company, which is as dysfunctional as a system that does not empower workers. Getting past this is a cultural change that appears necessary given the challenges and complexities that companies now face, but cultural change is not for the faint-hearted. 5 First Issue 2008 cost cutting as usually understood. Lean tools do transform processes to eliminate waste, but their deeper purpose is to transform people to improve processes almost autonomously. Lean tools expose many more problems than they fix. Unless people engage in daily problem solving to correct the problems that are exposed, the waste of their human abilities remains the mother of all those process wastes. The objective of lean is to develop people to develop great processes. These should serve customers exceptionally well. If they pay more than the processes cost, the company should make money. Sounds logical, so why is it hard to do? Because the culture in which this is done differs substantially from that which prevails in business - and most other work organizations. Tweaking systems to better support lean methods is not enough. Patterns of thought change dramatically throughout the company. Rather than the financial language of business, a common problem-solving language and discipline begin to glue the organization together. Let's call this a vigorous learning culture. (The term "vigorous" implies an organization that does things, not an academic one.) No company has shifted culture this dramatically, but a few have made a good start. Webster Plastics, Autoliv, Ventana Medical Systems, and others that have been featured in Target provide examples in progress.1 Figure 1 is a high-level framework illustrating common points, but a lot of explanation must be added to make it meaningful. Disciplined Learning Processes Lean tools are best regarded as learning tools. Using them to improve processes by eliminating the most obvious waste only begins the learning. The heart of the change is developing people to solve problems, eliminate waste autonomously, and teach others how to do this. As with learning music or a sport, the primary method of development is consistent practice. For that reason, Toyota refers to "creating" TPS, not "implementing" it. To develop people, Toyota introduces TPS by first coaching how developing and holding standard work is the basis for process improvement. Beyond that, their objective is to create better and better systems of common problem- solving for learning. Value stream maps, 5S, SMED, visibility systems, pull systems, and related techniques expose problems, preferably well before they result in major errors and waste. That does little good unless people eagerly hop right on the problems exposed. But to do that, they need well-honed skills using solid problem-solving logic. Many problems quickly yield a countermeasure by careful observation while "asking why five times." Others need more experimentation guided by Plan-Do-Check-Act (PDCA cycle or equivalent logic), using basic data collection and analysis techniques (like check sheets, fishbone, Pareto). Other problem-solving tools are unique to a company's technology. Examples of techniques from "non-lean" sources are After Action Reviews and the SOL Coaching Model.2 Everyone tending small problems prevents most of them from becoming bigger ones. This seems so obvious, why is it ignored? One reason is that for many people it doesn't seem emotionally satisfying. Solving big problems - putting out fires - is heroic; preventing them isn't. Often business logic can't make sense of it either. Cost systems can't assign a hard dollar figure to events that did not happen. Therefore they can't assign a firm value to disciplined learning processes that enable people to prevent them from happening. Common problem-solving tools are as much a part of lean as the creation of work flow. They are the structure of a common language in an organization that has disciplined learning processes, which partially defines a vigorous learning enterprise. Learning occurs in at least two different categories, refinement and exploration. Refinement improves existing processes using different names - continuous improvement, small-step improvement, kaizen, streamlining, tuning up. Normally the objective is to reduce process variance by eliminating waste, but ideas for small 6 Target Volume 24, Number 1 The objective of lean is to develop people to develop great processes. improvements may be very imaginative, and their cumulative effect very impressive. Exploratory learning goes for breakthroughs, innovation, disruptive improvements, or just curiosity, as in basic research. Instead of reducing variance, one examines it, investigating deviations or new phenomena; maybe looking for something to exploit. Both in technical research and research of customer processes, one may seek ideas far afield. Indeed, one definition of creativity is associating old, but different concepts in some novel way. Another is looking at a problem from a dif- 7 First Issue 2008 This figure takes study. Key points are: 1. Everyone, especially top leaders, needs to "see" situations beyond the core work organization and even beyond its total enterprise, consisting of all stakeholders. Vision must penetrate the fog of financial representations and systemic rules. 2. The organization needs both A) a rigorous, disciplined learning system (tools, techniques, methods of analysis, common language, and common knowledge libraries); and B) well-practiced behaviors that enable them to effectively use this system to work through problems together. 3. If sub-groups must act in concert while acting autonomously, all need a clear understanding of the organization's common mission (what is important to do regardless of rewards), plus any current goals for unified change or improvement. Otherwise, people cannot exercise discretion when it is necessary to stretch the system for something important. 4.Working in this way is unnatural because humans are "political animals." The discipline and behavior to objectively observe work and change processes must be built into the culture and constantly reinforced by it. Otherwise, we quickly regress into recrimination, inventing conspiracies, and "gaming the system." Figure 1. Overview of a Vigorous Learning Organization . ferent angle or in a different context to see the problem as an opportunity. Another difference in learning is induction versus deduction. Induction looks for patterns in existing practice, and draws conclusions from them. Deduction forms rational hypotheses and tests them.3 However, when done rigorously, all these kinds of problem solving follow the scientific method, as with PDCA. More important, no one learns this thoroughly until they have to teach, coach, mentor, and lead other people in it. Improve from a Knowledge Base Most companies have difficulty holding operating standards, and R&D may reinvent the same wheel several times - and possibly a wheel that doesn't work because no one documented negative findings where others can easily find them. Keeping the current state of practice or knowledge pegged is part of a learning discipline. Doing so eliminates much of the waste of solving the same old dog problems over and over. To stop this, improve organizational learning processes in operations, R&D, and elsewhere. In production (or other operations), a rule is to always improve from an existing, documented standard; never from what you see now. If the process being used has not been documented, do that first, restraining the impulse to "get going." When a new method results from kaizen, have those who do the work document it so that they understand it well enough to train someone else how to do it. When workers do this, how work is actually done is more likely to accord with its documentation. (When they can't, it is a team leader's responsibility.) Furthermore, everyone is more likely to do it the same way, especially if they understand that this is the base for further improvement. And improvement is just as much the workers' responsibility as the tasks they perform. Making this discipline a habit takes time. Toyota fosters it by coaching people to train others using the Job Instruction part of Training Within Industries (TWI).4 (This eliminates a lot of human variance in processes even if a company is not otherwise lean.) Autoliv does something similar. At Webster Plastics, where polymer processing created the value added, people were asked to "think like the plastic," and come up with a "tip of the day." At the end of each quarter, the best ideas became new standards reinforced by a training session; then everybody started another round of improvement. To avoid regression, organizations need a discipline to hold gains built into the learning process. Besides self-documentation and cross-training, visibility cues should alert people when they are not holding standard work or violating a process standard. If they can't hold it, changes are incomplete. While no methods to do this are perfect, without this discipline, gains fall into a black hole. Logic-based learning systems come in many variations, but if they don't address regression, they lack discipline. Problem documentation and work instructions in standard format, whether in paper or digital form, become a "living" process knowledge base showing the logic behind why and how a process is done the way it is. It should be easy to access. When a problem occurs, it should answer, "Have we seen this before? If so, what was done and why?" A standard format simplifies communication. It also guides efficient problem solving. But no system is effective if people are not coached, or if they neglect documenting their knowledge into the system. Both giving input and using the output are part of the discipline. Other practices also contribute to a vigorous learning work culture. For example, let people take equipment apart and reassemble it to learn how it really works. Even with sensitive instruments, increase the number of people that can do this. The same with software; if people don't understand the coding logic embedded in processes, their ideas for improving them are foggy. The objective is to develop a workforce able to deal with challenges that others can't imagine. In a high tech world, this is obviously not getting easier. Evidence that first line workers are into this 8 Target Volume 24, Number 1 More important, no one learns this thoroughly until they have to teach, coach, mentor, and lead other people in it. include signs of problem-solving on the job, and detailed improvements at each workstation, not just in the main flows. The need for a vigorous work culture extends well beyond internal operations and the boundaries that financially define a company. One can never know too much about customers and their environment (including associated ecological issues). Having a heavy technology base on which to draw is equally important. Customers cannot understand everything they need -too much specialized detail - so work organizations have long had to do much of this for them. Health care and automotive safety are two obvious examples. Because of this, a robust knowledge base for marketing, product development and operations is vital. For example, Delphi has a database of automotive crash profiles very useful for OEMs checking the safety of new designs while still in CAD. Were it an asset on a balance sheet, its value would be only a guess. The values of a vigorous learning system and knowledge base are even squishier, but they are key elements in designing new products the Toyota way, integrating proven technical solutions to address customers' needs, like Teledyne Benthos.5 This minimizes risk in new product development. More important, it illustrates that having a clear, organized understanding of a company's core competence is becoming vital. And a vigorous learning system has to keep that knowledge base refreshed. For example, Webster Plastics' engineers constantly trolled customer locations looking for opportunities to assist them, learning a lot and sometimes designing a Webster part into a solution. Ventana has links to scientific research in every part of the world. For organizations dedicated to superior performance, building a superior learning process will not remain optional much longer. Copying best practice does not go deep enough. They have to create a disciplined learning process to fit their unique needs. But for this to be effective, leadership has to reduce the waste in human behavior that impedes collaborative learning. Cultivating the Behavior Together, lean tools and problem-solving tools structure a common learning language. A vigorous learning culture must speak it fluently. However, a language consists of more than a logical syntax. Its idioms are a cultural medium, and cultures have norms of behavior. Successful kaizen by a group of people depends on everyone focusing on a common process, not seeking culprits associated with it. Otherwise kaizen doesn't get past the first "why" question. Collective problem solving requires everyone to 9 First Issue 2008 Figure 2. This diagram originated from Partners in Leadership, LLC. STEPS TO ACCOUNTABILITY® are its registered trademarks. TTHHEE LLININEE confusion/tell me what to do wait and see cover your tail it's not my job finger pointing ignore/deny see it own it solve it ABOVE THE LINE® do it STEPS TO ACCOUNTABILITY® THE BLAME GAME® BELOW THE LINE® ABOVE THE LINE®, BELOW THE LINE®, and Above the Line and Below the Line Behaviors regard processes and problems as common, neutral ground. Nearly all teambuilding struggles with this until members "sort of understand" it, but an example that draws a picture of it is Ventana's abovethe- line and below-the-line behavior (Figure 2). Adding these phrases to the language makes emotional discord visible. Uttered during any interchange, they help resolve conflict. Willingness to dig for facts and abide by them is the logical path to resolving conflict. Ability to engage in dialog (respectful exchange) is the emotional path to resolving conflict. And resolving issues is the reason for problem solving. Mere use of scientific methodology does not remove behavioral baggage from: seeking the context of problems; looking at data; going to the source to see the facts; letting facts guide conclusions; standardizing countermeasures in practice. The discipline to do this runs counter to the political instinct normal in any organization. It cannot be totally suppressed. Scientists who supposedly live scientific methodology have personal squabbles and jealousies. Collective problem solving is not done through harmonious passivity, but by enthusiastically working a disciplined learning process without fear or favor. The emotional waste impeding this process is just as real as any other waste, but unfortunately, well-meaning people creating the waste are often unaware that they are. Creating and sustaining a work culture for effective discipline learning processes takes a different kind of leadership. Bringing in a non-comprehending leadership (or ownership) undoes it quickly. In both Ventana and Autoliv, a key point is that standard work for every leader at every level includes routines to reinforce the behavior of the learning culture almost daily. Sustaining vigorous learning is that hard. At every turn, supporting vigorous learning contradicts prevailing business convention: how people are selected; how they are developed; what leadership requires; even what success means. Making money is a necessity, but it's directionless as a guide for operational learning. Instead, success is delivering on the mission better and faster by learning to do what no one else can do. Profit becomes a process specification. For example, that is how it is treated in target costing for product design. An example of this kind of mission statement is Ventana's: "to improve the quality of life." That's altruistic, but vague, so the number one goal under that clarifies the learning objective: "find cancer faster." For inspiration, that's hard to top. Organizational Control Vigorous learning requires free lateral communication in all directions, so it is incompatible with rigid, command-andcontrol hierarchy. Strict hierarchy has become so cumbersome today that work organizations consist of projects, networks, committees, teams, partnerships, and other eclectic groupings familiar to most readers. But control is maintained through bureaucratic budgets, promotions, and reward systems. This is becoming so dysfunctional that the waste in budget systems is becoming obvious to financial executives who are leading initiatives to eliminate detailed budgets.6 If detailed budgets disappeared, so would the accounting variances that are the bane of process improvement. However, accrual accounting would still bury the effects of recent changes on cash in a system that predicts current financial status assuming that its implicit forecasts come true. Lean leaders want simple, cash accounting for end-to-end processes (value streams). However, another attribute of these systems may be more dysfunctional. Budgets tend to smother initiatives in "Mother-may-I" routines when more autonomous flexibility is needed. Present managerial systems, in thrall to financial guidance, are as stultified as non-empowering worker systems. Suppose accounting and financial controls loosen; then what? Managers must develop true responsibility in their people - becoming a leader rather than a budget manager. But executives dependent on financial controls are terrified by the 10 Target Volume 24, Number 1 prospect that their organization would decay into cabals of complacent thieves preserving their entitlements. So displacing much of this control system with a disciplined learning system is serious business. It's necessary for big-step increases in organizational performance, but not doable in one swift leap. Mind set shifts take time. People have to train up for it. As they get better and better, the basis for organizational discipline shifts toward the learning system (with well-tuned operating processes, minimum waste, and high visibility). An example of a mind set change is financial audits of inventory, important if a company has lots of it, and leaky transactional controls over it. But if inventory is miniscule, and highly visible, so that any item out of place triggers a check on the cause by the disciplined learning system, financial implications are comparatively trivial. Frequent financial audits suggest so much waste lurking about that pockets are easy to line with it. In a vigorous learning organization with a disciplined learning process, control is quite different. The job of leadership is to audit whether the learning system is being vigorously used - whether the total workforce tends processes, improves them, seeks new ways to satisfy customers, and creates innovative ways to do that - without dictums. This auditing has to be almost daily, with immediate response to lapses. This is high-touch, high-energy leadership; budgetary control is low-touch, low-energy leadership. Vigorous Learning Organization A vigorous learning enterprise is a nexus of human relationships in neutral problem spaces - processes embodying the activities performed. That sharply contrasts with the short-form legal definition of a corporation as a nexus of contracts. The resulting transactional boundaries are barriers to information flow - and to vigorous use of a 11 First Issue 2008 Figure 3. Vigorous Work Organization "Structures" . - disciplined learning system. In our current state of progress with collective process improvement, we live in constant tension with the legal definitions of business. Everyone attempting to integrate ideas from a supply chain into a new product design is familiar with the stress of crossing intellectual property boundaries. Just to streamline the flow of logistics from suppliers, one must simplify the transactional complications. In the small Toyota enterprise in Aichi, where many ideas to do this emerged, one explanation is that, not knowing what business convention prohibited, they tried anything that made sense. Autoliv and Ventana are two companies that have started breaking from business convention, heading toward a more robust vision here called a "vigorous learning enterprise." No company is known to have crossed the culture gap shown in Figure 4 to stay. Even Toyota has trouble staying on the north side of that gap. While Figure 4 is "far out" compared with present practice, we need to push our thinking. The issue is not holding the gains from lean, which implies that our present 12 Target Volume 24, Number 1 Figure 4. Vigorous Learning Organizations Classification achievements are good enough. Rather it's how to build on that base to go much further, much faster. It's not 1985 any more; eliminating a little waste is not enough. 21st century challenges call for much more dramatic change. Two major points are intended in Figure 4. First, the basic concepts of progress by lean and quality methods need to expand into a broader arena. Vigorous learning enterprise is just a straw name to start talking about it. Second, breaking out of conventional business thought and language will be necessary to do this. For instance, people are not assets. A company is not just property. People are the work organization. And their value is not some stock cap; it is what they collectively do, and are capable of doing - or more exactly, what the processes for which they are responsible are capable of doing. (Even investment analysts vaguely sense this by trying to factor more "intangibles" into their thinking.) A disciplined learning process is an intangible of very high value, but one cannot assign a dollar value to it by conventional market tests. Leadership Toward Vigorous Learning As with an athletic team, the objective of developing a vigorous learning organization is to develop people collectively to perform at the peak of their ability. Forget ultra-lean headcounts. Go for tiptop conditioning and performance at all times. The military version of this is clear: constant readiness for any sensible mission. Financial logic muddies this by rejecting anything not yielding a payback, preferably as big as possible as fast as possible, and if money is the goal, success is apt to induce complacency. The leader has to be a servant leader: mission first; people second; me last.7 Training on problem solving logic and methods is only the beginning. Leaders have to give people responsibility for learning and process improvement, expecting them to demonstrate their use of the system. That is, managers transform into learning leaders. To begin, start asking questions, expecting better and better responses. Questioning pushes responsibility onto the respondent. Mentor people; keep expanding the contextual understanding from which they can respond. Keep this up, and in time, you no longer have to ask many questions. People automatically tell you what problems they are working on and what needs to be done. Furthermore, mentor managers to become mentors. If someone tells you that they have no problem, respond, "That's a problem." When reviewing a problem countermeasure, three good questions are: 1.What did we learn from this? 2.What did we learn to keep it from happening again? 3.What have we set up to learn more about this process? The first two questions correspond to first and second loop learning, which originated from Chris Argyris.8 The third assures that respondents are already seeking more problems. (Few scientific papers end without noting that more research is needed on a subject, and in what areas.) Questions also involve the leader in the learning process. The most important role of the leader of a learning process is modeling the behavior that promotes it. No one can be expert in every complex process; no use pretending. But leaders can become expert in the learning processes themselves, including the behaviors expected. Finally, top leaders have to explain and promote the vigorous learning enterprise everywhere within it, including to its bankers, stock analysts, and auditors. Robert W. Hall is editor-in-chief of Target and a founding member of AME. 13 First Issue 2008 References: 1. Webster Plastics was featured in "Building a Vigorous Culture," Issue 3, 2005; Autoliv in Issue 5, 2007, and Ventana Medical Systems in Issue 6, 2007. 2. After Action Reviews (AAR) originated with the U.S. Army. The SOL Coaching Model is used by the Society for Organizational Learning (Peter Senge's organization). Both have been used by companies. AAR is used by small units and large task forces, and has a behavioral component as well as a methodology. The SOL Coaching Model emphasizes dialog behavior (listening), but also has a disciplined structure. The number of such frameworks in existence is unknown. 3. Induction and deduction are terms used in so many ways as to be confusing. However, when applied to logic, induction tries to reason a general conclusion from specific cases, while deduction reasons whether a general concept applies to a specific case. "Lean" problem solving is mostly induction. In the absence of a rigorous problem-solving methodology, the hazard of deduction on a process problem, for instance, is jumping to a conclusion without making the effort to see whether observations or data support that conclusion. (Japanese sometimes refer to that in "Janglish" as being "too smart.") 4. The Job Instruction part of Training Within Industries (TWI) was briefly explained by Jim Huntzinger in "Why Standard Work is Not Standard," Target, Issue 4, 2006. But actually learning to do it requires ten hours of instruction, preferably over two weeks, because you have to practice and reflect. 5. Teledyne Benthos is one of the first North American companies to structure a learning process for new product development similar to that of Toyota. See Pat Panchak, "Teledyne Benthos Adapts the Toyota New Product Development System," Target, Issue 3, 2007. 6. See Robert W. Hall, "Eliminating the Budget: Park Nicollet Health Services," Target, Issue 5, 2007. 7. Servant leadership is inculcated in military academies, if not by that name, but it's not inculcated in business schools. However, servant leadership in business does have a long, if quiet history. It is most closely associated with Robert K. Greenleaf, Servant Leadership, The Greenleaf Center, Indianapolis, IN, 1977. Independently, a pioneer of similar thinking was William J. O'Brien, former CEO of Hanover Insurance. He promoted "lean" in business long before the term become popular in manufacturing. 8. Google double loop learning: It's poke-yoke thinking applied to organizational learning processes. © 2008 AME® For information on reprints, contact: AME Association for Manufacturing Excellence www.ame.org 14 Target Volume 24, Number 1