Dear Robert,

A few weeks ago I walked through the Arsenale in Venice, which has been in continuous use for building and overhauling military ships since 1104. The Arsenale is still an Italian naval base and visiting requires permission from navy headquarters in Rome. But my interest was not in present-day activities. It was in the remarkable history of the Arsenale as a landmark in the long history of lean thinking.

Jim Womack (right); Arnaldo Camuffo (right center), scientific director, Lean Enterprise Center Italy; and John O'Donnell (center), executive director, Lean Global Network; visit the Arsenale in Venice on November 14, 2009.

During the 1400s the Venetians made a remarkable leap in thinking about the organization of design and production. They adopted a standard design for their war galleys and began to think systematically about producing these ships in volume. By the mid-1500s they were manufacturing in advance and kitting the parts for the hundreds of ships built each year. Then, when military need dictated, pre-built hulls were caulked at the first production station and lowered into the water. They were then floated past numerous assembly stations for installing the masts, the steering, the flooring, the seats, the oars, the guns, and all of the items needed to complete the most complex manufactured product of that era. Thus we have the first known example of "flow" production in process sequence using standard designs. And by 1574, when King Henry III of France visited to observe the process, the Arsenale could caulk and final assemble a galley in only an hour. (Fortunately for us, the basic layout of the Arsenale hasn't changed since the 1500s so it is still possible to imagine how the work was done.)

One of the floating assembly halls in the Arsenale, Venice

As I walked through the Arsenale and reflected on what the designers and production workers there contributed to human knowledge and best practice, I found myself wondering what we will add on our watch 500 years later. By my reckoning, our watch (or at least my watch) is the period after 1979 when the practices developed at Toyota, Honda, and their hundreds of suppliers in Japan were first transferred to new countries and new industries.

It seems to me that we have already achieved several things of lasting value:

• We have transferred and adapted lean process tools for production, product development, supplier management, and customer support to a wide range of industries in a wide range of countries.
• We have experimented with all of the management tools - policy deployment, A3 analysis, and standardized management with kaizen - that are needed to introduce and sustain these process tools.

But we haven't combined all of these tools and management methods in more than a few organizations and even these are struggling in the turbulent conditions that the world economy will apparently continue to provide. So what is reasonable to hope for in the remainder of our watch, which in my case is probably the next ten years?

It's not realistic to hope that all organizations will become lean. And this wish makes no sense in any case because lean thinking will continue to evolve with changing circumstances and as the customer problems needing solutions change. The journey never ends as long as there is ever-changing life.

But can we hope to reach a landmark intermediate destination? Specifically, can we create at least one complete lean enterprise in every major industry and activity by 2020?  By a lean enterprise I mean an organization that focuses clearly on the customer problems it seeks to solve and implements lean product development, fulfillment, supplier management, and customer support processes tied together by lean management to cost-effectively solve them.

This surely is a stretch goal and seems all the more daunting given the recent struggles of our lean exemplar companies. But we already have most of the knowledge in hand so no fundamental invention is required. And all of the elements of lean enterprise have been tested individually in real organizations. So the question is whether we think we can. And the only way to find out is to conduct a wide range of experiments within the global Lean Community, freely sharing our findings in the spirit of PDCA. For my part, I'll continue to think that we can until our experiments conclusively prove we can’t.

Best regards,
Jim

James P. Womack
Founder and Chairman
Lean Enterprise Institute, Inc.

Dear Robert,

I've had a big smile on my face for much of the last month because I've had the opportunity to visit progressive organizations on three continents to look at their efforts to create lean value streams. Walking through any process, good or bad, seems to put a smile on my face for one of two reasons. If the process is awful it's easy to see how it could be better. And, if it has already been significantly improved from its original condition, I'm both pleased by the progress and aware that the next layer of waste is now visible and ready for elimination.

However, I also found myself frowning as I walked along some value streams. This happened when I heard improvement teams complaining about the difficulty of gaining and sustaining the engagement and cooperation of everyone and every part of the organization touching the process being improved.

For example, on one walk through an information processing activity in a large service company, the team was complaining about the resistance of the company's information technology department to substantially modifying the company-standard software package in order to support the improved process. In another case, a team was bemoaning the resistance of experienced financial service workers to sharing the details of how they worked their way around the problems in the existing process. In both cases I found the teams defaulting to the most comfortable explanation for the lack of engagement: Bad people.

When this happens I try to take off my technical analysis hat and put on my human empathy hat. I ask, "How do the team's requests feel to the individuals or departments being asked to do something different?" As I do this I remember our old friend Vilfredo Pareto (1848-1923), who gave us the 80/20 rule. (Pareto's first statement of this rule was based on the discovery that throughout history 80% of the wealth in societies was controlled by 20% of the population. Joseph Juran later (1941) extended the 80/20 rule to quality problems where he found that 80% of a problem is typically caused by 20% of the possible causes. And today the 80/20 rules seems to find application in practically every activity.)

Pareto had a second insight of direct relevance to what I saw on my walks. This was his concept economic optimality which states that any event in society should be judged in a positive light when no one is worse off and some individuals and organizations are better off. Public policy analysts (of which I was one early in my career) later realized that this concept applied particularly well in evaluating policy changes by governments. Not only were "Pareto Optimal" outcomes desirable on grounds of equity. Achieving them, by transferring some of the winners' gains to compensate losers (creating Pareto Optimality if it was not otherwise present), also made such policies much more feasible politically because potential losers were much less likely to resist change.

Applying this idea to the value stream improvements I was observing, I asked if the IT department and the experienced employees would be better off with the changed process. And the answer, after a bit of discussion, was clearly "no". The IT department would seriously overrun its annual budget in responding promptly to the team’s request while falling behind on other projects. The experienced employees would very likely be replaced by younger, lower-paid employees able to operate the new process without the need for all the veterans' workarounds.

The root cause of the problem was therefore not bad people. In fact, those affected were reacting quite rationally to protect their interests because they would be hurt by the changes. Instead the problem was a lack of discussion and negotiation between the heads of IT, HR, and the improvement teams about how winners could compensate losers to make everyone whole.

What was particularly striking to me as the outside observer was that Pareto Optimality easily could be achieved in these value streams by reallocating the substantial savings gained from improving both processes. The total saving would be much more than adequate to compensate IT for the additional hours and cost incurred in modifying the software quickly. And the substantial savings from the revised financial process were ample for giving the experienced employees, most of whom were near retirement, a generous severance package or transferring them at similar compensation to other jobs opened up by the organization's high turnover. Yet the implicit, unexamined thinking of the improvement teams was that all of the savings (plus the positive customer response to the improved processes) would be captured by the departments at the end of the processes and that everyone else should just get used to this new reality.

Understanding how change affects every participant in a value stream takes an extra effort and I often find that improvement teams shudder at the prospect of negotiations with leaders of all affected parts of the organization. But my experience over many years is that making visible efforts to make everyone whole -- by striving for Pareto Optimality whenever possible -- is the best way to make and sustain big improvements in core processes. So please give this concept a try in your organization the next time you find "bad people" standing in the way of valuable improvements in your value streams.

Best regards,
Jim

James P. Womack
Founder and Chairman
Lean Enterprise Institute, Inc.

Dear Robert,

Recently I've been spending most of my gemba time walking through value-creating processes in organizations far away from manufacturing. And the further away I get -- for example, all the way to healthcare -- the more I find myself asking, "Who designed this wasteful, incapable, unavailable, inadequate, inflexible, uneven, and disconnected value stream in the first place? And who is responsible for its performance now?" The answer in most cases is that usual suspect, No One.

To be fair to those readers in services industries, I must note that process awareness and process thinking - note that I use the terms "process" and "value stream" interchangeably -- are inherently easier in the factory. The product is there for all to see and the industrial, manufacturing, and production engineering departments and professions are all in place. In addition, almost all manufacturing businesses periodically introduce fundamentally new products, involving at least some new manufacturing methods, which offer the opportunity to rethink the entire value stream. The proof that these value streams are the easiest to transform is that process awareness declines rapidly in manufacturing companies once the factory floor with its physical transformations is lost from view. Product development, sales, purchasing, human resources, and finance are all processes -- series of actions that must be performed correctly at the right time in the proper sequence to create the value needed by their customers. Yet the design of these value streams often receives little or no attention even in firms that have become quite proficient in mapping and then redesigning their factory value streams.

And if your task is managing customer support in a bank, or claims adjusting in an insurance company, or the flow of patients through a hospital, the chances are great that you need to fundamentally rethink a value stream that was never designed in any rigorous way and for which no one has clear responsibility. Many of these processes seemingly just happened and have continued to happen for years or decades so that the deficient current state seems sanctioned in some way by history. "It's always been like this. But we are smart and work around problems every day."

What's missing here is what I call the value-stream architect. This is the person, ideally a line manager given a special assignment, who is asked to take responsibility for fundamentally re-thinking an existing value stream to create a lean value-creating process. (This may be a primary value stream directly touching the customer, such as product development or fulfillment from order to delivery, or a support stream necessary to operate some primary stream.)  The first step is to determine customer need and the type of process required to meet the need. Then the value stream can be mapped by walking along the stream with everyone touching it and the current state can be determined along with the gap in meeting customer need while permitting the organization to prosper. This analysis will suggest potential improvements and the architect then strives for agreement on the best alternative to pursue, followed by experiments using the familiar Plan-Do-Check-Act cycle to determine if the alternative works.

I like the term architect because in its conventional usage this is a person who talks with customers to determine their needs for a building and also talks with all the players involved in design and construction (at a minimum the engineer and the general contractor) to create a process that will deliver the desired value. The difference, of course, is that traditional architects are creating buildings while value-stream architects are creating coherent, lean processes.

This all sounds hard and it is. Most organizations tackle process improvement as a staff (or consultant) activity at isolated points along extended value streams and often without asking what is really important to the customer or the implications for the whole organization. In consequence, the most important processes -- the ones flowing across the organization -- often aren't tackled and, even when they are, the improved process can't be sustained because the new process isn't understood by the line managers who don't see maintaining and improving the entire value stream as their responsibility. (This points to the need to designate a value-stream manager for each process once it is transformed, to continually monitor the performance of the value stream and steadily improve it through kaizen.)

I wish we in the Lean Community could simply draw on Toyota or Honda practice in pursuing lean process architecture. But I fear we can't. I've made many enquiries but I haven't found any clarity in the way those organizations design robust, lean value streams for their value-creating activities outside the factory. And in any case, the vast majority of the world's processes operate in organizations far removed from manufacturing. So I think we are pretty much on our own as lean thinking continues to spread far beyond its original point of origin.

What we need now is experiments in many industries in ways to make value-stream architecture a core skill of line management. And I'm hoping members of the Lean Community will lead the way and share their experimental findings. Indeed, we are hoping to examine these experiments as part of our new LEI initiative on lean management and to share our conclusions about promising approaches. (If you have experiences to share I hope you will send them to Chet Marchwinski who prepares our "success stories" for our website at www.lean.org.)

To cite one example of the type of experimentation needed, my long-time co-author Dan Jones and his collaborators have been working with several healthcare organizations in the UK to create value-stream architects to transform the flow of patients all the way through the hospital. This end-to-end process is particularly challenging because the many steps involved from admission to discharge have never been treated as a connected activity and -- in the few cases where they have -- neither patient nor organizational needs have been clearly specified.

Dan has recently asked his collaborators, Marc Baker and Ian Taylor, to write up their approach in a workbook, Making Hospitals Work, and I hope members of the Lean Community in every industry will find it of interest. In reviewing their work I find it particularly important that they have embedded value-stream redesign in an A3 analysis to focus on the needs of the patient and the organization before tackling patient flow. And their discussion is truly provocative on the role of the value-stream manager (the term they use instead of architect) and his or her key contribution as the cross-departmental, horizontal force for transformation in an otherwise vertical organization.

Every industry and organization is different and we will need a lot of experiments in the years ahead if we are going to make everything work. Indeed, I believe that systematic experimentation to discover the best way for lean managers to create lean value streams for every process in every industry is the most important task before the Lean Community.

Best regards,
Jim

James P. Womack
Founder and Chairman
Lean Enterprise Institute, Inc.